Mo Bulabula, ka Bula Balavu WISHING YOU A HEALTHY LIFE, AND A LONG LIFE Fiji Health Sector Review 2024 Investing in People Fiji Health Sector Review 1 Table of Contents Figures, Tables, and Boxes............................................................................................................................... 4 Acknowledgements........................................................................................................................................... 5 Abbreviations.................................................................................................................................................... 8 Executive Summary........................................................................................................................................... 10 Introduction....................................................................................................................................................... 28 Health Outcomes.............................................................................................................................................. 33 Health Financing............................................................................................................................................... 43 Health Service Organization and Delivery......................................................................................................... 56 Human Resources ............................................................................................................................................ 74 Going Beyond the Health Sector to Improve Population Health....................................................................... 86 Policy Suggestions and Programs.................................................................................................................... 92 About the Report: Scope and Methods............................................................................................................. 108 ANNEXES......................................................................................................................................................... 111 Fiji Health Sector Review 3 Fiji Health Sector Review 3 Figures, Tables, and Boxes Figure 1: Health spending is projected to rise in both real terms and relative to gross domestic product...................................................................................................................... 12 Figure 2: Major gaps in diagnosis, treatment, and control of hypertension and diabetes in Fiji............................................................................................................ 13 Figure 3: There were large variations in bed occupancy rates across Fiji’s hospitals in 2019........................................................................................................ 14 Figure 4: Uneven distribution and utilization of skilled health workers in the public sector......................................................................................................... 15 Figure 5: From To  Flipping Fiji’s service delivery around, to strengthen the foundation for preventative care and chronic disease management............................................... 20 Figure 6: Fiji’s life expectancy is low compared to other countries with similar levels of income........... 34 Figure 7: The gap between healthy and overall life expectancy has increased and is wider for women than for men..................................................................................... 35 Figure 8: Neonatal and infant mortality rates in Fiji have stagnated since 2000.................................... 36 Figure 9: Fiji is facing a rising non-communicable diseases burden...................................................... 37 Figure 10: Fiji has one of the highest shares of premature deaths from non-communicable diseases in the world...................................................................................... 37 Figure 11: Non-communicable diseases impose a greater burden in Fiji than in Malaysia at all ages........................................................................................................... 38 Figure 12 : Work participation among Fijian women has been stagnant despite declining fertility rates (1990-2019)............................................................................. 40 Figure 13: Per capita spending on health has been rising in nominal terms, but only grown in real terms in recent years (2000–21)......................................................................... 44 Figure 14: Fiji’s spending on health is low relative to its aspirational and structural peers (2021)........................................................................................ 45 Figure 15: Private health spending in Fiji is rising.................................................................................. 46 Figure 16: Loss ratios for private medical insurance in Fiji are very high............................................... 47 Figure 17: Out-of-pocket expenditures in Fiji are low compared to aspirational peers but high compared to the region and rising (2021)............................................... 48 Figure 18: Staff costs make up a large and rising share of government health spending in Fiji......................................................................................................... 49 Figure 19: Government spending on drugs increased with the introduction of the Free Medicine Program in 2015......................................................................... 50 Figure 20: Government health expenditure is predominantly spent in hospitals.................................... 50 4 Fiji Health Sector Review 4 Fiji Health Sector Review Figure 21: The proportion of older people in Fiji is projected to rise substantially.................................. 51 Figure 22: Per capita spending by age in Fiji, using age-weights from Malaysia and Australia, suggest that healthcare spending per capita will rise with age.............................................. 52 Figure 23: Health spending is projected to rise in both real terms and relative to GDP......................... 53 Figure 24: MHMS Structure.................................................................................................................... 58 Figure 25: Diagnosis, treatment, and control of hypertension and diabetes in Fiji are limited................................................................................................................ 61 Figure 26: There were large variations in bed occupancy rates across Fiji’s hospitals in 2019........................................................................................................ 65 Figure 27: Hospitals in the Eastern Division had much lower bed occupancy rates than subdivisional hospitals elsewhere in Fiji in 2019.......................................... 66 Figure 28: The use of hospitals for antenatal care visits is inversely related to the availability of health workers in primary health care facilities............................ 67 Figure 29: Primary and secondary healthcare facilities do not offer the range of diagnostic services needed............................................................................. 71 Figure 30: Combining the numbers in the public and private sector brings Fiji above the minimum threshold of 1 doctor per 1,000 population..................... 76 Figure 31: Filling existing positions would meet minimum staffing thresholds for nurses, but additional positions would be needed for doctors............................ 76 Figure 32: The number of allocated positions has risen since 2011....................................................... 77 Figure 33: Large numbers of doctors and nurses have left the public sector in recent years for other jobs....................................................................................... 78 Figure 34: An increasing number of nurses left for New Zealand post Covid, while the numbers for doctors have remained constant......................................................................... 78 Figure 35: Women now make up most of the current and future workforce of doctors.......................... 80 Figure 36: Fiji has seen an increase in public sector doctors, nurses, and midwives per population, but stagnating numbers of dentists and pharmacists............................. 80 Figure 37: The number of students specializing in nutrition and rehabilitation is not keeping pace with growing needs........................................................................... 81 Figure 38: Fiji National University is producing a steady flow of skilled health workers for the Fijian (and the broader Pacific) workforce............................................... 82 Figure 39: Due to population growth, Fiji will need to increase the number of doctors, nurses, and midwives to reach and maintain minimum Sustainable Development Goal thresholds to 2050............ 83 Figure 40: Allocated public health positions are unevenly distributed by subdivision and facility level............................................................................................. 84 Figure 41: Regional policies on eliminating trans-fats............................................................................ 89 Figure 42: From To Flipping service delivery around, to strengthen the foundation for preventative care and chronic disease management................................................................................................. 91 Figure 43: Master plan for CWM Hospital: Identifying three buildings with high patient volumes, the Children’s Ward, the Maternity Ward, and the 1965 building................................................................. 96 Fiji Health Sector Review 5 Fiji Health Sector Review 5 Table 1: Implementation framework....................................................................................................... 18 Table 2: Indicative costing and timeline for reform area 1...................................................................... 21 Table 3: Indicative costing and timeline for reform area 2...................................................................... 22 Table 4: Strategies for the reallocaton of public sector skilled health workers....................................... 23 Table 5: Indicative costing and timeline for reform area 3...................................................................... 24 Table 6: Indicative costing and timeline for reform area 4...................................................................... 25 Table 7: Indicative costing totals over seven years (two phases of implementation)............................. 26 Table 8: Lifestyle factors remain the main drivers of disability adjusted life years in Fiji........................ 39 Table 9: Spending on health in Fiji is largely public and domestically funded (2021)............................. 46 Table 10: The distribution of health facilities is uneven across the country............................................ 67 Table 11: A significant share of medical registrations in 2023 were for private doctors, but not so for nurses and midwives........................................................................................................ 75 Table 12: SSB taxes in Fiji and selected neighbouring countries........................................................... 90 Table 13: Implementation framework..................................................................................................... 92 Table 14: Indicative costing and timeline for reform area 1.................................................................... 97 Table 15: Indicative costing and timeline for reform area 2.................................................................... 98 Table 16: Strategies for the reallocaton of public sector skilled health workers..................................... 99 Table 17: Indicative costing and timeline for reform area 3.................................................................... 100 Table 18: Indicative costing and timeline for reform area 4.................................................................... 103 Table 19: Examples of indicators for improvements in data analysis..................................................... 104 Box 1: Salt and Sugar Intake in Fiji........................................................................................................ 39 Box 2: Public-Private Partnerships (PPPs) in Health............................................................................. 59 Box 3: Diagnosing and Managing Ambulatory Care Sensitive Conditions (ACSCs) at the Primary Healthcare Level............................................................................ 63 Box 4: Examples of Community Health Worker Roles in Service Delivery............................................ 64 Box 5: Lessons from Costa Rica’s Integrated PHC Service Delivery..................................................... 69 Box 6: Towards Climate-Smart Health Systems..................................................................................... 70 Box 7: The Sri Lanka Health Service Pyramid....................................................................................... 94 Box 8: PHCorp: Towards Strategic Purchasing in Malaysia................................................................... 102 6 Fiji Health Sector Review 6 Fiji Health Sector Review Acknowledgments This report presents the key findings, policy suggestions, and programs of the Fiji Health Sector Review, which was completed by a World Bank team led by Margareta Norris Harrit (Senior Health Specialist and Task Team Leader), Mesulame Ratu Namedre (Health Specialist), and Maude Ruest (Health Economist/Consultant), working with Professor Ajay Mahal (Senior Technical Advisor) and his team at the Nossal Institute for Global Health at the University of Melbourne. Carried out at the request of the Ministry of Finance, Strategic Planning, National Development and Statistics (MFSPNDS) and in close consultation with the Ministry of Health and Medical Services (MHMS), this review would not have been possible without the outstanding leadership, support, and guidance from: • MFSPNDS provided by: Honorable Biman C. Prasad, Deputy Prime Minister and Minister for Finance, Strategic Planning, National Development and Statistics and his team, including Shiri K. Goundar (Permanent Secretary), Munesh S. Deo (Head of Fiscal Policy Research & Analysis), Kelera Kolivuso Ravono (Acting Head of Budget), Isoa R. Talemaibua (former Head of Budget), Laurie O’Neal Singh (Principal Policy Planning Officer), Sundhia Ben (Manager Economic & Fiscal Policy), Freeda Fremlin (Principal Budget Analyst), Ema Rokowaqa (Budget Analyst), and Fipe Rabo (System Administrator). • MHMS provided by: Honorable Atonio R. Lalabalavu, Minister for Health and Medical Services and his team, including Dr. James Fong (former Permanent Secretary), Dr. Jemesa Tudravu (Permanent Secretary), Dr. Eric Rafai (former Head of Research and Innovation), Muniamma Gounder (former Director Policy and Planning), Idrish Khan (Head of Administration and Finance), Dr Devina Nand (Head of Wellness), Dr. Rachel Devi (Head of Family Health), Joe Fuata (Director Human Resources), Jiosefa Draunidalo (Principal Administration Officer), Makarita Tikoduadua (Head of Executive Support Unit), Dr. Anaseini Maisema (Focal Point for Primary Health Care), and Rachel Fotofili (Data Analysis and Management Unit), as well as the many medical doctors, nurses, community health workers and staff who have contributed toward this effort during workshops, health system site visits, and interviews. The team also extends its gratitude to colleagues and partners who have generously shared time, data, and information, contributions without which this assessment would be incomplete: Professor Paul A Iji and Dr. Donald Wilson (Fiji National University), Pratyasha Acharya (University of Melbourne), Professor Shaista Shameem, Dr. Abhijit Gogoi, Nathasha Mudaliar, Dr. Neelam Zaidi, and Sharon Biribo (University of Fiji), Fiji Medical and Dental Secretariat, Fiji Nursing Council, Medical and Nursing Council Boards of New Zealand, colleagues from the Pacific Community and from the UN agencies and bilateral offices (Australia, New Zealand, Japan, and Korea) in the Pacific, and colleagues from across the World Bank, Kate Mandeville (Senior Health Specialist), Manuela Villar Uribe (Senior Health Specialist), Christopher Herbst (Senior Health Specialist), Wayne Jeremy Irava (Health Specialist), Lander Bosch (Health Economist), Naoko Ohno (Senior Operations Officer), Fang Yang (Health Economist), Libby Hattersley (Nutrition Consultant), Vika Raica Waradi (External Affairs Analyst), Tuimasi Radravu Ulu (Research Analyst), Eseta Cokanasiga (Team Assistant), and Sakura Iwasaki (Team Assistant). The team would like to thank Juan Pablo Uribe (Global Director for Health, Nutrition and Population), Ronald Upenyu Mutasa (Practice Manager, Health Nutrition and Population, East Asia and the Pacific), Aparnaa Somanathan (Practice Manager for Health Nutrition and Population, South Asia) and Stefano Mocci (Country Manager, South Pacific) for overall guidance, and peer reviewers, Ajay Tandon (Lead Economist), Jumana Qamruddin (Global Program Lead, Health Service Delivery), Mickey Chopra (Lead Health Specialist), Son Nam Nguyen (Lead Health Specialist), Tamer Samah Rabie (Lead Health Specialist), Jeremy Veillard (Lead Health Specialist), Rifat Hasan (Lead Health Specialist), Agnes Couffinhal (Global Program Lead, Health Financing), Mehwish Ashraf (Senior Country Economist), and Thomas Walker (Lead Economist) for their constructive feedback and useful suggestions. The team values the strategic advice, guidance and support by Stephen N. Ndegwa (Country Director for Papua New Guinea and Pacific Islands) and Alberto Rodriguez (Regional Director, Human Development, East Asia and the Pacific). Sally Hinchcliffe provided invaluable editorial services and the Greenhouse Studio in Fiji developed the creative design of the report. The Fiji Health Sector Review has benefited from contributions by the Government of Australia and Government of New Zealand through the World Bank’s Pacific Health Program of Advisory Services and Analytics. Fiji Health Sector Review 7 Fiji Health Sector Review 7 Abbreviations ACSC Ambulatory Care Sensitive Condition Ministry of Health and Medical MHMS Services ALOS Average Length of Stay NCD Non-Communicable Disease ANC Antenatal Care NGO Non-Governmental Organization ASDR Age-Specific Death Rate BOR Bed Occupancy Rate NZ New Zealand BTR Bed Turnover Rate Organisation for Economic Co- OECD CHE Current Health Expenditure operation and Development OOP Out-of-Pocket CHW Community Health Worker Public Expenditure and Financial COVID-19 Coronavirus Disease PEFA Accountability Cardiovascular Risk Assessment and PFM Public Financial Management CRAM Management PHC Primary Healthcare CWM Colonial War Memorial Primary Health Care PHCPI DALY Disability-Adjusted Life Year Performance Initiative Equipo Básico de Atención Integral de PIC Pacific Island Country EBAIS PPP Public-Private Partnership Salud Financial Management Information Reproductive, Maternal, FMIS RMNCAH Neonatal, Child, and Adolescent System Health FNU Fiji National University SCM Supply Chain Management GDP Gross Domestic Product SDG Sustainable Development Goals GoF Government of Fiji SEG Standard Expenditure Group GP General Practitioner SHW Skilled Health Worker HALE Healthy Adjusted Life Expectancy SSB Sugar-Sweetened Beverage HRH Human Resources for Health TB Tuberculosis HSP Healthy School Policy TC Tropical Cyclone HSR Health Sector Review UMIC Upper Middle-Income Country Information and Communications VAT Value-Added Tax ICT Technology WB World Bank Institute for Health Metrics and WDI World Development Indicator IHME Evaluation WHO World Health Organization Ministry of Finance, Strategic Planning, MFSPNDS National Development and Statistics 8 Fiji Health Sector Review 8 Fiji Health Sector Review Photo Credit: Fiji Ministry of Health and Medical Services Fiji Health Sector Review 9 Executive Summary 1. This work was conducted at the request of the Ministry of Finance, Strategic Planning, National Development and Statistics (MFSPNDS). MFSPNDS requested the World Bank’s support to conduct a systematic review of the health sector, to take stock of its current performance in tackling the country’s major health challenges and propose policy suggestions and programs for improving its performance. A working group with members from MFSPNDS and the Ministry of Health and Medical Services (MHMS) was established to facilitate data collection and oversee progress of the review. It was agreed that the review would cover: (i) health financing, including health expenditure projections to 2030; (ii) service delivery, including a supply-side readiness assessment of services at all levels of the health system, and redevelopment options for Colonial War Memorial Hospital; (iii) human resources for health, including assessment of stock and flow as well as projections of needs and supply to 2030; and (iv) options and models for the Government of Fiji to engage private healthcare providers to improve population health outcomes. Fijians are not reaching their full productive potential because of poor health outcomes 2. Fiji is an upper middle-income country (UMIC), with the health outcomes of a lower middle-income country, and a catastrophic disease burden from non-communicable diseases (NCDs). Relative to its national income, Fiji underperforms in several key population health indicators. Life expectancy at birth is just 68 years, well below the UMIC average of 77 years and below the lower middle-income country average of 69 years. Gains in healthy life expectancy, a measure of longevity that discounts for time in illness or injury, have been small relative to peers, while infant and child mortality rates have stagnated since 1995, following large improvements during the preceding decades. Fiji has an infant mortality rate of 23 per 1,000 live births and an under-five mortality rate of 28 per 1,000 live births, closer to the average for lower middle-income countries than UMICs. There are recurrent infectious disease outbreaks, driven by environmental factors including climate change and urbanization, and infectious diseases such as tuberculosis (TB) and HIV present risks to population health. Its major health challenge, however, is from NCDs. Fiji has one of the highest rates of adult mortality due to NCDs; about 85 percent of deaths in 2019 were linked to NCDs, with 64 percent of these deaths occurring in the working-age population. 3. Fijians enjoy relatively low out-of-pocket (OOP) expenditure on health, a feature that Fiji should seek to maintain. Currently, around 70 percent of health spending in Fiji is publicly funded, with the balance equally split between private health insurance and OOP payments, meaning health expenditure in Fiji is largely financed by the public sector through general taxation, which is a suitable approach to ensuring equitable access and financial protection. This is one of the major factors that explains Fiji’s low levels of OOP spending, although it is possible that Fijians may also be forgoing care with adverse implications for efficiency and equity. Financial protection is at the core of universal health coverage and is one of the objectives of effective health financing policies: the burden of health expenses should not create financial hardship or threaten people’s living standards and is an achievement that Fiji should seek to protect. The key to protecting people is to ensure prepayment and pooling of resources for health (e.g., via taxation and/or social health insurance), rather than relying on people paying for health services OOP at the time of use. 10 Fiji Health Sector Review 10 Fiji Health Sector Review 4. Fiji’s ability to reap its full economic potential is dependent on ensuring that its working-age population remains healthy and in productive employment. A combination of declining infant mortality and birth rate has reduced the share of young dependents, resulting in an increase in the share of Fiji’s working-age population. Relative changes in the various segments of the population have increased the ratio of working-age people to dependents (children 0–14 years and the elderly, 65 years and over), rising from 1.2 in 1973 to 1.9 in 2021. The growing share of the working-age population translates to a larger labor force pool and opens a window for Fiji to benefit from accelerated economic growth (sometimes referred to as the demographic dividend), provided it can ensure that its population remains healthy and educated, and is supported by conducive economic policy, labor policy, and financial systems that generate productive employment opportunities. This favorable population composition is expected to remain until about 2050. 5. Managing Fiji’s burden of NCDs is essential to harness its demographic dividend and strengthen its fiscal position. The Ministry of Health and Medical Services (MHMS) estimates that NCDs cost Fiji approximately FJ$591 million (US$263 million) per year. This estimate includes direct costs such as treatment and prevention, as well as indirect costs incurred by employees who leave the workforce to care for an affected relative, or who miss or underperform at work because of illness.1 Women’s ability to participate in paid work is adversely affected given their caregiving responsibilities. Left at its current trajectory, we assess that a high and growing NCD burden and an ageing population will gradually increase pressure on public health spending in Fiji. In contrast, ensuring that Fiji’s population lives longer and ages healthily, will boost female labor force participation and further reduces dependency on the productive segment of the population. 6. Should Fiji continue its current trajectory of epidemic NCD levels, health spending needs are projected to reach about 7 percent of GDP by 2050 (Figure 1). This represents an increase by between 200 percent and 250 percent between 2019 and 2050 in real per capita terms. This trend poses a threat to Fiji’s achievements in protecting its population from financial risks, and means, in the absence of increased public spending, households will be forced to spend more from their own resources or forgo care. Fiji can limit the adverse impacts of health spending, while improving population health and economic productivity, by stemming the rise in NCDs. As an illustration, if Fiji could match the health outcomes of UMIC peers by 2050, this would result in a 10-year increase in life expectancy at birth. Additionally, the relatively larger labor force and longer life expectancy translates to an income boost which will keep healthcare needs affordable. 1. MHMS and UNDP Assessment, National Wellness Strategic Plan 2024-2030 Costing Exercise, July 2024. Fiji Health Sector Review 11 Figure 1: Health spending is projected to rise in both real terms and relative to gross domestic product 1800 8 Per Capita Health Expenditure (in FJ$) 1600 7 Share of Health Expenditure Needs 1400 6 1200 5 1000 in GDP (%) 4 800 600 3 400 2 200 1 0 0 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 Year HE/GDP Ratio Per Capita Health Expenditure Source: Authors’ estimates. Fiji’s health system is neither appropriately structured nor equipped to prevent and manage its alarming disease burden 7. Access to good quality health services remains an ongoing concern and Fijians with chronic conditions are at risk of hospitalization for conditions that could be effectively managed by ambulatory care services. Gaps are evident in reproductive, maternal, neonatal, child, and adolescent health (RMNCAH) services, but crucially also in the prevention and effective management of NCDs. Global evidence highlights that a larger share of preventable deaths occurs due to poor quality of care, in comparison to access to care.2 There are major gaps in the quality of NCD services available for Fijians at primary health care facilities. Data from the most recent STEPS survey in Fiji suggests that fewer than one- third of individuals with hypertension and diabetes have been diagnosed or are aware of their conditions (Figure 2). Among those diagnosed, fewer than one-third of people with diabetes and less than one in seven of those with hypertension are on a treatment regimen. Of those enrolled for treatment, less than 10 percent had their blood sugar or blood pressure under control. The current state of this care cascade indicates that only a few Fijians with hypertension or diabetes are being effectively managed. Thus, large numbers of Fijians with chronic conditions are at risk of hospitalization for conditions that can be effectively managed through ambulatory care services (medical services performed on an outpatient basis, without admission to a hospital). 8. Recent studies in Fiji allude to knowledge gaps for staff (including adherence to clinical guidelines) for managing patients with chronic conditions. This is further compounded by limited availability of diagnostic tests and drugs, as well as low levels of patient adherence to treatment.3 2. Kruk et al. 2018. “Mortality Due to Low Quality Health Systems in the Universal Health Coverage Era: A Systematic Analysis of Amenable Deaths in 137 countries“ Lancet 392:2203-12. 3. Kumar and Mohammadnezhad. 2022. “ Separately, the Primary Health Care Performance Initiative (PHCPI) assessment undertaken by the World Bank found that indicators of RMNCAH service coverage, such as antenatal care services have declined in recent years, consistent with the findings of stagnant child health outcomes” (World Bank. 2023. Primary Health Care Performance Initiative). 12 Fiji Health Sector Review Figure 2: Major gaps in diagnosis, treatment, and control of hypertension and diabetes in Fiji 40 35 30 25 Percent 20 15 10 5 0 % Diagnosed % Treated % Controlled Diabetes Hypertension (25-64) Source: Fiji STEPwise approach to NCD risk factor surveillance (STEPS), 2011, also quoted in journals in 2020 and later. 9. There is evidence of inefficiency in resource utilization in the Fijian health sector. A resourcing imbalance between primary care and hospital care can have implications for both allocative efficiency (where resources are being directed) and technical efficiency (how resources are being used). Spending on primary health care (PHC) has increased in absolute terms but fallen as a share of current health expenditure. While more than half of public health expenditure is directed to PHC,4 79 percent of government health expenditure was on hospitals in 2019. This suggests that PHC spending is predominantly allocated to outpatient services in hospital settings, rather than to ambulatory or preventative care providers in primary health care facilities (e.g., health centers, nursing stations) which are a less costly alternative and are more physically accessible to patients. MHMS data on antenatal care (ANC) visits suggest that more than half of all ANC visits to hospitals are by pregnant women who are not deemed to be at risk, and thus could be managed by personnel at lower-level healthcare facilities. 10. Fiji’s national referral hospital, the Colonial War Memorial (CWM), and major urban health facilities are running at close to maximum capacity. CWM consistently operates at full capacity and is often required to offload patients to other care facilities. The Valelevu health center, with a catchment population of 60,000 people, serves as one of CWM’s offload facilities for both outpatient clinics and inpatient care, even though it is not classified or equipped to serve as a hospital. This center recently had to close its diabetic foot care clinic and reduce outreach activities to a minimum due to its large patient load. Similar trends are observed at other urban health centers with large catchment areas, limiting their ability to provide preventative care, which in turn leads to delayed diagnosis of disease and avoidable complications that ultimately require costly treatment at higher levels of care. It also causes stress and burnout among healthcare staff. 11. In contrast, many of Fiji’s rural hospitals have very low utilization rates (Figure 3). Hospitals in Fiji’s Northern Division average a mere 30 percent bed occupancy, and hospitals in the Eastern Division reach an average bed occupancy of 10 percent, indicating limited provision of inpatient services relative to investments in the hospital sector in these divisions. Human resource shortages and gaps in complementary 4. PHC expenditure includes general, dental, and other outpatient curative care; home based curative care; outpatient and home-based long-term health- care; part of medical goods and health system administration/ governance costs; and preventive care e.g., immunization and health education (Global Health Expenditure Database, World Health Organization). Fiji Health Sector Review 13 inputs (e.g., equipment and diagnostics) appear to be part of the explanation for excess capacity alongside capital investments that may not be well matched to population needs in remote island locations. Figure 3: There were large variations in bed occupancy rates across Fiji’s hospitals in 2019 90.0 80.0 70.0 Bed Occupancy Rate (%) 60.0 50.0 40.0 30.0 20.0 10.0 0.0 0 100 200 300 400 500 600 Number of Hospital Beds Source: MHMS data. Underperformance of Fiji’s health system is underpinned by a set of key drivers 12. In addition to inefficient use of sectoral resources, Fiji is also underinvesting in health relative to its UMIC peers, limiting its ability to strengthen cost-effective primary care services. With total health expenditure of 4.25 percent of gross domestic product (GDP),5 Fiji spends less on health than peers with similar levels of income such as Jamaica (6.1 percent) and Belize (4.8 percent). Fiji’s public spending on health was 2.6 percent of GDP in 20196 which is less than most of its UMIC and regional peers, such as Jamaica (4.1), Samoa (4.1), and Belize (3.3).7 This limits Fiji’s ability to strengthen PHC. For instance, good quality and resilient physical infrastructure is crucial for patient and health providers’ safety and comfort and is a key input to the provision of health services. More than half of Fiji’s PHC facilities need upgraded electricity and plumbing, and maintenance of physical infrastructure and equipment. Resourcing the strategic investments highlighted in this report is vital to bring Fiji’s health system up to par with its economic development and to meet the needs of the population. 5. Using 2019 Global Health Expenditure Database to exclude the impact of COVID-19 on health spending and GDP in 2020 and 2021. 6. We use 2019 data to exclude the effect of COVID-19 which distorted spending on health (2021 is the latest year available in the global database). 7. Global Health Expenditure Database. 14 Fiji Health Sector Review 13. Fiji is not providing adequate access to diagnostic services and medicines in PHC facilities (health centers and nursing stations). PHC facilities are poorly equipped with diagnostic resources and other consumables that allow healthcare professionals to initiate timely interventions, saving lives and money. The 2021 Service Availability and Readiness Assessment survey indicated that rapid diagnostic tests were patchily available in PHC facilities, requiring patients to visit secondary and tertiary healthcare facilities for some tests. Only 20 percent of health facilities had all the standard safety precautions or equipment. Last-mile delivery of essential medicines remains a challenge at PHC levels. Supply chain management (SCM) challenges related to regulations, forecasting, procurement, and distribution have led to exposure to counterfeit medicines, inadequate access to drugs in government pharmacies, and the common practice of supplementary orders, which result in inefficient public sector spending and increased OOP expenditure on medicines. 14. Fiji reaches the indicative minimum threshold of skilled health workers (SHWs) when both the public and private sectors are considered. In 2016, the World Health Organization (WHO) Global Strategy on Human Resources for Health identified the availability of 4.45 doctors, nurses, and midwives per 1,000 population as an indicative minimum density of skilled healthcare workers (SHWs).8 More recent estimates suggest that the 4.45 minimum threshold may underestimate the workforce needed to achieve universal health coverage.9 Healthcare workforce densities and disparities are strongly related to sociodemographic development. Low-income countries generally have low densities of SHWs, and high-income countries generally have significantly higher densities of SHWs than the quoted minimum thresholds. Across the public and private sectors, Fiji has approximately 1.2 doctors per 1,000 people and about 3.2 nurses and midwives per 1,000 population, totaling approximately 4.4 SHWs per 1,000 population. Fiji‘s SHW numbers lie in the middle of the range for structural and aspirational peer UMICs: Jamaica and Tonga have fewer doctors (0.8 and 0.4 per 1,000 population, respectively) and nurses and midwives (2.4 and 2.8 per 1,000 population, respectively), whereas Maldives has almost 3.4 doctors and 4.7 nurses and midwives per 1,000 population. Figure 4: Uneven distribution and utilization of skilled health workers in the public sector Allocated MHMS Staff per level of facility, by subdivision, per 1000 people 9.0 8.1 8.0 7.1 7.1 7.0 6.0 5.5 5.2 5.3 5.2 5.0 4.0 2.7 2.8 2.9 3.0 2.4 1.7 1.6 1.8 1.6 2.0 1.3 1.7 1.6 1.5 1.6 1.2 0.9 0.9 1.3 1.2 1.3 1.31.0 1.1 1.2 1.4 1.3 1.3 1.6 1.7 1.0 0.6 0.6 0.8 0.9 0.5 0.0 Tavua Suva Nadroga/Navosa Bua Cakaudrove Kadavu Lakeba Lautoka/Yasawa Lomaloma Naitasiri Tailevu Rakiraki Rotuma Serua/Namosi Ba Macuata Taveuni Rewa Nadi Lomaiviti Central Western Northern Eastern Health centers and nursing stations staff per 1,000 Hospital staff per 1,000 Source: Ministry of Health and Medical Services Administrative Data and authors’ calculations. Note: CWM hospital staff are included in the Suva subdivision for this analysis (refer to Annex 10 for more detailed data). 8. In 2016, WHO adopted a method that quantified how many health workers are needed to achieve a median performance on an SDG index composed of 12 tracer indicators: 5 related to infectious diseases; 3 with respect to maternal, newborn and child health; and 4 for non-communicable diseases. 9. Global Burden of Disease 2019 Human Resources for Health Collaborators. 2022. “Measuring the Availability of Human Resources for Health and its Relationship to Universal Health Coverage for 204 Countries and Territories from 1990 to 2019: A systematic analysis for global burden of disease study 2019” The Lancet 399:2129-54. Fiji Health Sector Review 15 15. Fiji’s deployment of its healthcare workforce is uneven and SHWs are stretched thin in urban PHC facilities. There are wide geographical variations in the allocation of SHWs across regions and facilities (Figure 4). As an example of an urban PHC facility with a large catchment area, the Nuffield Health Center, in Fiji’s Central Division, serves a catchment population of over 56,000 people, with an actual staffing ratio of just 0.67 SHWs per 1,000 people. Similarly low levels of SHWs can be observed at PHC facilities across Fiji, where staff are unable to provide essential preventative services to patients, while hospitals in areas such as Rotuma, Lomaloma, Lakeba, Vunisea, Vunidawa, Tavua, and Waiyevo have lower utilization rates and higher ratios of SHWs to the population served. 16. Emigration increased in the years after COVID-19 but the greatest movement of SHWs has been from the public to the private sector. According to exit data collected by the MHMS, most doctors and nurses who leave Fiji go to New Zealand, followed by Australia. In the period immediately following the COVID-19 lockdown, Fiji experienced public sector SHWs leaving their jobs in a post-pandemic exodus. There was an increase in number of SHWs registering with New Zealand’s nursing and medical councils in 2023, when a total of 79 Fijian nurses and 10 Fijian doctors registered in New Zealand. This was up from 32 nurses and seven doctors in 2022 and the average of nine nurses and seven doctors a year in the past decade.10 However, most SHWs leaving the public health sector in the post-COVID exodus appear to have stayed in Fiji and joined the private health sector. 17. Fiji has endorsed a digital health strategy for 2023–30 but a lack of basic infrastructure, equipment, and digital literacy is delaying its implementation. The evidence available on telehealth interventions suggests that health systems that effectively adopt digital solutions can boost access to health services; reduce need for, time, and costs of travel for patients seeking care; and reduce the time interval between first symptoms and management for patients who need coordinated care, for example involving specialists and primary care providers.11 Gains can also arise from easier access to continuing professional education. At the time of drafting this report, 133 health facilities in Fiji are yet to be connected to the internet. Many of the healthcare facilities that have been upgraded to high-speed internet lack computers to make use of the digital enhancements. Fragmented implementation of digital interventions and a lack of skills among the healthcare workforce on the use of digital platforms and tools mean the intended gains from the digital transformation have only been partly realized. 18. Private healthcare services are growing and could support healthcare delivery, including for prevention and primary healthcare services, but are currently not effectively used for this purpose. Currently, approximately 30 percent of SHWs work in the private sector. Private sector providers are predominantly located in or near urban areas and primarily in the Central and Western Divisions, offering a potential source of services for the urban poor. A growing trend in UMICs globally is for governments to use purchasing entities to engage private sector providers for healthcare and other services. Public-private partnership arrangements exist in various forms in Fiji, but there is little evidence about their effectiveness in improving priority outcomes and overall monitoring capacity is weak. The Government of Fiji (GoF) appears to be limited to being a passive purchaser of services and should consider developing a more strategic purchasing role. This will require not just greater investments in human resources engaged in purchasing private care, but also clearer policies on reporting, data sharing, and data governance to help assess the capacity, quality, and cost of public and private service delivery. 10. Nursing and Medical Councils of New Zealand data, 2023. 11. Rajit et al. 2021. Review of Policy and Use of Tele-health to Support Health System Strengthening in the New COVID Normal. Report to the Western Pacific Regional Office, WHO, Manila. Melbourne: Nossal Institute for Global Health; WHO. 2019. Recommendations on Digital Interventions for Health Systems Strengthening. Geneva: World Health Organization. 16 Fiji Health Sector Review 19. Strengthening Fiji’s health system requires effective solutions to public financial management (PFM) challenges. Improvements in public expenditure and financial accountability (PEFA) scores are associated with improvements in indicators of child mortality.12 The budget execution rate of MHMS has remained stable over the past five years, averaging 80 percent across all standard expenditure groups. This is well below the UMIC average of 98 percent and equivalent to a public expenditure and financial accountability (PEFA) score of D,13 which is a sign of major deficiencies. There was some improvement during the 2023 fiscal year (FY23), with MHMS utilizing 88 percent of its total budget. In FY23, a key bottleneck for MHMS execution was the post-pandemic departures of SHWs from MHMS to the private sector and opportunities abroad. The Ministry of Finance, Strategic Planning, National Development and Statistics (MFSPNDS) allocated dedicated budget for filling the vacant positions, but the recruitment of such a large cohort of staff proved challenging to complete in just one year, creating a surplus in the balance allocated for MHMS salaries. Capital construction is a more consistent bottleneck for social sector ministries which generally are less experienced and resourced compared to infrastructure-heavy ministries such as the Ministry of Infrastructure and Meteorological Services.14 Strengthening Health Sector Engagement with Other (Non-Health) Sectors 20. A whole-of-government approach to effectively address lifestyle risk factors, such as obesity, physical inactivity, and unhealthy diets is critical to achieve population-level change. The high prevalence of lifestyle-related risk factors in the Fijian population and the significant NCD disease burden suggests that policy makers need to step up measures to reduce lifestyle-related risks to help delay the age at first onset of disease and slow the transition to multi-morbidity (living with two or more chronic illnesses) among people with NCDs. Although risk profiling, screening, treatment, and effective management of NCDs are actions that are in the direct domain of the health sector, concerted efforts to promote healthy lifestyles must extend beyond the health sector. These will require building on ongoing engagements with the finance, education, labor, urban and rural development, and agriculture and waterways ministries, and other agencies. 21. Fiji has introduced many of the “best buy” population health interventions, but implementation can be strengthened, especially regarding health taxes. Although Fiji has introduced policies and interventions that have provided value for money in UMIC settings, they are not ambitious enough to tackle Fiji’s NCD emergency. Implementation, enforcement, and monitoring are generally weak. A recent WHO report on tobacco confirms that cigarettes were less affordable in Fiji in 2022 than they were a decade earlier but that tobacco taxes only accounted for 38 percent of the retail price of cigarettes—about the same as in Tuvalu and Vietnam, but considerably lower than Australia (77 percent) and Cook Islands (73 percent). Measures to promote healthier diets are also lagging despite these now being major risk factors for NCDs in Fiji. Fiji taxes foods that are rich in sugar, sugar-sweetened beverages and unhealthy snacks but salt and sugar are exempted from the standard value-added tax (VAT). The tax system could also be used more effectively to promote consumption of healthier foods including vegetables, fruits, nuts, and seeds. 22. A promising initiative is the Healthy School Policy (HSP) by MHMS and the Ministry of Education, which targets NCD risk factors among younger Fijians at a time when their health habits and lifestyles are being formed. This GoF initiative is expected to introduce health-promoting content into the school curriculum related to physical activity and healthy eating, and age-appropriate information on healthy choices. Through a gradual roll-out, it will provide children aged from 3 years to 19 years with routine, systematic screening in early childhood education centers, and primary and secondary schools nationwide. The HSP also promotes school gardens that make nutrient-rich vegetables available in the school setting, which can be a tool to tackle Fiji’s triple burden of malnutrition (undernutrition, overweight, and micronutrient deficiencies) among Fiji’s children and youth. 12. Moritz and Smets. 2019. Public Financial Management, Health Financing and Under-five Mortality: A Comparative Empirical Analysis. World Bank and Inter-American Development Bank. 13. PEFA assessment scores countries on a scale of A-D with A ranking the highest and D the poorest. In budget credibility, D reflects a budget execution deviation of 15 percent or worse from the original budget. 14. The average utilization of capital construction allocation for Ministry of Education 2017 - 2022 is 44 percent, while the average utilization for Ministry of Infrastructure and Meteorological Services over the same time period is 88 percent. Fiji Health Sector Review 17 Policy Suggestions and Programs 23. In this report, we propose a new vision of healthcare, centered around the expression Mo Bulabula ka Bula Balavu: a Fijian greeting meaning “Wishing you a healthy life, and a long life”. The report estimates the financing required for the country to develop a health system that will be fit for the 21st century and meet its population’s expectations. The policy suggestions and programs aim to preserve Fiji’s achievements in low OOP spending, while enhancing the economic gains from a healthier working-age population. Leading Fiji toward healthier lifestyles and greater well-being depends not only on expanding access to quality care, but also creating pathways for Fijians to make healthier choices that improve longevity and reduce the risks and impacts of chronic disease. 24. Fiji will require a combination of policy reforms, actions, programmatic investments, and service delivery shifts to achieve this vision. This section presents a concise set of costed policy suggestions and programs the GoF and its partners can take now, to improve health outcomes and reduce the long- term fiscal burden of the health sector. Table 1 summarizes the proposed priority policy reforms, institutional changes, and programmatic investments required to transform the health sector. Table 1: Implementation Framework Vision: Mo Bulabula ka Bula Balavu Wishing you a healthy life and a long life Reform area 1: Reform area 2: Reform area 3: Reform area 4: Redesign health service Spur the adoption of Build a modern health Strengthen stewardship delivery to meet the health healthy behaviors in the workforce for the future. and data for a modern challenges of Fiji and the population. health system. Pacific. Cross-cutting: Enhance health resilience through health emergency prevention, preparedness, and response, using climate-smart solutions. Improve health equities though pro-poor and gender-sensitive interventions. 25. Implementation will be phased over time. The proposed sequencing will enable Fiji to take a stepwise and phased approach to its reform journey given the fiscal cost and the technical and administrative demands of such reforms. Phase I interventions address challenges that affect people’s lives daily. Their implementation, alongside process and impact evaluations will allow MHMS to “get the basics right” of its health sector transformation, which aligns with the strategic direction of Fiji’s national development plan, and its vision to enhance the provision of foundational policies and services that can tangibly improve the quality of life of ordinary Fijians. 18 Fiji Health Sector Review Reform area 1: Redesign health service delivery to meet the health challenges of Fiji and the Pacific 26. Fiji’s health service structure needs to be ‘flipped’, from a system focused on hospital-based, curative care to one that prioritizes preventive care and disease management at the primary level (Figure 5). Primary healthcare-oriented systems offer the most cost-effective, equitable, and accessible route to improving population health. An effective PHC system promotes healthy lifestyles, diagnoses diseases early, and provides effective disease management. Hospitalizations for complications related to diabetes and hypertension can often be avoided if these conditions are managed early and adequately at PHC levels. The goal is to reduce the number of patients with chronic conditions and associated complications who end up seeking care in secondary and tertiary facilities. Despite this emphasis on strengthening PHC, quality tertiary care capacity remains essential, considering Fiji’s hefty disease burden. Although some of Fiji’s secondary and tertiary facilities have benefitted from recent upgrades, investments are still needed to improve healthcare services and safety at Fiji’s main referral hospital, the Colonial War Memorial (CWM). 27. Proposed priority 1: Transform PHC services with the introduction of a credentialed, digitally-enabled, climate-resilient, team-based PHC model. This includes the redesign and roll-out of a new PHC model with the following characteristics: (i) a comprehensive PHC services package that reflects Fiji’s evolving disease burden, demographic shifts, and climate change impacts; (ii) investments in PHC infrastructure nationwide, inclusive of equipment upgrades in Fiji’s health centers and nursing stations, informed by global best practices to facilitate early diagnosis and management of NCDs; and (iii) improved access to inputs such as pharmaceuticals and medical commodities, including during pandemics and weather emergencies. The government can implement this PHC reform through a phased approach (Table 2). Given the urgent need to improve population health at scale, some of Fiji’s largest catchment areas—for example, the medical areas of Valelevu, Nuffield, and Raiwaqa, collectively reaching over 150,000 people—could serve as implementation hubs for urban areas, accompanied by evaluations to inform a national roll-out. Similarly, evaluations of phase I implementation interventions in selected rural areas can be used to inform the roll-out and necessary adaptation of health sector policies and programs to rural settings. 28. Proposed priority 2: Unleash the power of digital technology and data for health. Enhanced connectivity and the implementation of Fiji’s 2023–27 digital health strategy will successfully move Fiji toward digital patient records, NCD registries, telehealth services, access to just-in-time consultations with clinicians at higher-level facilities, and online learning, including in the Outer Islands. Fiji’s digital strategy, which was launched by MHMS in 2023, advocates for an enterprise architecture-based approach, ensuring that all digital solutions are patient-centered, accessible, compatible, interoperable, and sustainable in the long term. By following this structured framework, the policy will facilitate the integration of new technologies, promote human and financial resource optimization, and enhance the overall efficiency and resilience of the health system. Its successful implementation is contingent on (i) ensuring the basic infrastructure is in place at all levels of care to pave the path for a digital transformation; (ii) building digital literacy among healthcare workers and the general population to use and trust technology and also be aware of its limitations; and (iii) dedicated human resources with the technical skills to design, maintain, and promote effective use of digital technologies and systems. 29. Proposed priority 3: Transform hospital services by creating a modern, integrated, digitally enabled diagnostic and treatment network for Fijian and Pacific patients. Strengthen the functioning and appropriate utilization of CWM as a tertiary care hospital and training facility, enhance the availability of diagnostics, and management of complex conditions in subdivisional facilities (secondary care), which are critical to improve patient pathways and outcomes. Investments in health facilities across the care spectrum (PHC, secondary, and tertiary) are highly symbiotic. Effective health promotion and a strengthened PHC system are crucial to preventing premature disability and deaths among the working-age population. It reduces the burden on Fiji’s secondary, tertiary, and specialist hospitals. At the other end of the spectrum, increased access to high quality tertiary care and clinical training is necessary to provide advanced health care services Fiji Health Sector Review 19 that cater to the Fijian demographic profile and disease burden, and by extension, the neighboring countries that rely on Fiji for specialized care. CWM, the Pacific’s largest tertiary and training hospital, serves as Fiji’s national referral hospital, and also cater to patients from other Pacific Island Countries, such as Tuvalu and Kiribati, where access to comprehensive tertiary care is limited. CWM also serves as a training hospital for SHWs from across the Pacific, majority of whom are trained in Fiji. The proposed Priority 3 does not include the recommendations of the CWM infrastructure master plan, which is expected before the end of-2025, generously supported by the Government of Australia. In the meantime, based on a recently concluded architectural and structural assessment conducted as part of the World Bank financed Fiji-COVID Emergency Response Project, engineering recommendations propose enhancing services and safety in three buildings that provide critical services to high patient volumes and provide critical services, the Children’s Ward, the Maternity Ward, and the 1965 building. Therefore, the proposed investments covered in this report is limited to activities required to bring these buildings up to Code in critical areas of seismic and structural stability,15 asbestos,16 and fire safety17 notwithstanding future resourcing needs for CWM per the Master Plan. Figure 5: From To Flipping Fiji’s service delivery around, to strengthen the foundation for preventative care and chronic disease management From: a curative health system with To: A healthcare system that prevents and effectively large burden on hospitals manages disease through effective PHC • Tertiary prevention: Reduce the Tertiary Care effects of the disease and need for Tertiary Care hospitalization through effective disease management • Secondary prevention: Early Secondary screening programs Secondary Care Care • Primary prevention: E.g. Immunization • Primordial prevention: Risk factor Primary Care reduction (e.g. enabling physical Primary Prevention and Health activity and curbing malnutrition. Care Promotion Example: Healthy School Policy) Fiji’s Health System Today Ideal scenario, post reform 15. Review against New Zealand Society for Earthquake Engineering guidelines. 16. Various work, health and safety standards, the investigation was only for asbestos-containing materials and excluded other potentially hazardous materials, such as lead paint. 17. In line with local and international building codes including Australia and New Zealand. 20 Fiji Health Sector Review Table 2: Indicative costing and timeline for reform area 1 Implementation of reform area 1: proposed phasing and indicative costing of interventions Phase I over the first 1-3 years Followed by gradual roll-out (phase II) over 4-7 years Define Fiji’s PHC service package and test if it is Roll-out PHC upgrades in strategically FJ$ fit-for-purpose in urban (e.g., Valelelvu, Nuffield located health centers nation-wide. FJ$61–81m 20m and Raiwaqa) and rural implementation hubs. Implement a digital transformation nation-wide FJ$ Roll-out PHC upgrades in select nursing and ensure digital foundations are in place at FJ$19m 24m stations nation-wide. all levels of care, including the Outer Islands. Implement highest priority upgrades at CWM’s FJ$ Additional operational costs of Fiji’s digital Children’s Ward, Maternity Ward, and 1965 FJ$8m 29m transformation. building. TBD by forthcoming CWM renovation or rebuild. Master Plan Phase 1 Phase 2 Total Indicative Cost: FJ$73m Total Indicative Cost: FJ$88m–108m (US$33m) (US$40–49m) Reform area 2: Spur the adoption of healthy behaviors in the population 30. Fiji needs bold action to tackle its NCD crisis, beyond what the health sector can do by itself. It needs to implement effective strategies that focus on lifestyle risk factors to delay the transition to NCDs and the development of multi-morbidity among people with NCDs: this includes partnering with finance, education, agriculture, and other ministries and bodies. Although Fiji has introduced policies and interventions that have provided value for money in other UMIC settings, they are not ambitious enough to tackle its current emergency and their implementation is weak. Fijian leaders can take measures that are commensurate with the scale of its NCD crisis. 31. Proposed priority 4: Improve risk profiling, diagnosis, and management of NCDs through screening programs and use of digital NCD registries. Fiji has already adapted the Package of Essential Non- communicable Diseases services, as recommended by the WHO, to the Fijian context, in the form of its Cardiovascular Risk Assessment and Management (CRAM). The CRAM system uses a color-coding system to show the level of advancement of the disease. Currently, the folders of many of Fiji’s patients enrolled in CRAM are red or deep red, which means they are at an irreversible stage of the disease. Implementing large- scale screening programs for hypertension and diabetes in workplace settings, schools, and PHC facilities will allow for earlier diagnosis and improved disease management and a more robust referral system to secondary and tertiary care for complex cases. Integrated digital patient records and disease registries that can be disaggregated by sex, age, and other sociodemographic determinants are essential to effective risk profiling, screening, diagnosis, and the treatment process. This will be enabled through the implementation of Fiji’s 2023–27 digital health strategy, digital infrastructure (proposed priority 2), and increased capacity for data entry, disaggregation, management, and quality assurance. Fiji Health Sector Review 21 32. Proposed priority 5: Implement and evaluate population-level social and behavior change programs. The government has recently adopted promising policies centered around prevention, including the National Policy on Healthy Catering and Sale of Food and Beverages for Government Ministries and Institutions, through which it intends to lead by example. The HSP, approved by Fiji’s Cabinet in April 2024 has significant potential to promote healthy behaviors among children and youth if well implemented and adopted by most Fijians. Progress will be measured by (i) the share of students from early childhood to secondary school who are malnourished18 who are identified and referred for treatment; and (ii) the share of students meeting recommended guidelines for physical activity (disaggregated by boys and girls). Similar indicators could be promoted widely, including in workplace and community settings. The HSP also promotes school gardens that make available nutrient-rich vegetables in the school setting, which can serve as a tool to tackle Fiji’s triple burden of malnutrition among its children and youth. Behavior change interventions that assist health professionals, community leaders, and citizens to engage in health promotion are likely necessary considering high levels of acceptance of NCD risk factors. Evaluations should monitor progress against targets and inform improvements to increase impact. 33. Proposed priority 6: Leverage tax policies to encourage better diets. Policies to encourage healthier diets can offer quick wins that have translated into health gains in other countries. The government can optimize its current value-added tax (VAT) and tax policies to reduce Fijians’ high consumption of salt and sugar, which fuels the diabetes epidemic, and to encourage the consumption of healthier food. Fiji should also seek to eliminate trans-fats from the food chain as their consumption substantially raises the risk of heart disease. The implementation of MHMS/Ministry of Education forthcoming Healthy Canteen Policy (estimated to be approved by Cabinet by May 2025) offers a good opportunity for this. It can be complemented by efforts to educate the public to choose healthier options through improved labelling in restaurants, shops, and market stands. Many countries have adopted successful policies in this area and their knowledge on how to achieve goals like the elimination of trans-fats can be leveraged and adapted to the Fijian context. Table 3 Indicative costing and timeline for reform area 2 Implementation of reform area 2: proposed phasing and indicative costing of interventions Phase I over the first 1-3 years Followed by gradual roll-out (phase II) over 4-7 years Enhance capacity for risk profiling, screening Mainstream tailored risk profiling thanks programs in workplaces, schools and PHC FJ$7m to digital NCD registries and roll-out FJ$9m facilities, with data entered on digital platform screening programs in all PHC facilities. and sex-disaggregated NCD registries. Enable healthy behavior and tackle risk Implement Healthy Canteen Policy, factors among children and youth in school FJ$20m eliminating trans-fats from school FJ$2m (e.g., HSP). canteens. Catalyze population-level behavior change FJ$ Operating costs of school-based (behavior change communication, digital 16.5m FJ$6m prevention. reminders, availability of treatment). Remove salt and sugar from VAT exemption Minimal Expanded behavior change interventions and maximize SSB policies. (possible FJ$22m on NCD prevention and control. revenue) Phase 1 Phase 2 Total Indicative Cost: FJ$43.5m Total Indicative Cost: FJ$39m (US$19.5m) (US$18m) 18. Underweight, stunted, wasted, overweight, or obese. 22 Fiji Health Sector Review Reform area 3: Build a modern health workforce for the future 34. A new human resource model combined with PHC upgrades will improve access to care. Meeting Fijians as close to home as possible with health promotion and preventative care will require a combination of solutions that leverage Fiji’s existing human resources in health and that are tailored to the needs of the population they are entrusted to serve. 35. Proposed priority 7: Retain and deploy Fiji’s healthcare workers to meet the needs of the population they serve. Fiji’s density of SHWs is in line with the WHO minimum threshold when both public and private sector SHWs are considered, however, the current allocation of SHWs is uneven (Annex 10). A comprehensive revision of the role delineation policy, which defines the range and services to be delivered at different levels of care and more flexible staffing models can be developed, to align resources with workloads and needs in urban and rural settings (Table 4). Given the remoteness of many communities in Fiji, interventions to enhance basic care, including NCD prevention and control will benefit from upskilling the existing cadre of approximately 1,640 community health workers (CHWs). International studies that assessed interventions where appropriately supervised and equipped CHWs were trained to deliver blood pressure monitoring and counselling for lifestyle behavior change found they improved health and were cost effective.19 The updating of CHWs roles and responsibilities, supervision, training, and renumeration packages can be informed by global lessons that can be adapted and tested in a phased approach to meet Fiji’s needs, with possible variations in urban versus rural areas. It is also important to monitor the effects of recent salary increases in the public sector, emigration trends, and the impact of phase I interventions on SHW retention. Monitoring non-monetary drivers of resignations such as unclear career paths, poor working conditions, and burnout is an important element of an updated and improved human resource strategy. Table 4: Strategies for the reallocation of public sector skilled health workers • Redeploy SHWs from low utilization hospitals to high utilization PHC facilities and update role delineation plan. • Shift resources toward PHC facilities • Create a roster of weekly PHC outreach activities from hospitals to understaffed health centers. • Enable subdivision administrators (cost center managers) to allocate and rotate staff with greater flexibility between • Create greater flexibility for rotation of SHWs facilities within their subdivision, adding resources to high- utilization facilities when appropriate. 36. Proposed priority 8: Build human resource capacity to improve adherence to clinical guidelines and skills to use data for decision making. Studies have identified clinical training gaps at all levels of the health system. Fiji will update key clinical guidelines which need to be disseminated through in-person and online training. Further skills development in digital literacy, using data for decision making and improved patient outcomes, are worthwhile investments as part of Fiji’s efforts to digitize its health sector. In collaboration with teaching institutions such as the Fiji National University (FNU), the Pacific Community, and an international partner institution, an assessment should be carried out to identify opportunities to improve SHW training materials, including on NCD screening and treatment guidelines. Systems-level training, such as the World Bank Health Systems Flagship programs can also be offered, tailored to the Pacific and building on recent analytical assessments such as the Primary Health Care Performance Initiative (PHCPI) published in January 2024. Considering the high prevalence and tolerance of risk factors such as obesity, an independent study into the attitudes of health workers will help to improve the effectiveness of health promotion and adherence 19. Gamage et al. 2020. “Effectiveness of a Scalable Group-Based Education and Monitoring Program, Delivered by Health Workers, to Improve Control of Hypertension in Rural India: A Cluster Randomized Controlled Trial.” PLOS Medicine; Krishnan et al. 2019. “Cost-effectiveness and Budget Impact of the Community-based Management of Hypertension in Nepal Study (COBIN): A Retrospective Analysis.” Lancet Global Health 7(10):E1367-74; Jafar et al. 2020. “A Community-based Intervention for Managing Hypertension in Rural South Asia.” The New England Journal of Medicine 382(8):717-26. Fiji Health Sector Review 23 to the CRAM protocol. A broader capacity building program to strengthen the monitoring and evaluation capacity on program performance and cost effectiveness will help create a health systems performance culture that is evidence based and outcomes focused. Table 5: Indicative costing and timeline for reform area 3 Implementation of reform area 3: proposed phasing and indicative costing of interventions Phase I over the first 1-3 years Followed by gradual roll-out (phase II) over 4-7 years Update the role delineation policy and MHMS Roll-out SHW redeployment and capacity HR strategy for improved redeployment and FJ$5m building on data for decision making FJ$2.5m guidelines for SHWs. (building on digitization of the sector). Upskill and strengthen governance of community health workers (CHWs) and CHW / auxiliary worker scale-up programs FJ$16m FJ$37.5m auxiliary worker programs (training, digital (training, digital tools, and allowances). tools and allowances). Strengthening retention of nurse and Stock-taking of current training programs auxiliary programs (e.g., dieticians) FJ$2.5m (initial and continued education) and testing of following training and labor-force FJ$5m new materials. assessments, including on the impact of recent reforms. Phase 1 Phase 2 Total Indicative Cost: FJ$23.5m Total Indicative Cost: FJ$45m (US$10.5m) (US$20m) Reform area 4: Strengthen stewardship and data for a modern health system 37. Proposed priority 9: Strengthen public sector capacity for strategic purchasing from the private sector to get better value for money from private providers. The private sector is playing a growing role in Fiji’s health system. With careful planning, it could be leveraged to support health service delivery, including among under-served populations in urban settings. To fully harness the contribution of the private sector, the GoF can strengthen its capacity for strategic purchasing in MFSPNDS and MHMS through capacity building and learning exchanges to countries that have successful public-private sector engagement models in place. Existing programs to purchase services from the private sector could, if better monitored, be transformed into strategic opportunities for screening, preventive care, and health promotion. The GoF could build on its existing efforts to harness private general practitioners and healthcare providers to expand the reach of its NCD prevention programs. Evaluations to ascertain the cost effectiveness of current and future arrangements are necessary to assess value for money. 24 Fiji Health Sector Review 38. Proposed priority 10: Strengthen public financial management capacity. This review has identified three areas of focus to support MHMS in close coordination with MFSPNDS on broader PFM reforms: (i) Strengthen budget execution across standard expenditure groups and capital works: This includes creating a cross-sectoral task team (MFSPNDS, the Ministry of Civil Service, local government, and MHMS) and conducting a root-cause analysis of MHMS’ poor budget execution, with guiding actions for each government agency to take. (ii) Strengthen MHMS capacity for the roll out of GoF’s updated financial management information system (FMIS) and ensuring interoperability with sector specific tools. This transition will be accompanied by capacity building activities to use the enhanced FMIS systems and data for improved accountability and evidence-based decision making. (iii) Strengthen capacity to manage public finances effectively in emergencies. This includes building capac- ity to enable Fiji to access context-appropriate surge financing instruments, strengthen agile treasury systems (e.g., ensure appropriate procurement rules are triggered during an emergency), and set up systems to prioritize and track climate preparedness spending. Table 6: Indicative costing and timeline for reform area 4 Implementation of reform area 4: proposed phasing and indicative costing of interventions Phase I over the first 1-3 years Followed by gradual roll-out (phase II) over 4-7 years Strengthen capacity for strategic purchasing of Enhance digital capacity for strategic FJ$1m FJ$2.5m PHC services from the private sector. purchasing. Undertake bottleneck analysis and Enhance digital capacity to support PFM at recommendations for improved MHMS budget FJ$0.3m FJ$2.5m MHMS. execution. Strengthen PFM capacity at MHMS. FJ$1.7m Phase 1 Phase 2 Total Indicative Cost: FJ$3m Total Indicative Cost: FJ$5m (US$ 1.3m) (US$ 2.3m) Fiji has also obtained a grant from The Pandemic Fund to strengthen its pandemic preparedness 39. through a One Health approach. Fiji’s intersecting climate, antimicrobial resistance (AMR), zoonoses, and infectious disease threats, compounded by increasing human health vulnerabilities related to NCDs and social inequities, demand a One Health approach to all-hazards emergency preparedness and response. The activities align with an overall health systems transformation and center around three areas: (i) health emergency workforce capacity; (ii) community-based surveillance; and (iii) One Health information management and analysis. They represent a paradigm shift from reactive, response-based approaches to a deeper understanding of multi-hazard systemic risk, and investment in pre-emptive and anticipatory measures. Fiji Health Sector Review 25 Table 7: Indicative costing totals over seven years (two phases of implementation) Reform Area Phase I (millions) Phase II (millions) FJ$73 FJ$88–108 1. Redesign health service delivery to meet the health challenges of the Pacific (US$33) (US$40–49) FJ$43.5 FJ$39 2. Spur the adoption of healthy behaviors in the population (US$19.5) (US$18) FJ$23.5 FJ$45 3. Build a modern health workforce for the future (US$10.5) (US$20.2) FJ$3 FJ$5 4. Strengthen stewardship and data for a modern health system (US$1.3) (US$2.3) FJ$143 FJ$177–197 Health System Reform, Total (excluding CWM rebuild) (US$64) (US$80–89) We estimate that implementing the activities proposed in this report would amount to between 40. FJ$320 million and FJ$340 million20 or approximately FJ$50 million annually over seven years. This aggregate includes spending on physical and digital infrastructure under the four reform areas identified to achieve the vision for Mo Bulabula ka Bula Balavu. This includes upgrades in three buildings at CWM prioritized by the CWM management team: the Children’s Ward, the Maternity Ward, and the 1965 building. The full costs of the CWM renovation are not included in this estimate but are expected to be considerably larger. Recurrent spending arising from investments across the four reform areas, such as salaries and training of CHWs, or maintenance of physical and digital infrastructure, will continue into the period beyond the initial seven years, at about 10 percent of the initial aggregate investment. The above investments promise substantial gains 41. Successfully implementing these proposed investments could enable Fiji’s health outcomes to catch up with peer UMICs by 2040 and is likely to yield high rates of return. We conservatively estimate that annual rates of return to investment, after adjusting for inflation, would range from 7.6 percent to 10.2 percent over the period from 2024 to 2050, with the higher return reflecting the inclusion of the monetary value of longevity gains as part of the benefits. By way of comparison, the long-term real rate of growth of the S&P Index is 6.4 percent. In other words, the returns on investment are comparable to high-yielding index funds. Even if it took Fiji until 2050 to catch up with its UMIC peers, the annual real rates of return would range from 4 percent to 8 percent for the period from 2024 to 2050, which is still very impressive. 42. These health gains would also translate into significant economic gains. Achieving the health out- comes of a typical UMIC by 2040 would add 0.3 percentage points to the annual real GDP per capita growth rate over the period from 2024 to 2050. Catching up by 2050 would mean the annual growth rate of real GDP per capita would be a still-significant 0.15 percentage points higher relative to the status quo. If the monetary value of added life years associated with improved health was also included, the average annual economic gains of catching up by 2040 would increase to 0.90 percentage points over the period from 2024 to 2050 relative to the status quo, or 0.55 percentage points if equivalence is achieved by 2050. The greatest returns would come from health promotion, preventive services, and the effective manage- ment of chronic non-communicable conditions. 20. US$142–151 million. 26 Fiji Health Sector Review Fiji Health Sector Review 27 Introduction 43. This work was conducted at the request of the Ministry of Finance, Strategic Planning, National Development and Statistics (MFSPNDS). The MFSPNDS requested the World Bank’s support to conduct a systematic review of the health sector, to take stock of its current performance in tackling the country’s major health challenges and propose policy suggestions and programs for improving its performance. A working group with members from MFSPNDS and the Ministry of Health and Medical Services (MHMS) was established to facilitate data collection and oversee progress of the review. It was agreed that the review would cover: (i) health financing, including health expenditure projections to 2030; (ii) service delivery, including a supply-side readiness assessment of services at all levels of the health system, and redevelopment options for Colonial War Memorial Hospital; (iii) human resources for health, including assessment of stock and flow as well as projections of needs and supply to 2030; and (iv) options and models for the Government of Fiji (GoF) to engage private healthcare providers to improve population health outcomes. 44. This Health Sector Review (HSR) aims to identify transformational investments needed in Fiji’s health sector to respond to its changing epidemiological and demographic needs. This review includes an assessment of Fiji’s health system performance, and the health financing needs of Fijians going forward, followed by an analysis of proximate drivers of performance. These include quality of care in service delivery, efficiency in resource use, regulatory issues, infrastructure and human resource gaps, and behaviors at risk of causing ill health. It also highlights the economic implications of actions to address performance gaps using well known models of economic growth, and measurement of the economic value of longevity (the annexes of the report provide the detailed methodologies underpinning this work). The review concludes with policy suggestions and programs to guide short- and medium-term health sector investments to help address performance gaps and the resource implications of these investments. The expected returns on investment in terms of health and economic gains under different reform scenarios are reported using standard techniques of cost-benefit analysis, as described in the annexes to this review. MFSPNDS had also expressed interest in assessing the effectiveness of outsourcing health services to private practitioners, however, access to the data needed for a full assessment of private health service practitioners was not available during the drafting of this report. Fiji Upper middle-income country status Population (2023):21 FJ$11,164 912,302 Gross national income (GNI) per capita (2022):22 (US$5,390) 21. Population data from the Ministry of Health and Medical Services. 22. World Development Indicators DataBank, https://databank.worldbank.org/source/world-development-indicators. 28 Fiji Health Sector Review Report 28 Fiji Health Sector Review 45. The Republic of Fiji is an upper middle-income country (UMIC) with a total population of roughly 912,000. It is one of the most remote countries in the world—New Zealand is 2,000 kilometers away, Australia 3,000 kilometers, and the United States 5,000 kilometers. It extends across 18,000 square kilometers and more than 330 islands, of which about 110 are inhabited, although 87 percent of its total population lives on the two major islands, Viti Levu and Vanua Levu. 56 percent of Fiji’s population lives in urban areas, with 20 percent of the urban population living in informal settlements. Fiji is divided administratively into four divisions: the Central Division, which includes Fiji’s capital Suva and which has a population of approximately 385,000; the Western Division, with an approximate population of 349,000; the Northern Division, with a population of approximately 139,000; and the Eastern Division, which includes Fiji’s most remote outer islands, with an approximate population of 39,000. Fiji Health Sector Review 29 46. By International standards, extreme poverty is low, but Fiji trails its UMIC peers in delivering higher living standards to its population. Per capita income was estimated at FJ$11,791 (US$5,356)23 in 2022, comparable to that of Samoa and Tonga. Although the World Development Indicators (WDI) show that only 1 percent of the population lives on less than FJ$4.73 (US$2.15)24 per person per day, the international marker for extreme poverty, other measures of poverty tell a different story: prior to the Coronavirus disease (COVID-19) pandemic, 24.1 percent of Fijians lived below the national poverty line (the amount needed to meet the cost of basic needs) of FJ$5.97 (US$2.71) per person per day,25 and 52.6 percent lived below the World Bank poverty line for UMICs of FJ$15.08 (US$6.85) per person per day.26 Basic needs poverty in 2019–20 was highest in the Outer Islands, where 39.2 percent were living below the national poverty line, but rural areas in the main islands of Viti Levu and Vanua Levu also experienced much higher poverty (36.2 percent) than urban centers (13.8 percent).27 Projections suggest that poverty increased sharply during the pandemic but will return to pre-pandemic levels in 2024.28 47. The COVID-19 pandemic, coinciding with multiple tropical cyclones (TCs), caused an economic crisis in Fiji. International border closures halted tourism, which had generated 40 percent of gross domestic product (GDP) and 30 percent of jobs, with ripple effects across all segments of the economy. Concurrently, Fiji was hit by multiple tropical cyclones: Category 4 TC Harold in April 2020, Category 5 TC Yasa in December 2020, and Category 3 TC Ana in January 2021, illustrating the country’s high exposure to climate and disaster risks. The combined impact of COVID-19 and these natural disasters resulted in a 17.0 percent contraction in real GDP in 2020 and a further 4.9 percent fall in 2021.29 The economic crisis and the increased spending that was a key component of the GoF’s pandemic response, led to a sharp increase in public debt, from 52.1 percent of GDP in 2019 to 92.8 percent in 2021.30 48. The new government transitioned into office following elections in December 2022 and put a high priority on investments in health. The GoF established a Fiscal Review Committee that produced a blueprint for its socio-economic priorities, and highlighted Fiji’s need to invest in its physical and human capital,31 after a long period of underinvestment in health and education relative to its peers. The Fiscal Review Committee report suggested that Fiji’s economic growth would be under threat without additional investments in health and essential infrastructure to support social services32 and to enhance its resilience to shocks including extreme weather. It emphasized the need to strengthen preventative healthcare and improve physical infrastructure and the conditions of the healthcare workforce to mitigate the loss of skilled health workers (doctors, nurses, and midwives; SHWs).33 49. The economy has been recovering strongly on the back of a tourism rebound and improved confidence post-election, but its fiscal buffers and sources of growth remain limited. Since the reopening of international borders in December 2021, and on the back of rapid tourism growth, economic recovery has been strong with real GDP growth estimated at 20 percent in 2022 and 8 percent in 2023. However, the high level of public debt means the country has limited buffers to address future shocks, highlighting the scale and urgency of the need for fiscal consolidation, resilience building, and expanding the economy. Beyond building a more sustainable and higher value-added tourism sector, investment in education, health, connectivity, resilient infrastructure, and creating more opportunities for women will help create higher, sustainable, and inclusive growth. 23. World Bank. 2023. World Development Indicators DataBank. 24. Unless otherwise specified, exchange rate data used for conversions are from the World Development Indicators DataBank. See Annex 2 25. Fiji Bureau of Statistics (FBS). 2021. House Income and Expenditure Survey 2019-20: Main Report. Suva, Fiji: FBS. 26. World Bank. 2023. “Fiji: Poverty & Equity Brief”, October 2023. Washington, DC: The World Bank. 27. Fiji: Poverty & Equity Brief (op.cit.). 28. Fiji Bureau of Statistics. 29. World Bank. 2023. Fiji Public Expenditure Review 2023. Washington, DC: The World Bank. 30. Fiji Ministry of Finance, International Monetary Fund and World Bank staff estimates. 31. Human capital refers to the knowledge, skills, and health that people accumulate over their lives. 32. Such as health centers, hospitals, and schools, including improvements to water and sanitation. 33. Fiji Fiscal Review Committee: Report of the Committee. Suva, Fiji: Government of Fiji. 30 Fiji Health Sector Review 50. Fiji is also undergoing a process of demographic transition,34 creating the potential for a demographic dividend. The key driver of the demographic dividend is growth in the share of Fiji’s population who are working-age that has resulted from declines in fertility and infant mortality over the last half-century. The ratio of the working-age population (15–64 year-olds) to the number of dependents (aged 65 years and over, or younger than 15 years) in Fiji rose from 1.2 in 1973 to almost 1.9 in 2019, and this trend is expected to continue until 2045 according to United Nations population projections.35 Influential literature for East and South Asia has shown that rising shares of working-age populations can contribute to sharp rises in real income per capita.36 Declining fertility rates could also enable Fijian women to participate more widely in employment, further contributing to rising income per capita. 51. This Health Sector Review (HSR) aims to identify the investments needed to transform Fiji’s health sector to respond to its changing epidemiological and demographic needs, and its economic potential, considering the fiscal context. Fiji has major health sector achievements to its credit, such as its relatively low out-of-pocket (OOP) payments (expenditure borne directly by the patient at the point of care), something that Fiji should protect and maintain to strengthen universal health coverage. As this review highlights, however, there are significant unrealized economic and health benefits that Fiji could exploit by effectively addressing its large and growing burden of non-communicable diseases (NCDs). Moreover, health expenditure projections for Fiji suggest that, based on current trends, the gap between health spending needs and public resources is likely to grow sharply in the future. Fiji will need to design strategies that impose less stress on secondary and tertiary healthcare, while using public and private resources efficiently. The HSR investigates proximate drivers of Fiji’s health sector performance, including the effectiveness and efficiency of its healthcare delivery system, the role of private healthcare providers, gaps in infrastructure, human resources, and the health financing and regulatory environment. The review, which benefitted from guidance from the MFSPNDS and MHMS, concludes with a set of policy suggestions and programs to guide short-, medium-, and long-run health sector investments by the GoF, its development partners, and other stakeholders and the expected returns on investment under different reform scenarios (in health, longevity, and economic gains) between 2024 and 2050. 52. Fiji’s performance is compared to other countries using an international benchmarking exercise, in line with the 2023 Public Expenditure Review.37 This report draws upon data from domestic and global sources to measure Fiji’s health sector performance and for international comparisons. The benchmarking framework introduced in the Public Expenditure Review chose two sets of comparator countries, based on the following characteristics: (i) population; (ii) GDP per capita (current US$); (iii) tourism receipts; and (iv) island states or continental countries with lower foreign market access. These peer countries, drawn from the Caribbean and other small island developing states, will be used as reference points throughout this report: a. Aspirational peer countries: Barbados, Maldives, St. Lucia, and Mauritius. b. Structural peer countries: Belize, Jamaica, Samoa, and Tonga. c. At various points, Fiji will be compared to UMICs that have a strong track record of investments in primary care and health promotion. 34. Characterized by declining fertility rates and, notwithstanding the stagnation Fiji has experienced recently, long-term declines in infant mortality rates. 35. World Population Prospects 2022 (database). United Nations. https://population.un.org/wpp/. 36. Bloom and Williamson. 1998. “Demographic Transitions and Economic Miracles in Emerging Asia.” World Bank Economic Review 12(3):419-455; Bloom et al. 2010. “The Contribution of Population Health and Demographic Change to Economic Growth in China and India.” Journal of Comparative Economics 38:17-33. 37. World Bank. 2023. Fiji Public Expenditure Review. Washington, DC: The World Bank. Fiji Health Sector Review 31 32 Ministry Photo Credit: FijiFiji of Health Health and Medical Sector Review Services Health Outcomes Summary: • Relative to countries at its level of GDP per capita, Fiji underperforms in life expectancy at birth. • The gap between life expectancy at birth and the expected number of years lived in full health (healthy life expectancy) has widened over time. While Fijian women have a higher life expectancy, the gap between their life expectancy and healthy life expectancy is wider than for men. • Despite achieving high coverage rates in key services such as childhood immunizations, Fiji’s infant and neonatal mortality have stagnated since the turn of the century, after experiencing large gains in preceding decades. • Fiji faces recurrent infectious disease outbreaks, including in densely populated urban areas, where inadequate water supply, sanitation, drainage, and waste collection predisposes residents to diarrhea, typhoid, leptospirosis, and dengue. Emerging and re-emerging infectious diseases such as tuberculosis (TB) and HIV also present risks to population health. • Fiji has a high and increasing burden from NCDs driven by lifestyle factors, including unhealthy diets and insufficient physical activity. Premature mortality from NCDs is high in Fiji. Women have a higher prevalence of NCD risk factors compared to men. • The high disease burden of NCDs, alongside stagnant infant and neonatal health outcomes constitute a serious constraint in exploiting the economic opportunities generated by a growing working-age population. NCDs are also limiting Fijian’s women’s ability to participate in paid employment both because of the impacts on their own health and through their caregiving roles for adults with NCDs and children in the household. • If Fiji could lower its disease burden so as to attain the average health and mortality outcomes for UMICs by 2040 or 2050, the life expectancy at birth of an average Fijian would rise by almost 10 years, and the country would increase its annual average growth of real income per capita by between 0.15 and 0.30 percentage points. Health Sector Review Report 33 53. Relative to its level of economic achievement, Fiji’s life expectancy at birth is low, and has remained stagnant over the past three decades. A Fijian born today can expect to live 4.1 fewer years than would be predicted by Fiji’s national income per capita (i.e., compared to other countries with similar levels of income per capita). Although Fiji’s life expectancy at birth has increased from 65.5 years in 1990 to 68 years in 2019,38 it remains considerably lower than its aspirational, structural, and regional peers (Figure 6). Concerningly, life expectancy at birth is lower than the average for lower middle-income countries, about 69 years. Figure 6: Fiji’s life expectancy is low compared to other countries with similar levels of income Relative to its GDP per Relative Capita, to its Fiji GDP per underperforms Capita, ininLife Fiji underperforms Expectancy Life Expectancy at Birth at Birth 90 90 85 85 Life expectancy at birth (years) Life expectancy at birth (years) 80 80 75 75 70 70 Fiji 65 65 60 60 55 55 50 50 0 0 20000 20000 40000 40000 60000 60000 80000 80000 100000 100000 120000 120000 140000 140000 per capita GDP GDP (I$) (I$) per capita Other countries Other Countries Fiji Fiji Aspirational Aspirational Peers Peers Peers Peers Regional Structural Structural Regional PeersPeers Source: World Development Indicators DataBank, World Bank, https://datatopics.worldbank.org/world-development-indicators/. 54. The gap between life expectancy and the expected number of years lived in full health, the latter accounting for periods of illness and injury, has widened. Healthy adjusted life expectancy (HALE) has stagnated since 2010, while the length of time during which people are living but in poor health has increased from 7.8 years in 2000 to 8.4 years in 2019 (Figure 7). Women in Fiji have a higher life expectancy than men, but a wider gap with respect to HALE: on average there was a 10-year difference between women’s life expectancy and healthy life expectancy in 2019, compared to seven years for men. 38. Country Profiles: Fiji (database), United Nations Children’s Fund, New York (accessed November 13, 2023). 34 Fiji Health Sector Review Figure 7: The gap between healthy and overall life expectancy has increased and is wider for women than for men 70 10 12.0 68 9.5 10.0 9 8.9 66 9.4 9.6 8.5 9.3 64 8.1 8.0 Gap in Years Gap in Years 7.8 8.4 8 8.2 62 7.2 7.4 7.1 Years 7.5 6.0 6.8 60 7 68 4.0 6.5 56 6 2.0 54 5.5 52 5 0.0 2000 2010 2015 2019 2000 2010 2015 2019 Life expectancy at birth Health life expectancy at birth Gap Female Male Source: Global Health Observatory (database), World Health Organization, https://www.who.int/data/gho. 55. Despite considerable progress in childhood vaccination rates, child health outcomes have stagnated, and child malnutrition remains a major concern. Infant mortality rates in Fiji have been relatively stagnant since the late 1990s, after large gains during the preceding 30 years: the rate has declined by only 4 percent in the period since 2000, compared to a 59 percent decline in the three decades prior.39 These trends are underpinned by a lack of progress on newborn survival in the 28 days following birth (neonatal mortality) (Figure 8). Inadequate nutrition (only 43 percent of Fijian children are exclusively breastfed in the first six months of their life), and service delivery gaps related to skilled delivery, newborn resuscitation, cord care, and infection management are among the factors influencing neonatal mortality in Fiji. Although childhood immunization rates are high (96–99 percent for all doses of required vaccines in 2021),40 infant health outcomes have been challenged by recurrent vaccine preventable disease outbreaks, including measles, rubella, typhoid, and diarrhea. Based on the World Health Organization (WHO) Child Growth Standards, 7.2 percent of Fijian children aged under five are stunted (have a height-for-age which is two standard deviations below standard), and 7.4 percent of children under five are overweight (weight- for-height > +2 standard deviation from the median).41 56. Emerging, re-emerging, and endemic infectious diseases also pose a threat to service delivery and outcomes. Fiji faces recurrent infectious disease outbreaks, including in densely populated urban areas, where inadequate water supply, sanitation, drainage, and waste collection predisposes residents to diarrhea, typhoid, leptospirosis, and dengue. In the period between 2009 and 2019, the incidence of dengue fever increased from 49 to 326 cases per 100,000 population while the leptospirosis incidence increased from 18 to 124 cases per 100,000 population.42 Total TB incidence in Fiji increased from 21 cases per 100,000 population in 2009 to 66 per 100,000 in 2019.43 Fiji also reported a 260 percent increase in HIV cases in the period between 2010 and 2020 and the increase in infections has continued to rise at unprecedented rates, necessitating both immediate and long-term solutions, with strengthened preventative measures and increased availability of point of care testing. Other neglected tropical diseases 39. Administrative data report a 54 percent increase in infant mortality in the period between 2010 and 2020. 40. WHO. Immunization Dashboard: Fiji (database). World Health Organization (accessed November 13, 2023). https://immunizationdata.who.int/pages/ profiles/fji.html. 41. Fiji Bureau of Statistics. 2022. Fiji Multiple Indicator Cluster Survey 2021: Survey Findings Report. Suva, Fiji: Fiji Bureau of Statistics. 42. MHMS Annual Reports. https://www.health.gov.fj/publications/. 43. WHO. 2023 Global Tuberculosis Report. https://www.who.int/teams/global-tuberculosis-programme/tb-reports. Fiji Health Sector Review 35 such as filariasis are still contributing to preventable morbidity and deaths and often leave people vulnerable to other infections and diseases. As in many countries, the COVID-19 pandemic increased hospitalizations and deaths in Fiji and strained the health system as demand for services increased and investments were diverted to stop its spread.44 Figure 8: Neonatal and infant mortality rates in Fiji have stagnated since 2000 30 25 20 Per 1,000 Live Births 15 Infant Mortality Rate 10 Neonatal Mortality Rate 5 0 89 91 93 95 97 99 01 03 05 07 09 13 15 17 19 21 11 19 19 19 19 19 19 20 20 20 20 20 20 20 20 20 20 20 Year Source: World Development Indicators DataBank, https://databank.worldbank.org/source/world-development-indicators. 57. The share of the disease burden resulting from NCDs has been rising over time and the country has one of the highest rates of premature deaths from NCDs globally. NCDs rose from a share of 64.1 percent of Fiji’s disease burden in 1990 to 77.1 percent in 2019 (Figure 9). In 2019, almost 64 percent of all deaths due to NCDs were among Fijians aged between 30 and 70 years (i.e., prematurely).45 This is a high share of all NCD deaths compared to other countries with similar levels of income, and compared to Fiji’s aspirational, structural, and regional peers (Figure 10). Diabetes, ischemic heart disease, stroke, and chronic kidney disease are the major causes of death from NCDs among all Fijians. 44. WHO COVID-19 Dashboard (database). Geneva, Switzerland: World Health Organization (accessed November 13, 2023), https://covid19.who. int; Wilson et al. 2023. ”Country Case Study: Fiji.” Republic of Korea – World Bank Group Partnership on COVID-19 Preparedness and Response. Washington, DC: The World Bank. 45. WHO. Global Health Observatory (database), World Health Organization, https://www.who.int/data/gho. 2019 latest data available. 36 Fiji Health Sector Review Figure 9: Fiji is facing a rising non-communicable diseases burden 350 300 DALYs lost (in thousands) 250 200 250 100 50 0 1995 1990 1991 1992 1993 1994 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2012 2013 2014 2015 2016 2017 2018 2019 2011 Year Communicable diseases Injuries NCDs Source: Institute for Health Metrics and Evaluation. Figure 10: Fiji has one of the highest shares of premature deaths from non-communicable diseases in the world 90 Share of Deaths due to NCDs Occuring Prematurely (%) 85 80 Fiji 75 70 65 60 55 50 0 20000 40000 60000 80000 100000 120000 140000 0 GDP per capita (ppp) Other Countries Fiji Aspirational Peers Structural Peers Regional Peers Log. (Other countries) Source: World Population Prospects 2022 (database). United Nations. https://population.un.org/wpp/. 58. Across all age groups, Fiji’s disease burden from NCDs is high compared to countries with a similar epidemiological profile. As an illustration, Figure 11 compares Fiji’s disease burden from NCDs per person (in DALYs lost per capita) by age group, with that of Malaysia, a UMIC in the region at a broadly similar stage in its epidemiological transition. The per capita disease burden from NCDs is higher in Fiji than in Malaysia in all groups, with the difference especially noticeable from age 35 upwards. Fiji Health Sector Review 37 Figure 11: Non-communicable diseases impose a greater burden in Fiji than in an average UMIC country at all ages 3.0 DALYs from NCDs Lost per Person 2.5 2.0 1.5 1.0 0.5 0.0 4 9 4 4 9 4 9 4 9 4 9 4 9 4 9 4 9 4 9 + 0- 5- -1 -1 -2 -2 -3 -3 -4 -4 -5 -5 -6 -6 -7 -7 -8 -5 -9 95 10 15 20 25 30 35 45 50 55 60 65 70 75 80 85 40 90 Age Intervals (in years) Average UMIC country Fiji Source: Institute for Health Metrics and Evaluation. 59. The top ten risk factors for NCDs in Fiji relate to lifestyle, and there is evidence that the impact of lifestyle factors on the disease burden has been rising over time. Data from the Institute for Health Metrics and Evaluation (IHME) show that DALYs lost (per 100,000 population) in Fiji attributable to tobacco, diet, and low physical activity rose from 2009 to 2019 (Table 8). Data from food consumption surveys show that 63 percent of the adult population consume well above the recommended daily energy intake.46 Convenience foods, such as sugar-sweetened beverages, crackers, noodles, and fried foods, have replaced healthy, Fijian-grown traditional foods. As a result, Fijians have much higher salt and sugar intake than recommended levels (Box 1). 60. Changing diets and other lifestyle factors have been a particular issue for women, with intergenerational consequences. Among Fijian girls and women aged 15-49 years, 70 percent are overweight, and 45 percent are obese.47 More women (18.3 percent) also perform insufficient physical activity compared to men (11.3 percent). These factors are linked to women’s higher risks of diabetes and raised blood pressure, and significantly increases their NCD risks. Being overweight during pregnancy increases the risk of gestational diabetes and complications during pregnancy and delivery to the mother and child. A lack of diversity in mothers’ diets (and high-calorie food intake) translates into poor dietary diversity for their infants and young children, given their primary role in providing food in the household. This predisposes children to unhealthy food choices as they grow up. 46. Fiji’s most recent nutrition survey was conducted in 2015. 47. Fiji Bureau of Statistics. 2022, Fiji Multiple Indicator Cluster Survey 2021: Survey Findings Report. Suva: Fiji Bureau of Statistics and the United Nations Children’s Fund (UNICEF). 38 Fiji Health Sector Review Box 1: Salt and Sugar Intake in Fiji Mean sodium intake in Fiji is four times higher than the recommended daily intake range. An intake breakdown shows that salt was the main source for sodium (14 percent) in daily consumption with other sources provided by wheat flour products, fresh fish, seafood, and meat, including sausages.48 A population survey conducted in Fiji’s Central Division in 2023 found similar consumption patterns as well as limited knowledge about the sodium content in common household food items.49 Sugar consumption in Fiji is three times higher than the WHO recommended limit. The main sources of sugar in the diet are table sugars, honey, and related products (24 percent); non-alcoholic beverages (21 percent); and bread and bakery products (18 percent). The consumption of sugars, honey, and related products was higher in the iTaukei population while consumption in bread and bakery products as well as mixed cooked dishes was higher in Fijians of Indian descent and other groups. A population survey conducted in Fiji’s Central Division in 2022 found average sugar consumption of 74 grams per day. Table 8: Lifestyle factors remain the main drivers of disability adjusted life years in Fiji Top ten risks contributing to 2009 rank 2019 rank Change in DALYs per 100,000 people disability adjusted life years (DALYs) High fasting plasma glucose 1 1 1,090.2 High body-mass index 2 2 881.0 High blood pressure 3 3 680.5 Dietary risks 4 4 445.5 Tobacco 5 5 202.0 High low-density lipoprotein cholesterol 7 6 157.6 Malnutrition 6 7 -636.3 Air pollution 8 8 -393.0 Kidney disfunction 9 9 220.7 Low physical activity 10 10 113.8 Source: Institute for Health Metrics and Evaluation (2019 latest data available) and Global Burden of Disease 2019 Risk Factors Collaborators.50 Note: Red means an increase in DALYs between 2009 and 2019, green means a decrease in DALYs in that same period. 48. MHMS. 2015. National Nutrition Survey. 49. Silatolu et al. 2023. “Salt and Sugar Intakes of Adults in the Central Division of Fiji – Findings from a Nutrition Survey of over 500 People” Preprint. DOI: 10.21203/rs.3.rs-2896390/v1. 50. Global Burden of Disease 2019 Risk Factors Collaborators. 2020. “Global Burden of 87 Risk Factors in 204 Countries and Territories, 1990–2019: A Systematic Analysis for the Global Burden of Disease Study 2019.” Global Health Metrics 396(10258) 1223–1249. https://doi.org/10.1016/S0140- 6736(20)30752-2. Fiji Health Sector Review 39 An Economic Opportunity under Threat 61. Fiji’s ability to exploit the demographic dividend depends on its working-age population remaining healthy and in employment. The high burden of NCDs, including premature mortality, among Fijian adults directly limits their productive employment. Moreover, children in Fiji are facing a worrying triple burden of malnutrition with concerning trends of undernutrition, overweight, and micronutrient deficiencies impacting the country’s youth. This leaves them more susceptible to infections, and puts them at increased risk of developing diabetes and heart disease, lower cognitive ability, and worse learning outcomes. Upon reaching adulthood, malnourished children face increased risk of NCDs, reduced productivity, and premature mortality.51 62. Declining fertility has not led to the expected increase in participation in work among Fijian women, with NCDs likely a contributing factor. International evidence suggests that a 1-point reduction in the fertility rate raises women’s work participation by 5–9 percentage points, but this has not been observed among Fijian women (Figure 12). Although the high burden of NCDs among women in Fiji is one factor affecting their work participation, they also have a primary caregiving role for children as well as for adult household members with NCDs.52 Fijian women spend 2.5–2.9 more time than men on unpaid domestic work, with elderly caregiving being an important component.53 This makes it harder for them to participate in or return to the workforce, forgoing significant economic opportunities for themselves, their families, and for Fiji. Almost two-thirds of elderly Fijians live with their family members, especially those who are in poor health.54 Although data for Fiji are scarce, research in high-income countries has found that adult caregiving responsibilities reduced women’s participation in employment by 2–6 percentage points,55 with a roughly similar impact from unpaid childcare responsibilities.56 Figure 12: Work participation among Fijian women has been stagnant despite declining fertility rates (1990–2019) 4 90.0 3.5 80.0 Labor Force Participation on 15-64 (percent) 70.0 3 60.0 2.5 Total Fertility Rate 50.0 2 40.0 1.5 30.0 1 20.0 0.5 10.0 0 0.0 1990 1992 1994 1996 1998 2000 2002 2004 1990 2008 2010 2012 2014 2016 2018 2020 Fertility Rate W-LFPR M-LFPR Source: World Development Indicators database, World Bank. Note: W-LFPR: women’s labor force participation rate; M-LFPR: men’s labor force participation rate. 51. Alves and Alves. 2024. “Early-life Nutrition and Adult-Life Outcomes.” Jornal de Pediatria 100(S1):S4-S9. 52. IFC. 2019. Tackling Childcare: The Business Case for Employer-Supported Childcare in Fiji. International Finance Corporation. https://www.ifc.org/en/ insights-reports/2019/tackling-childcare-the-business-case-for-employer-supported-childcare-in-fiji 53. ADB. 2016. Fiji: Country Gender Assessment 2015. Manila: Asian Development Bank; Johnston. 2021. Understandings and Experience of Dementia in Fiji. Doctoral dissertation. Townsville, Australia: James Cook University. 54. Seniloli, Kesaia, and Rupeni Tawake. 2014. ”Living Arrangements of the Elderly in Fiji.” The Journal of Pacific Studies 34(12). 55. Leigh. 2010. “Informal Care and Labor Market Participation” Labour Economics 17(1):140-9; Maestas, Messel, and Truskinovsky. 2021. “Caregiving and Labor Force Participation: New Evidence from the Survey of Income and Program Participation” Working paper no. W120-12. Madison: University of Wisconsin-Madison. 56. Anukriti et al. 2023. Filling the Gaps, Childcare Laws for Women’s Economic Empowerment. Washington, DC: World Bank. 40 Fiji Health Sector Review 63. Healthier older populations can help reduce the burden of caregiving among adult women, enabling them to take greater part in paid work. Healthy ageing could also permit older individuals to contribute by providing childcare. Studies from China show that elderly childcare support can translate into increases in workforce participation rates among adult women of 28–43 percent.57 Similar findings from the United States, Australia, Burkina Faso, and elsewhere suggest that childcare support could lead to large positive work participation effects for women, with knock-on effects on their earnings and household well-being.58 64. Improving the health of Fiji’s population would translate into significant economic gains. Were Fiji to lower its disease burden to that of a typical UMIC by 2040, it would enjoy a significant boost to its health outcomes, including life expectancy gains of almost 10 years from their current levels, and an extra 0.30 percentage points of growth in real GDP per capita annually during 2024–50 relative to the status quo (Annexes 4 and 5). If it were to catch up by 2050, the extra growth would be a still-significant 0.15 percentage points annually, alongside life expectancy gains of 10 years. When the monetary value of longevity gains is also included, the average annual rate of growth could increase by an additional 0.90 percentage points relative to the status quo if Fiji catches up with a typical UMIC by 2040 or 0.55 percentage points per year if it does so by 2050. 57. Du et al. 2019. “Grandparent-provided Childcare and Labor Force Participation of Mothers with Pre-school Children in Urban China” China Population and Development Studies 2:347-68; Miao et al. 2023. “Impacts for Family Care for Children and the Elderly on Women’s Employment: Evidence from Rural China” Frontiers in Psychology 14; Shen et al. 2016. “Coresidence with Elderly Parents and Female Labor Supply in China” Demographic Research 35:645–70. 58. Ishida et al. 2022. “Heart Disease and the Economic Contributions of Elderly Men and Women: Evidence from Australia” Australian Economic Review 55(4):461-76. Ajayi et al. 2022. “The Effects of Childcare on Women and Children: Evidence from a Randomized Evaluation in Burkina Faso” Policy Research Working Paper no. 10239. Washington, DC.: The World Bank. Fiji Health Sector Review 41 Photo Credit: Fiji Ministry of Health and Medical Services 42 Fiji Health Sector Review Health Financing Summary: • Fiji spends less on health than other countries with similar levels of income per capita, despite recent increases in both nominal and real health spending. • Expenditure on health in Fiji is largely from the public sector and domestically financed, albeit with growing reliance on privately provided healthcare services financed by public resources. • Private spending on health has been rising steadily, due to increases in out-of-pocket (OOP) expenditure and private insurance spending, raising concerns about efficiency in resource use and equity. However, OOP expenditure for health is relatively low, a feature Fiji should strive to protect, although not at the cost of forgoing necessary health services. • Human resources for health account for a large and increasing share of expenditure at just under 60 percent. • More than three-quarters of government health expenditure was allocated to hospitals in 2019, including a substantial portion of the spending on primary care, suggesting that primary healthcare (PHC) services are being disproportionately delivered in hospital settings rather than health centers and nursing stations. • Increasing incomes, population size, ageing, urbanization, and other factors such as climate change, will put significant upward pressure on Fiji’s health spending and affect its composition. In a business- as-usual scenario, Fiji’s total health expenditure is expected to rise to almost 7 percent of GDP by 2050, well above its current level of 4–5 percent, and per capita health spending needs are expected to rise by between 200 and 250 percent. • In the absence of increased government health spending or interventions to increase the efficiency of resource use, including climate-smart approaches and prioritizing prevention and primary care, Fijians will have to increasingly rely on their own resources to fund their healthcare, or forgo care, with adverse implications for efficiency and equity. Health Sector Review Report 43 65. Spending on health per capita in Fiji has been rising in nominal terms over the last two decades but growth in real terms has largely been concentrated in the period after 2015. Per capita health spending in Fiji more than tripled in nominal terms between 2000 and 2021, from FJ$163 (US$75) to FJ$550 (US$250) (Figure 13). In contrast, real spending hovered at around FJ$311 (US$150) per capita (in constant 2021 dollars) between 2000 and 2016 but then grew by 40 percent between 2016 and 2021. Despite recent increases, Fiji’s spending on health is lower than its aspirational, regional, and structural peers, whether taken as a share of GDP or considered in per capita terms (Figure 14). Nine percent of Fiji’s government expenditure was on health, lower than both structural and aspirational peers. Figure 13: Per capita spending on health has been rising in nominal terms, but only grown in real terms in recent years (2000–21) 300 250 200 150 100 50 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2022 2021 Per capita Per capita real (2021) Source: Global Health Expenditure Database, World Health Organization. https://apps.who.int/nha/database. 44 Fiji Health Sector Review Figure 14: Fiji’s spending on health is low relative to its aspirational and structural peers (2021) 12 1,600 1,400 Spending per Capita (US$) Percent of GDP (%) 10 1,200 8 1,000 6 800 600 4 400 2 200 00 Fiji Samoa Saint Lucia Mauritius Fiji Tonga Belize Barbados Jamaica Jamaica Saint Lucia Mauritius Belize Maldives Samoa Barbados Maldives Tonga Aspirational peers Structural peers Aspirational peers Structural peers Source: Global Health Expenditure Database, World Health Organization. https://apps.who.int/nha/database. 66. Current health financing arrangements in Fiji are progressive, with health spending largely publicly financed. In 2021, 67 percent of current health expenditure (CHE) was financed by the government from general revenues, and 33 percent was from private spending (including OOP spending and private health insurance).59 Asante et al. (2017) assessed government spending on health in Fiji as being equitably distributed.60 Although OOP payments in Fiji have so far been concentrated in richer households and catastrophic health spending is low, these low OOP payments could also reflect forgone care. Externally funded expenditure has varied between 2 and 10 percent of CHE over the past decade, reaching 10 percent in 2021 during the COVID-19 pandemic. Given pre-COVID trends and the international aid environment, external funding is unlikely to increase in the future. 67. Private spending as a share of total health expenditure has been rising steadily over the last two decades (Figure 15). Private health spending primarily comprises spending supported by private voluntary health insurance (the main component of “other private expenditure”) and OOP spending, both of which have been rising over time in Fiji. Recent analyses for Fiji show that private health insurance is concentrated among the economically better off population.61 59. Global Health Expenditure Database, World Health Organization. 60. WHO. 2023 Global Tuberculosis Report. https://www.who.int/teams/global-tuberculosis-programme/tb-reports. Asante et al. 2017. “Financing for Universal Health Coverage in Small Island States: Evidence from the Fiji Islands.” BMJ Global Health 2017;2: e000200. 61. Ibid. Fiji Health Sector Review 45 Table 9: Spending on health in Fiji is largely public and domestically funded (2021) Current health As percent of current health As percent of current expenditure expenditure Gov. spending on GDP per health expenditure health (percent capita Country per Private total gov (USD) – percent spending) 2022 capita Public (excl. OOP Domestically Externally GDP (USD) OOP) funded funded Fiji 250 5.38 67 15 18 90 10 9 5,356 Aspirational 897 12 44 23 33 93 7 12 13,827 peers Structural 306 13 72 15 13 87 13 13 5,301 peers Source: Global Health Expenditure Database, World Health Organization, 2021 or latest data available. GDP data: World Development Indicators DataBank, World Bank. Note: External spending can be either public (i.e., channeled through the Ministry of Health and Medical Services) or private (i.e., through NGOs). In the case of Fiji, 43 percent of external funding was public, and 57 percent private. Figure 15: Private health spending in Fiji is rising 600 500 400 Constant FJ$ Per Capita 300 200 100 0 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Government Expenditure OOP Expenditure Voluntary Health Insurance External Private Expenditure Other Private Expenditure Source: Global Health Expenditure Database, World Health Organization. https://apps.who.int/nha/database. 46 Fiji Health Sector Review 68. Private insurance in Fiji is highly fragmented and characterized by high administrative costs. In 2022, there were 365 group insurance policies and 2,108 individual policies.62 Small insurance pools such as these tend to be characterized by large variations in claims, leading to shallow coverage and higher reinsurance expenses, limiting their risk protection benefits and the efficient use of resources. The individual medical insurance policies offered by Fiji Care and BSP Life (two of the companies offering health insurance cover in Fiji) illustrate shallow coverage: restrictions on the upper age at entry (50–55 years) into insurance, an upper age limit on insurance coverage (65 years), and limits on payouts for several categories of healthcare services. Loss ratios (the ratio of claim payments to premium income) for private medical insurance in Fiji have mostly ranged between 50 percent and 80 percent indicating overheads of 20–50 percent (Figure 16), higher than the rates of 7–28 percent in Organisation for Economic Co- operation and Development (OECD) countries.63 While underestimation of the risk profile can also raise loss ratios, this is unlikely to be the case in Fiji given the persistence of the high loss ratios over several years. Private insurance overhead costs are also considerably larger than the overhead costs of the Ministry of Health and Medical Services (MHMS) health spending in Fiji, estimated at 8–10 percent in National Health Accounts data.64 Figure 16: Loss ratios for private medical insurance in Fiji are very high 50000 100.00 45000 90.00 40000 80.00 35000 70.00 30000 60.00 FJ$ 000s) 25000 50.00 20000 40.00 15000 30.00 10000 20.00 5000 10.00 0 0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2011 Gross Premiums Gross Claims Loss Ratio Source: Reserve Bank of Fiji, Annual Reports, various years. 62. Reserve Bank of Fiji. 2023. Annual Insurance Report 2022. 63. Mathauer and Nicolle. 2011. “A Global Overview of Health Insurance Administrative Costs: What are the Reasons for Variations Found?” Health Policy 102(2-3):235-46. 64. MHMS. 2017. Fiji National Health Accounts: National Health Expenditure 2011-15. Suva: Government of Fiji. Fiji Health Sector Review 47 69. Without adequate regulatory safeguards, private health insurance in Fiji could have an adverse impact on preventive care, health expenditure inflation, and equity. Insurers are incentivized to provide shallow coverage directed towards healthier (and younger) patients with limited prevention needs and will underemphasize prevention. Unhealthy patients will tend to be dropped as they reach the upper age limit for private coverage, imposing costs on the public system. Data from India and Malaysia, countries with a similar a regulatory environment to Fiji, suggest a sharp drop-off in coverage by private insurers occurs at older ages.65 Insurance companies’ role as primarily third-party payers in Fiji also means they may be more inclined to pass on increases in treatment costs by local or international private providers in the form of premium increases to the insured, increasing the risk of health expenditure inflation. Equity may also be at risk if expanding private insurance coverage increases incomes for private healthcare providers, leading to more public sector doctors, nurses, and allied health workers moving to the private sector. 70. Fiji should seek to keep OOP expenditure on healthcare low. At 18 percent, Fiji has the highest OOP share of current health spending amongst Pacific Island countries (PICs) although the share is lower than most aspirational peer countries (Figure 17). The share of Fiji’s OOP spending is also low relative to its health spending and income per capita. But OOP spending per capita in Fiji has increased substantially since 2001. More than half of OOP expenditure is on drugs and medical goods, and just over one-third on curative care.66 International evidence shows that OOP payments contribute to increasing health inequality by linking healthcare access with ability to pay and can deter or delay the use of needed health services.67 Their fragmented nature also means that they are inefficient. Low OOP payments for healthcare are a key feature of universal health coverage and one that Fiji should actively seek to maintain. Moreover, efforts to reduce OOP payments via increased pooling of public and private resources for healthcare could enable the more effective purchase of healthcare services and lower the risks of higher costs of health services in the private sector. Figure 17: Out-of-pocket expenditures in Fiji are low compared to aspirational peers but high compared to the region and rising (2021) 50 45 40 OOP Share of Current Health 35 Expenditure, 2021 30 25 20 15 10 5 0 Solomon Islands Jamaica Vanuatu Saint Lucis Samoa Kiribati Tuvalu Belize Tonga Nauru Barbados Palau Mauritus Maldives FSM RMI Fiji Aspirational Peers Structural Peers Other Pacific Islands Countries 65. World Bank. 2023. Public Expenditure Review: Malaysia. Washington, DC: The World Bank. 66. Government of Fiji. Fiji Health Accounts: 2014-2019: National Health Expenditure. 67. Wagstaff, Eozenou, and Smitz. 2019. “Out-of-Pocket Expenditures on Health: A Global Stock take” Policy Research Working Paper 8808. Washington, DC: The World Bank. 48 Fiji Health Sector Review 60 50 40 OOP per Capita in Constant US$ 30 OOP as % of Current Health 20 Expenditure 10 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 Source: Global Health Expenditure Database, World Health Organization. https://apps.who.int/nha/. 71. Human resources for health (HRH) account for a major and increasing share of health expenditure, followed by drugs and medical equipment (Figure 18). As might be expected, given their central role in the provision of services, spending on HRH is increasing in Fiji, both as a share of total health spending and in nominal terms, growing from FJ$108 million (US$52 million) in 2017 to FJ$209 million (US$95 million) in 2022. In 2022. HRH accounted for 58 percent of Fiji’s MHMS spending in 2022. This share is at the upper end of the range observed in low- and middle-income countries, but is comparable to Malaysia, also an UMIC (60 percent in 2019),68 and slightly higher than the 56 percent share in Mauritius.69 HRH accounted for about 73 percent of health spending on average in OECD countries,70 and 55 percent in 33 low- and middle-income African countries,71 Spending on drugs and medical equipment is the second largest category of expenditure. Between 2016 and 2022, spending on this category fluctuated between 12 and 19 percent of MHMS expenditure, increasing from FJ$33.5 million (US$16 million) to FJ$49.0 million (US$22 million). The introduction of the Free Medicine Program in 2015 translated into an increase in government spending on drugs and medical equipment (Figure 19). Figure 18: Staff costs make up a large and rising share of government health spending in Fiji 2022 58 13 13 2021 58 14 13 2020 63 15 11 Average 2017-2019 52 18 15 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Staff cost Other goods and services Medical supplies Grants Capital Overseas medical referrals Source: Fiji Ministry of Finance BOOST data. 68. World Bank. 2023. Public Expenditure Review: Malaysia. Washington, DC: The World Bank. 69. Ministry of Health and Wellness. 2022. “Vote 18-1: Ministry of Health and Wellness.” Government of Mauritius. 70. Hernandez et al. 2006. ”Measuring Expenditure for the Health Workforce: Evidence and Challenges.” Background paper for the World Health Report. Geneva: World Health Organization. 71. Toure et al. 2023. ”Health Expenditure: How Much is Spent on Health and Care Worker Remuneration? An Analysis of 33 Low- and Middle-income African Countries.“ Human Resources for Health 21, 96 (2023). Fiji Health Sector Review 49 Figure 19: Government spending on drugs increased with the introduction of the Free Medicine Program in 2015 90 80 70 60 46.7 41.5 50 25 13.5 FJ$ million 40 9 30 20 35 39 39 35 39 10 0 2014 2015 2016-17 2017-18 2018-19 Drug expenditure (excluding government) Government drug expenditure Source: Fiji National Health Accounts Report 2014–2019. Note: Drug expenditure (excluding government) includes prescription and over the counter drugs only. From 2015, government drug expenditure includes the Free Medicine Scheme expenditure. 72. Hospitals account for a major share of public spending in Fiji. Spending per capita on PHC in Fiji increased from FJ$235 (US$112) in 2016 to FJ$313 (US$145) in 2019,72 but the share of PHC in national health spending fell from 64 percent of in 2016 to 57 percent in 2019. Although more than half of health expenditure is directed to PHC, 79 percent of government health expenditure was on hospitals in 2019 (Figure 20). The high share of hospital spending, and the utilization of resources allocated to hospitals have implications for efficiency in resource use as discussed below. Figure 20: Government health expenditure is predominantly spent in hospitals 19.8 27.6 21.6 29.4 27.4 Expenditure (FJ$ million) 122.3 127.1 159.7 207.8 244.8 21.3 14 15 23.3 28.1 37.4 37.3 18.7 26.8 19.9 2014 2015 2016-17 2017-18 2018-19 Providers of ambulatory healthcare Providers of preventative care Hospitals Other Source: Government of Fiji. Fiji Health Accounts: 2014-2019: National Health Expenditure. 72. Global Health Expenditure Database, World Health Organization. https://apps.who.int/nha/database. 2019 latest data available. 50 Fiji Health Sector Review 73. Public financial management (PFM) is an important part of health system strengthening and ultimately in achieving progress in terms of universal health coverage. A 1-unit increase in public expenditure and financial accountability (PEFA) scores is associated with a reduction of the under-five mortality rate of about 14 per 1000 live births.73 The budget execution rate for MHMS has remained stable over the past five years, averaging 80 percent across all standard expenditure groups (SEGs), which is well below the low-income country average of 87 percent and equivalent to a public expenditure and financial accountability (PEFA) score of D,74 a sign of major deficiencies. There was some improvement during the 2023 fiscal year (FY23), with MHMS utilizing 88 percent of its total budget. MHMS’s average execution rate for the established staff SEG75 in the period between 2017 and 2023, lies at 93 percent, which is consistent with the average budget execution rate (95 percent) for the same group across all ministries. A key bottleneck for MHMS execution of the established staff SEG budget in FY23 is the post-pandemic departure of skilled health workers to the private sector and opportunities abroad. Capital construction is a major bottleneck in MHMS budget execution. The average execution rate for the capital construction SEG in MHMS was 64 percent in the period between 2017 and 2022, which is higher than other social sectors,76 but generally lower than infrastructure-heavy ministries such as the Ministry of Infrastructure and Meteorological Services, which reported an average of 88 percent. Broader challenges with capital budget absorption are related to procurement approval and decision processes which become amplified during emergency resourcing of activities, as well as limited procurement and contract management capacity. Addressing Future Healthcare Needs 74. Fiji’s population will continue to age, which will put further pressure on health spending. United Nations population projections77 suggest that the share of the population aged 65 years and over in Fiji will rise from around 7 percent in 2023 to more than 13 percent in 2050 (Figure 21). Older populations tend to be sicker, with a greater incidence of chronic NCDs than their younger counterparts, and therefore have higher healthcare needs. Population ageing will therefore generate pressures to increase health spending and create a growing need for social care. Although age-specific healthcare use and spending data are not readily available for Fiji, the experience of other countries, such as Malaysia and Australia, confirms that health expenditures per person rise with age (Figure 22). Fiji’s growing NCD burden and premature acquisition of NCDs will also contribute to increased demand for healthcare. Figure 21: The proportion of older people in Fiji is projected to rise substantially Share of People Aged 60 and over, and 65 and over in Fiji: (2000, 2023, and 2050) 16.0 14.0 Share in Total Population (%) 12.0 10.0 8.0 6.0 4.0 2.0 0.0 2000 2023 2050 60+ yrs 65+ yrs Source : United Nations (UN) Population Projections. 73. Moritz and Smets. 2019. Public Financial Management, Health Financing and Under-five Mortality: A Comparative Empirical Analysis. World Bank and Inter- American Development Bank. 74. PEFA assessment scores countries on a scale of A-D with A ranking the highest and D the poorest. A score of D in budget credibility reflects a budget execution deviation of 15 percent or worse from the original budget. 75. Established staff include medical, nursing, dental, paramedical, and administrative staff. The budget books do not allow a disaggregation by cadre of health workers. Ward attendants and sanitation and maintenance staff are included under SEG 2 i.e., wage earners. 76. The average execution of capital construction SEG by the Ministry of Education during the same time period was 47 percent. 77. World Population Prospects 2022 (database). United Nations. https://population.un.org/wpp/. Fiji Health Sector Review 51 Figure 22: Per capita spending by age in Fiji, using age-weights from Malaysia and Australia, suggest that healthcare spending per capita will rise with age 1800 1600 Spending per person (FJ$) 1400 1200 1000 800 600 400 200 0 <5 5_9 9_14 15_19 20_24 24_29 30_34 35_39 40_44 45_49 50_54 55_59 60_64 65_69 70_74 75_79 Age Intervals (in years) Malaysia Australia Source: Malaysia data: HPRA, IHSR, and IHP. 2013. Malaysia Health Care Demand Analysis. Putrajaya, Malaysia: Institute for Health Systems Research; Australia data: WHO. 2020. “How Will Population Ageing Affect Health Expenditure Trends in Australia and What are the Implications if People Age in Good Health?” World Health Organization & European Observatory on Health Systems and Policies. 75. Fiji is also expected to become increasingly urban, putting pressure on provision of health services especially in informal settlements. Urbanization is predicted to reach 61 percent of the total population by 2030. The number of people living in urban areas has grown by 57 percent since 1976 and urban populations are growing by 1.5 percent per year, more quickly than their rural counterparts (0.7 percent annually).78 There has been rapid growth of the population living in urban informal settlements (currently estimated at around 120,000 to 140,000 Fijians) with limited access to public services, including healthcare, water, and sanitation, and high levels of income and housing insecurity.79 This includes climate migrants from Fiji and other parts of the Pacific (see below). At the same time, populations in rural areas, 40 percent of whom are poor, will continue to need health services. 76. The effects of climate change will exacerbate Fiji’s disease burden and put its healthcare delivery system under stress. Fiji is highly vulnerable to a range of climate-related hazards, including rising sea levels, coastal erosion, large-scale urban and river flooding, cyclones, and extreme heat, with associated water shortages and salination of water supplies.80 Changes in precipitation and temperature patterns will entail an increase in frequency and severity of these events, damaging health infrastructure and possibly accelerating climate-related migration. This will lead to an increased peak demand for healthcare in case of extreme events, and longer-term changes in vector-borne disease and NCD patterns.81 Fiji has experienced an increase in the frequency of hot days (over 35°C) by 2 percent every decade since 1951. By 2040, hot days will be twice as frequent,82 which is likely to cause people to engage in less physical activity. Changing rainfall patterns and extreme weather could lower the availability of fresh produce, pushing people to rely on more processed food, in turn driving obesity and nutritional deficits. Poorer households that rely on fishing for subsistence are particularly at risk of malnutrition if coastal fisheries are lost.83 Without adaptation, annual heat-related deaths in the region could increase by 437 percent by 2050. 78. UN Habitat. 2012. Fiji National Urban Profile. Nairobi, Kenya: UN Habitat. 79. Report of the 2023 Fiji Fiscal Review Committee, p. 38; Naidu et al. 2015. “Informal Settlements and Social Inequality in Fiji.“ Journal of Pacific Studies 35(1):27-46. 80. World Bank. 2024. Think Hazard! Fiji. Available at: https://thinkhazard.org/en/report/83-fiji. 81. IFRC Climate Centre. 2021. “Climate Change Impacts on Health and Livelihoods: Fiji Assessment.” Red Cross Red Crescent Climate Centre. 82. Climate Change Knowledge Portal (database). https://climateknowledgeportal.worldbank.org/. 83. World Bank Group. 2021. Climate Risk Country Profile: Fiji. Washington, DC: The World Bank Group. 52 Fiji Health Sector Review 77. Changes in income, population size, and a mix of other factors will also influence Fiji’s future health expenditure. Rising incomes would allow people to afford more and better-quality healthcare, with studies estimating that a rise in national income per capita leads to a rise in health spending of the same proportion (or an income elasticity of one).84 Efficiency gains in service delivery and healthcare inflation rates (which differ from general inflation measures such as the consumer price index) and technological advancements will also influence future health spending requirements. Because of the difficulty of assessing these individual factors, most health expenditure forecasts use their combined effects, sometimes referred to as residual factors, excepting the impact of income growth and population ageing, which are forecasted directly. 78. Fiji’s health expenditure is forecasted to rise over the next 30 years. Our analysis used historical data and forecasts of GDP, data on health spending during 2000–19 (pre-COVID), and estimates of the strength of association between income growth and health spending from international literature. The forecasts also incorporated age-specific population data from United Nations Population Statistics Division, and assumptions about the distribution of per capita health spending by age. A fuller description of the methodology, including projections of the residual factor, is provided in Annex 1. Under a business-as- usual (no intervention) scenario with existing service delivery mechanisms, health spending is projected to rise from 4–5 percent of GDP at present to 6–7 percent of GDP by 2050. In real per capita terms, health spending needs will increase by between 200 percent and 250 percent between 2019 and 2050 (Figure 23). Figure 23: Health spending is projected to rise in both real terms and relative to GDP 1800 8 1600 7 Per Capita Health Expenditure (in FJ$) 1400 6 Share of Health Expenditure Needs 1200 5 1000 4 800 3 in GDP (%) 600 400 2 200 1 0 0 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 Year HE/GDP Ratio Per Capita Health Expenditure Source: Authors’ estimates. 84. Organization; Khan. and Mahmud. 2015. “Is Healthcare a Necessity or Luxury? Empirical Evidence from Public and Private Sector Analyses of Southeast Asian Countries.” Health Economics Review 5(3) and studies cited therein. Fiji Health Sector Review 53 79. Meeting these additional needs will require a combination of increased government spending on health, investment in prevention and health promotion, and increased efficiency in the use of health resources. Unless resources are spent more effectively, public sector health spending would have to increase sharply—almost doubling as a share of GDP— to maintain its existing share of total health spending. Given the significance of NCDs in Fiji, efficiency gains will need to come from strategies that contribute to their prevention, effective management, and treatment. Improved efficiency in hospital spending will be necessary given their large share of health sector resources in Fiji. Investments will also need to focus on mitigating climate-related impacts on the health system. Given Fiji’s physical remoteness, and limited resources to finance care in overseas health facilities, these efforts will need to be placed alongside a policy priority to effectively provide some hospital-based curative care, such as services at the Colonial War Memorial (CWM) hospital and hospitals providing secondary care. Some of this health expenditure could also be supported by rising tax revenues from growing national incomes if Fiji is able to effectively exploit its demographic dividend. In the absence of these measures, however, Fijian households will have to rely on their own resources to meet their health needs, whether supported by voluntary health insurance or OOP spending, or even forgoing care, with adverse consequences for both equity and efficiency. 54 Fiji Health Sector Review Photo Credit: Fiji Ministry of Health and Medical Services Fiji Health Sector Review 55 Health Service Organization and Delivery Summary: • Health services in Fiji are largely publicly funded, managed, and delivered by the Ministry of Health and Medical Services (MHMS). • There is a growing private healthcare provider sector in Fiji and public-private partnership (PPP) arrangements exist in various forms, but little evidence is available about their fiscal implications, or effectiveness in improving priority outcomes. Harnessing private sector potential to better achieve health system goals in Fiji requires the strengthening of government capacity in strategic purchasing, and stronger information systems. • There are service quality gaps, especially in the prevention and management of chronic conditions, related to the availability, composition, and training of human resources for health (HRH), and complementary inputs such as drugs and diagnostics. Public sector infrastructure needs to be upgraded. Digital infrastructure could help to improve access to health services, but is lacking, especially in rural areas. • Hospital bed utilization rates are low, and bed capacity is unevenly used, with smaller hospitals in the Eastern and Northern Divisions characterized by bed occupancy rates (BORs) of 30 percent and below. Moreover, hospitals appear to be providing primary care services that are likely to be more efficiently provided at frontline facilities, such as health centers and nursing posts. 56 Fiji Health Sector Review Governance 80. Fiji’s public healthcare system is governed by the MHMS and organized in four tiers: national, divisional, subdivisional, and medical area. At the national level, the MHMS is organized by department and unit, each responsible for specific aspects of healthcare, public health, and medical services. Healthcare is delivered through four administrative divisions: Central, Eastern, Northern, and Western Divisions, where health officers are responsible for implementing and coordinating health services and public health initiatives to communities. Each division has several subdivisions, and each subdivision oversees several medical areas. Fiji has three divisional referral hospitals, the Colonial War Memorial (CWM), Labasa Hospital, and Lautoka Hospital, that offer specialized tertiary care. Inpatient care is provided by hospitals at the divisional and subdivisional levels. 81. MHMS has gone through several reorganizations involving centralization and decentralization. Prior to 1999, the public sector health services centralized, with a central office and just four corporate divisions (primary and preventative services, health planning, hospital services, and finance).85 Between 1999 and 2003, there were efforts to decentralize to bring services closer to the community. In 2008, these changes were rolled back, and the system was recentralized to a variation of its current functional structure (Figure 24). A second wave of decentralization was launched in March 2009, with a more modest aim of improving access to health services in the Suva subdivision.86 During these decentralization efforts, the workload increasingly shifted from tertiary hospitals to health centers.87 Responsibility for PHC is divided between programs and divisions. 85. Mohammed, North, and Ashton. 2016. ”Decentralization of Health Services in Fiji: A Decision Space Analysis.” International Journal of Health Policy and Management 5(3), 173–181. 86. Government of Fiji. 2011. “Cabinet Memorandum - Report of the Decentralization of the General Outpatient Department (GOPD) Services.” Suva, Fiji: Government of Fiji. 87. Workload is calculated by multiplying work time per staff by the rate of working i.e., clients per hour. Mohammed et al. 2016. op. cit. Fiji Health Sector Review 57 58 Minster Health & Medical Services Figure 24: MHMS Structure Permenent Secretary Fiji Emergecy Excutive Support Unit Health and Mediacal Medical Services Assistance Team Chief Medical Adviser Divisional Medical Head of Medical Head of Head of Head of Superintendents Research, Head of Officers Finance, Asset Head of Human Pharmaceuticals Nursing and Head of Head of Cwm, Lab, LTK, Innovation, Head of Head of Oral Planning Clinical Central, Management Resoucers & Biomedical Health Family Midwifery Data Wellness Health and Policy Support Eastern, Specialist Hospitals and Digital Services Protection Health Analysis and Development Services Northern, Health Management Western Function Function Function Function Function Function Function Function Function Function Function Function Function 1. Public Health 1. Hospital Admin 1. Financial 1. HR 1. Supply 1. Nurses & 1. Epidemiological 1. NCDs 1. Communicable 1. Material 1. Oral Health 1. Strategic, 1. Patient Safety & 2. Provincial Councils 2. Clinical Management Management Chain Midwife Advise Management Disease and Child Promotion Service & Quality 3. Capital Works Management 2. Asset 2. Civil Service Management Management 2. Research 2. Wellness Prevention & Health 2. Oral Health Operational 2. Visiting Medical 4. Clinical 3. Capital Works Management Reforms 2. Biomedical 2. Registration Advise Approach Control 2. Adolescent Standards Planning Teams Management 4. HSS Function 3. Information 3. Learning & Equipment 3. Community Evaluation 3. Food and 2. Communicable Health 3. Audit Dental 2. Policy 3. Overseas 5. HSS Function 5. Board of Visitors Technology Development Management Health Worker & Research Nutrition Disease 3. Sexual Clinics Development Referrals 6. Board of Visitors 6. Rehabilitation 4. Capital Works 4. Work, Life, 3. Medical & Capacity Security Epidemiological, & 4. Training and & Review 4. Blood and 5. Donor Balance Consumables Development 4. Dietetics & Surveillance & Reproductive Awareness 3. National Health Ambulance Engagement Management 3. Research Food Response Health Accounts Services 6. Procurement Governmance Services 3. National 4. External 7. Internal Audits 4. Innovation 5. Health Public Health Partnership 8. Occupational Process Promotion Laboratory (MoUs & MoA) Health & Saftey Management 6. Mental 4. Risk 5. Publications Health Communications 7. Physical 5. Environment Activity Health 8. Training and 6. Health Awareness Emergencies & Fiji Health Sector Review Climate Change Source: MHMS Annual Operational Plan 2023–2024. https://www.health.gov.fj/wp-content/uploads/2023/12/Annual-Operational-Plan-2023-2024.pdf. Clinical Services 7. International Health Network Regulations 82. The private sector is playing an increasingly important role in providing health services, funded by both private and public sources. The private sector includes formal, for-profit providers and services provided by non-governmental organizations (NGOs). There were an estimated 160 private general practitioner (GP) practices in Fiji in 2021, largely concentrated in urban areas. In its Strategic Plan (2020– 2025), the MHMS envisages a broader “collaboration with partners for a more efficient, innovative and higher-quality health system”, including private practitioners. One mechanism by which the Government of Fiji (GoF) has purchased services from the private sector is the Private GP Scheme, under which poorer Fijians have highly subsidized access to a defined set of services, including medicines. In addition, the Free Medicine Program funds private pharmacists to provide free medicines to poor Fijians, and there are also schemes that fund NGOs to provide specific services. Separately, the GoF has partnered with a private sector hospital operator (Aspen Medical) along with the Fiji National Provident Fund to manage two hospitals in Fiji’s Western Division: Lautoka Hospital, operational from 2021, and Ba Hospital, operational from 2022 (Box 2). Box 2: Public-Private Partnerships (PPPs) in Health Private GP Scheme: Building on efforts during COVID-19, Fiji created the publicly financed Private GP Scheme,88 partnering with private GPs, dental practitioners, and medical laboratories to provide services to patients at pre-agreed prices. There are 46 GPs, four dental practitioners, and seven laboratories currently in the scheme. Information on participating providers is available at the Free Health Scheme online portal. The cost of medical services under this scheme is borne by the GoF. As of August 2023, only patients with a combined household income of FJ$30,000 (~US$14,000) or less per annum qualified for the free private services available under this scheme. Free Medicine Program (introduced in 2015): While government hospital pharmacies and dispensaries provide medicines free of charge for all citizens, this program additionally allows eligible individuals to obtain free medicines from selected private pharmacies. Those eligible are (i) adults with an annual income of less than FJ$20,000 (~US$9,400); and (ii) children less than 18 years old whose combined parental income is less than FJ$20,000 (~US$9,400). Although Fiji’s Essential Medicines List—which guides the registration, procurement, distribution, and use of therapeutic products across the health sector—has not been updated since 2015, key informant interviews conducted during a recent World Bank mission89 on pharmaceutical supply chain management suggested that medicines not included in Fiji’s essential medicines list are also available to patients under the Free Medicine Program. The introduction of the Free Medicine Program in 2015 increased annual government spending on drugs and medical equipment from slightly over FJ$10 million (US$4.7 million) in 2015, to over FJ$40 million (US$19 million) in 2018/19.90 The 2024 audit report by the Auditor General of the Republic of Fiji noted anomalies in the free medicine program and the risk of misuse of funds allocated for the program. The report recommended that MHMS create a policy/guideline which will require pharmacies to update the Ministry at regular intervals on the stock of free medicines available at their pharmacy, including identifying medicine stocks that need replenishing; and that MHMS carry out stock-takes of medicines held by private pharmacies at the end of the financial year and disclose in the notes to the financial statements the monetary value of the closing stock of medicines held by private pharmacies.91 88. The Private General Practitioner Scheme (https://freehealthschemesfiji.com/general) was first implemented during the COVID-19 pandemic to reduce overcrowding at public health facilities and limit care to emergent and critical cases. The scheme has been extended with a comprehensive package of routine and emergency services to complement public health service provision. 89. World Bank technical mission to Fiji July 27 – August 9, 2023. 90. Fiji National Health Accounts Report 2014-2019. 91. Office of the Auditor General. 2024. Audit Report on Follow Up of High Risk Ministries / Departments. https://www.parliament.gov.fj/wp-content/ uploads/2024/04/45-Audit-Report-on-Follow-Up-of-High-Risk-Ministries_Departments-.pdf. Fiji Health Sector Review 59 PPP in hospitals: In 2019, the GoF entered a PPP with a private sector hospital operator (Aspen Medical) in partnership with the Fiji National Provident Fund to develop, upgrade, equip, and operate two hospitals—a secondary care level facility, Ba Hospital, and a tertiary care level facility, Lautoka Hospital—in a bid to raise the quality of health services in Fiji’s Western Division. Aspen Medical is expected to provide staff training and medical equipment at both hospitals, and clinical services across a range of specialties. These include the provision of free and subsidized services for eligible populations for a pre-identified service package. Aspen Medical is also contracted to upgrade the infrastructure of the two hospitals and build staff accommodation.92 Financing is based on a two-part tariff, with a base annual payment by the government, and extra payments based on the volume and composition of services provided to Fijian citizens. The PPP arrangement became operational in 2022.93 Grants and transfers to NGOs: The GoF, through the MHMS, also finances NGOs to complement the provision of services by the public and private-for-profit sectors. In 2024, the GoF allocated FJ$8.5 million (US$4 million) to St John’s Association, Frank Hilton Organization’s disability screening program, and other NGOs. Supported by grants, NGOs typically provide screening, preventive, and health promotion services in communities along with referrals to the public health system. A FJ$2.5 million (~US$1 million) capital grant was provided for Kidney Dialysis Treatment Subsidy and FJ$4 million (~US$2 million) to the Sai Prema Foundation, a not-for-profit organization which provides medical, education, and social service initiatives, including the provision of free heart surgery for children. These niche services utilize skilled personnel and specialized equipment, providing an opportunity for capacity building and skills transfer to strengthen health service delivery in the public sector. 83. Although performance assessments of existing PPP arrangements are not yet available, Fiji could benefit from strengthening its monitoring and evaluation framework for such schemes and consider moving towards a more strategic purchaser role. PPP schemes can bridge significant gaps in service delivery, but information on their effectiveness including the volume and quality of services delivered, cost, and population groups using these services is not readily available. It is unclear if existing PPP financing arrangements are linked to quality-of-care indicators, and available evidence suggests that these arrangements restrict the GoF’s role to being a passive purchaser of services. Moving forward, GoF should consider progressing to a strategic purchaser role. This will require not just greater investments in human resources engaged in purchasing private care, but also clearer policies on reporting, data sharing, and data governance to help assess the capacity, quality, and cost of service delivery. 84. Fiji serves as a medical hub for complex or specialized medical conditions for some neighboring countries, and there may be additional opportunities for growth through engagement with the private sector. Other PICs refer patients to neighboring countries, including Fiji, for further diagnosis and treatment. Tuvaluans are supported by the Tuvalu Overseas Medical Referral Scheme, that complements its own currently limited health services. Patients with chronic renal insufficiency who require dialysis are usually referred to Fiji where, due to the lack of dialysis centers in Tuvalu, they end up living in Fiji for the rest of their lives.94 At approximately FJ$82,222 (US$37,000) per dialysis patient per year, Fiji offers a cheaper service than Australia (FJ$127,000, or US$57,000 per year for urban and rural centers, and FJ$182,000, or US$82,000, for remote centers).95 Given the high incidence of diabetes and high blood pressure in Tuvalu and other PICs such as Kiribati, the number of patients requiring dialysis and other specialized care services will continue to increase, a demand that should be considered in the Master Plan for CWM as it may have implications for the broader health system. 92. Aspen Medical https://aspenmedical.com.fj/ (accessed January 1, 2024). 93. Fiji, BOOST data 2022, Ministry of Finance, Strategic Planning, National Development and Statistics. 94. Tuvalu Healthcare Needs Assessment, phase 1, Dec 2023, unpublished draft, conducted in the context of the World Bank financed Health Systems Strengthening Project for Tuvalu (P175170). 95. Gorham et al. 2019. ”Cost of Dialysis Therapies in Rural and Remote Australia – A Micro-costing Analysis.” BMC Nephrology (2019) 20:231. 60 Fiji Health Sector Review Quality of Care: Strengthening Prevention and Management of NCDs in Fiji 85. Quality of care at frontline facilities remains an ongoing concern in Fiji, with gaps evident in reproductive, maternal, neonatal, child, and adolescent health (RMNCAH) services, but crucially also in the prevention and effective management of NCDs. A Primary Health Care Performance Initiative (PHCPI) Assessment undertaken by the World Bank found that indicators of RMNCAH service coverage, such as antenatal care services have declined in recent years, consistent with the findings of stagnant child health outcomes.96 There are also gaps in service coverage of NCDs at frontline facilities. Data from the most recent STEPS survey from Fiji (a new round is planned this year) suggests that fewer than one-third of individuals with hypertension and diabetes are likely to be diagnosed (Figure 25). Among those diagnosed, fewer than one-third of people with diabetes and less than one in seven of those with hypertension are likely to report being treated. With less than 10 percent of people under treatment reporting that their blood sugar or blood pressure is under control, few Fijians with hypertension or diabetes are being effectively managed. Thus, large numbers of Fijians with chronic conditions are at risk of hospitalization for conditions that could have been effectively managed by ambulatory care services (Box 3). Figure 25: Diagnosis, treatment, and control of hypertension and diabetes in Fiji are limited Care Cascades for Hypertension and Diabetes in Fiji 40 35 30 25 Percent 20 15 10 5 0 % diagnosed % treated % controlled Diabetes (Adults) Hypertension (25-64) Source: World Health Organization STEPS Survey 2011, Fiji. 96. World Bank. 2023. Primary Health Care Performance Initiative. Fiji Health Sector Review 61 86. Strategies to increase awareness and early diagnosis of NCDs are desirable but primary care will need to be strengthened to improve chronic disease management. Poor diagnosis of hypertension and diabetes in Fiji reflects a combination of individuals’ lack of awareness of the conditions (presumably reflecting relatively small investments in public health education), and a strategy of essentially opportunistic identification (e.g., when a patient visits the provider).97 Early diagnosis, including through mass screening strategies, can be cost-effective in this context, especially if there is follow-up for the treatment/management of identified conditions in primary care facilities.98 Increased investments in public health education may be effective if backed up by primary care, including support for programs engaging community volunteers for health promotion and screening, and training healthcare providers to support health promotion for lifestyle changes and drug adherence among patients. 87. Continuity and coordination of care in the management of chronic conditions have large benefits but are not being effectively implemented in Fiji. Continuity and coordination involve the relationship between the patient and healthcare providers, the accessibility of medical information over time and across providers, and bringing together health professionals with different areas of expertise (rehabilitation, drug adherence, general practice, nutrition, etc.). Such efforts are known to improve the management and effective treatment of chronic conditions, including patients’ adherence to their drug regimen, but require strong linkages across healthcare providers, and data systems that are suitable for care integration and continuity. Fiji has invested in efforts to enhance NCD care, but recent studies highlight gaps in availability of drugs and diagnostic inputs, inadequate diabetes management in outpatient clinic services even at secondary hospitals, fragmented patient record keeping that is unsuited to effective continuity and integration of care, underutilization of opportunities for care coordination, inadequate adherence to clinical guidelines, and overloaded healthcare workers, translating into poor patient outcomes.99 97. Dearie et al. 2019. “A Qualitative Exploration of Fijian Perceptions of Diabetes: Identifying Opportunities for Prevention and Management.” International Journal of Environmental Research and Public Health 16(7):100; Jonathan et al. 2023. “I think taking herbal medicine first can help prevent. If it doesn’t work then can start taking medication given by doctors: Patients’ Perceptions Towards Hypertension in Fiji.” Plos One 18(8): e0285998; World Bank. 2022. “Assessing PHC Effectiveness: Analyzing Hospitalizations for Ambulatory Care Sensitive Conditions.” Asia and the Pacific Health Financing Forum. September. 98. Kaur et al. (2022). “Cost-effectiveness of Population-based Screening for Diabetes and Hypertension in India: An Economic Modelling Study.” Lancet Public Health 7: e65-e73; Wolcherink et al. 2023. “Health Economic Research Assessing the Value of Early Detection of Cardiovascular Disease: A Systematic Review.” Pharmacoeconomics 41(10):1183-1203 ; Ding et al. 2021. “The Effects of Chronic Disease Management in Primary Health Care: Evidence from Rural China.” Journal of Health Economics 80: 102539. 99. Ibrahim and Lawrence. 2022. “Improving Diabetes Care: A Fijian Diabetes Care Improvement Study.” International Journal of Chronic Diseases 9486679; Kumar and Mohammadnezhad. 2022. “Perceptions of Patients on Factors Affecting Diabetes Self-management Among Type-2 Diabetes Mellitus (T2DM) Patients in Fiji: A Qualitative Study.” Heliyon 8(6): e09728. 62 Fiji Health Sector Review Box 3: Diagnosing and Managing Ambulatory Care Sensitive Conditions (ACSCs) at the Primary Healthcare Level ACSCs are conditions for which timely and effective community or outpatient care can help to reduce the risks of hospitalization by either preventing the onset of an illness or condition, controlling an acute episodic illness or condition, or managing a chronic disease. ACSCs are broadly categorized into (i) vaccine-preventable ACSCs e.g., bacterial pneumonia and measles; (ii) acute ACSCs e.g., pediatric gastroenteritis and urinary tract infections; and (iii) chronic ACSCs e.g., diabetes and hypertension. Hospitalizations for complications related to diabetes and hypertension can often be avoided if these conditions are managed properly at the primary healthcare level. High-income countries widely monitor data on preventable hospitalizations and ACSC hospitalizations as an indicator of access, delivery, quality, and effectiveness of the primary care system. Strengthening primary healthcare to address NCD-related ACSCs at the primary care level can be multi-tiered: (i) targeting risk factors including behavioral/lifestyle modifications, and public health interventions; (ii) identifying risks and early onset of disease using proactive population- based screening and early detection by families, in the community, during routine visits; leading to (iii) early initiation of treatment, and long-term (often lifelong) adherence to treatment to improve the control of symptoms and prevent complications; and (iv) effective management, with ongoing monitoring and adjustments based on co-morbidities and multi-morbidities. Source: World Bank. “Data Driven Decision Making for Universal Health Coverage.” Prince Mahidol Awards Conference side event. Jan 2024. 88. Strengthening Fiji’s health system to diagnose and manage NCDs close to the community will require investments in human resources including expanding the roles of community health workers (CHWs) and improving training. Health workers in frontline facilities appear to have a heavy workload, suggesting that additional responsibilities related to screening and management of chronic NCDs might not be feasible without further support.100 Given the remoteness of many communities in Fiji, interventions to enhance NCD care management could benefit from using the existing cadre of approximately 1,640 CHWs. International studies that assessed interventions where appropriately supervised and equipped CHWs were trained to deliver blood pressure monitoring and counselling for lifestyle behavior change found they improved health and were cost effective.101 The CHW models used in India and Ghana are profiled in Box 4. The provision of integrated care will also require the assessment of the current and future availability and training of allied health workers (e.g., nutritionists, podiatrists, and rehabilitation staff). Separately, additional training might be necessary for SHWs on risk factors, as they appear to have a high level of cultural acceptance of risk factors such as over-weight and obesity. Studies have also identified clinical training gaps at all levels of the health system.102 100. Nawaqaliva. 2022. Exploring Job Satisfaction of Nurses in Fiji. Doctoral dissertation. Auckland: Auckland University of Technology. 101. Gamage et al. 2020. “Effectiveness of a Scalable Group-Based Education and Monitoring Program, Delivered by Health Workers, to Improve Control of Hypertension in Rural India: A Cluster Randomized Controlled Trial.” PLOS Medicine; Krishnan et al. 2019. “Cost-effectiveness and Budget Impact of the Community-based Management of Hypertension in Nepal Study (COBIN): A Retrospective Analysis.” Lancet Global Health 7(10):E1367-74; Jafar et al. 2020. “A Community-based Intervention for Managing Hypertension in Rural South Asia.” The New England Journal of Medicine 382(8):717-26. 102. Ibrahim and Lawrence. 2022. op. cit.; Kool, Webber, and McCool. 2015. ”DR Services in Fiji: Attitudes, Barriers and Screening Practices.” Community Eye Health 28(92): s6-s11; Pickmere. and Booth. 2017. ”Anti-Microbial Stewardship in Fiji.” Faculty of Pharmacy and Pharmaceutical Sciences, Monash University, unpublished. Fiji Health Sector Review 63 Box 4: Examples of Community Health Worker Roles in Service Delivery India: Primary healthcare focus with opportunities for additional health project involvement India’s Accredited Social Health Activists (ASHAs) program was launched in 2005 and with about 1 million registered ASHAs, they represent the world’s largest group of community health workers (CHWs). ASHAs are crucial to India’s primary healthcare system, serving as a link between health staff and rural communities, with each ASHA typically responsible for a population of 1,000 people. Following their selection by local village governments (Panchayats), ASHAs undergo government training for maternal and child health care but are now increasingly becoming more involved in addressing non-communicable diseases given the rising burden of diseases and risk factors. ASHAs are contractual workers and are remunerated based on the same performance accountability mechanisms as if they were fulltime permanent employees. However, ASHAs have flexibility to work on other projects during their free time. Many non-government organizations use this opportunity to involve ASHAs in different projects and compensate them for their time. The additional duties allocation has sparked discussions about their remuneration and workload, with a growing movement among ASHAs advocating for recognition as government staff, w hich w ould include additional benefits and job security. Ghana: A tiered approach Ghana’s Community Health Planning and Services (CHPS) program was established in 1999 and uses two different cohorts of CHWs: full-time paid Community Health Officers (CH Os) and Community Health Volunteers (CHVs). CHOs receive training as a community health nurse with two years of training at an accredited community health nursing training school, followed by two weeks of community engagement training and finally an internship program. CHOs are responsible for a population of approximately 5,000 people in a demarcated CHPS zone and are responsible for maternal and reproductive health services, neonatal and child health services, treatment of minor ailments, health education, and referrals, many at the level of the household. Supervision of duties for CHOs are conducted by public health nurses, physicians assistants and CHPS coordinators. CHVs work part time and are unpaid. CHVs are selected by communities and their main role is to assist CHOs. There are usually at least two CHVs allocated to support one CHO for each CHPS zone. CHVs support the CHO with service delivery tasks, including assisting community members in their homes and community mobilization. Their activities are supervised by the responsible CHO. Source: Ministry of Health and Family Welfare, Government of India. National Health Mission. About Accredited Social Health Activist (ASHA) (2022). Available at: https://nhm.gov.in/index1.php?lang=1&level=1&sublinkid=150&lid=226; CHW Central. 2020. Ghana’s Community Health Officers and Community Health Volunteers. https://chwcentral.org/ghanas-community-health-officers-and-community-health-volunteers/. 89. Fiji can build on existing public funding schemes that involve private sector providers to help address the needs of key population sub-groups (urban informal settlements) or NCD prevention and management as the government strengthens its strategic purchasing function. Although data on private providers by medical area are not readily available, it is unlikely that a substantial private sector presence exists outside the major islands, and even there they are likely to be concentrated in urban areas. However, SHW gaps in urban service delivery in public facilities do need to be filled. Services could be purchased from the private sector to address gaps in health service delivery for poor populations living in urban informal settlements. Currently, 25 clinics in the Western Division, 21 in the Central Division, and four in the Northern Division participate in government PPP schemes for outpatient services, but none in the Eastern Division. Fiji should explore strategic purchasing options (involving the private sector) for services related to health promotion, diagnostics, and chronic disease management. 64 Fiji Health Sector Review Efficiency in Service Delivery: Hospitals in Fiji 90. The high share of hospitals in public spending in Fiji has implications for both allocative efficiency (where resources are being directed) and technical efficiency (how resources are being used). Although more than half of health expenditure is directed to PHC, PHC spending is predominantly allocated to outpatient services in hospital settings in Fiji, rather than to ambulatory or preventative care providers in frontline facilities (e.g., health centers, nursing stations). These would be a less costly alternative and usually more physically accessible to patients. For example, MHMS data on antenatal care (ANC) visits suggest that slightly more than half of ANC visits in hospitals are by pregnant women who are not deemed to be at risk. Data on Fiji’s public hospitals also show that the aggregate bed occupancy rates (BORs) are well below those of OECD countries. Using pre-COVID measures to ensure that comparisons were unaffected by the pandemic, Fiji had a BOR of 59 percent in 2019 (16 percentage points below the OECD average). This figure hides significant cross-hospital differences in BORs (Figure 26, Figure 27). Rates range from below 10 percent in some hospitals to almost 80 percent in the two largest divisional hospitals by bed capacity. Strikingly, there are large variations in BOR within the sub-group of hospitals with smaller capacity (fewer than 100 beds). Figure 26: There were large variations in bed occupancy rates across Fiji’s hospitals in 2019 90.0 80.0 70.0 Bed Occupancy Rate (%) 60.0 50.0 40.0 30.0 20.0 10.0 0.0 0 100 200 300 400 500 600 Number of Hospital Beds Source: MHMS data. 91. Examining occupancy indicators suggests that only about six hospitals in Fiji could be considered well-functioning. The three divisional hospitals fall into the category of better performers with relatively high occupancy rates. Of the other three hospitals in this group, two are in the Central Division (Navua and Wainibokasi) and one in the Western Division (Nadi). All the hospitals in the Eastern and Northern Division, except the divisional Labasa hospital, have low occupancy rates. Unlike other countries with a significant private hospital sector, these findings for the Eastern and Northern Division do not reflect a superior private sector option available to people living there. Instead, the more likely explanation is some combination of input shortfalls (e.g., the lack of adequate numbers of staff, infrastructure, drugs, and consumables), indicating a lack of capacity to deliver services of adequate quality, or excess beds relative to population needs. Fiji Health Sector Review 65 92. The Eastern Division has a greater number of hospital beds per population than the rest of Fiji, yet bed turnover is low. Even when beds in divisional hospitals are included, the ratio of beds to population (per 100,000) is 47–69 percent higher in the Eastern Division than the Central and Western Divisions. Excluding divisional hospital beds leads to an even greater disparity in the number of beds available per person in the Eastern Division. Excess capital investments are therefore likely to be part of the explanation for the observed low BOR in the Eastern Division. Additional analysis of utilization data reveals that occupancy in Fiji’s hospitals increases with the availability of human resources and declines with access to frontline facilities. Unless the goal is to maintain a degree of surge capacity, maintaining the current levels of inpatient infrastructure in the Eastern Division may be inefficient. Figure 27: Hospitals in the Eastern Division had much lower bed occupancy rates than subdivisional hospitals elsewhere in Fiji in 2019 50.0 Bed Occupancy Rates (%) 40.0 30.0 20.0 10.0 0.0 Central Western Northern Eastern Source: MHMS data. 93. Globally, countries have used a variety of methods to balance the issue of low hospital utilization rates with the needs of populations living in remote areas. This includes initiatives that support smaller hospitals via networks linking them to larger hospitals. In high-income countries these also include limiting the provision of acute inpatient care in remotely located facilities and focusing on emergency care (rural emergency hubs), support and referral through telehealth, reliance on air transport, use of frontline PHC facilities to provide emergency care, and initiatives that support smaller hospitals via networks that link them to larger hospitals. Fiji has adopted the last of these strategies, through its existing healthcare deliv- ery structure where divisional hospitals sit atop a collection of subdivisional hospitals with CWM assigned to both the Central and Eastern Divisions, alongside outreach visits. 94. Subdivisional and divisional hospitals account for a high volume of outpatient visits with MHMS data suggesting a lack of primary care options. In per capita terms, there were 0.84 outpatient visits per person to hospitals in the Eastern Division, compared to 0.08 outpatient visits to subdivisional hospitals in the Central Division and 0.43 per person in the Western and Northern Divisions. However, once divi- sional hospitals are included in the mix, the utilization rates of outpatient services in the four regions are comparable, indicating that divisional hospitals in the Central, Western, and Northern Divisions are sup- plementing outpatient services provided by subdivisional hospitals. Our analysis of MHMS data reveals a strong inverse association between a lack of availability of adequate primary care options and use of hospital outpatient services (Figure 28). For instance, 80 percent of all ANC visits in Fiji were in hospitals, almost half- consisting of visits by mothers not deemed at risk. There is some evidence that the share of ANC visits accounted for by hospitals declines with the availability of frontline care options, indicating an opportunity for resource savings by strengthening services provided by health centers, nursing stations, and community-based workers. 66 Fiji Health Sector Review Figure 28: The use of hospitals for antenatal care visits is inversely related to the availability of health workers in primary health care facilities 2500 2000 1500 1000 500 0 0.000 0.500 1.000 1.500 2.000 2.500 3.000 Ratio of ANC visits in Frontline Facilities to ANC visits in Hospitals Source: Authors’ estimates for 2019 based on utilization data for public hospitals for Fiji provided by MHMS. Physical and Digital Infrastructure in Public Facilities 95. The MHMS has four levels of health facilities: divisional referral hospitals, subdivisional hospitals, health centers, and nursing stations. There are 206 facilities in the country. The three divisional referral hospitals offer specialized tertiary care and inpatient care. The 19 subdivisional hospitals also offer inpatient care and significant outpatient care. There are 86 health centers and 101 nursing stations. Table 10: The distribution of health facilities is uneven across the country Central Western Northern Eastern Total Population 385,142 349,095 139,011 39,054 912,302 Number 1 1 1 0 3 Divisional Hospitals Ratio per pop. 1: 424,196 (*) 1: 349,095 1: 139,011 (*) 1: 304,101 Number 6 5 3 5 19 Subdivisional Hospitals Ratio per pop. 1: 64,190 1: 69,819 1: 46,337 1: 7,811 1: 48,016 Number 22 29 20 15 86 Health Centers Ratio per pop. 1: 17,506 1: 12,038 1: 6,951 1: 2,604 1: 10,608 Number 23 26 21 31 101 Nursing Stations Ratio per pop. 1: 16,745 1: 13,427 1: 6,620 1: 1,260 1: 9,033 Number 51 60 44 51 206 Total Facilities Ratio per pop. 1: 7,552 1: 5,818 1: 3,159 1: 766 1: 4,429 *CWM Hospital is the tertiary referral hospital for both the Central and Eastern Divisions. Source: MHMS data. Fiji Health Sector Review 67 96. Efforts to ensure equity in access to care across Fiji’s territory have left some facilities responsible for large populations while others serve very few people (Table 10). Of the four divisions, only the Eastern Division does not have its own tertiary hospital, instead depending on CWM in the Central Division. But the Eastern Division does have many (small) hospitals catering to a relatively small population in the outlying islands. This has implications for efficiency in the use of hospital resources as noted previously. 97. CWM requires major infrastructure upgrades. As the national referral hospital in Fiji and for residents of many Pacific countries, major renovations or a rebuild of the 100-year-old CWM is a key government priority. CWM is also the South Pacific’s largest training hospital for SHWs from Kiribati, Nauru, Palau, Samoa, Solomon Islands, Tonga, Tuvalu, and Vanuatu. 98. Many health centers and nursing stations also need upgrading, but relatively few facilities require major works (extensive upgrades or relocation or being converted from “old” to new”). Sound physical infrastructure is crucial for patient and health providers’ safety and comfort and is a key input into the provision of health services. Being old, the vast majority of health centers need upgraded electricity and plumbing systems, maintenance of physical infrastructure and equipment, and access to power generation systems. Some health centers, such as Nuffield and Raiwaqa, require quite major works. Addressing these gaps will be vital if, as this report recommends, Fiji is to prioritize primary care at its frontline facilities. 99. Enhancements to digital infrastructure (including telehealth facilities) could contribute to improved population access to services. The limited evidence available on telehealth interventions, mostly from high-income countries, suggests that their introduction improves access to health services, without any decline in patient satisfaction. There is also some evidence of savings in costs of travel and time seeking care; and reductions in the time gap between first symptoms and management for patients who need coordinated care, for example involving specialists and primary care providers.103 Gains from digital infrastructure could also arise from easier access to continuing education programs, better and more easily linked patient records, and effective utilization of disease registries. However, the literature also identifies weak digital infrastructure as a major bottleneck in realizing these gains. Only 15 percent of health facilities in Fiji had access to all basic information and communications technology (ICT),104 64 percent had access to the internet, and 42 percent had a laptop, desktop, or tablet. These statistics will have improved recently, as internet coverage was identified as a priority under the World Bank-financed Fiji COVID-19 Emergency Response Project, Fiji’s first International Development Association investment in the health sector. The project has enhanced internet connectivity in 56 health centers nationwide and invested in ICT procurement to support the implementation of Fiji’s Digital Health Strategy 2023–2027. Costa Rica’s reorientation towards integrated primary care, with a focus on underserved areas, offers an example of how uneven distribution of facilities, as in Fiji, could be tackled, including via digital tools (Box 5). 103. Rajit et al. 2021. Review of Policy and Use of Tele-health to Support Health System Strengthening in the New COVID Normal. Report to the Western Pacific Regional Office, WHO, Manila. Melbourne: Nossal Institute for Global Health; WHO. 2019. Recommendations on Digital Interventions for Health Systems Strengthening. Geneva: World Health Organization. 104. The ICT checklist for the survey includes a functioning desktop or laptop computer, tablet, internet reception at the health facility, and connectivity through either mobile phone, Wi-Fi or ethernet; MHMS administrative data 2024. 68 Fiji Health Sector Review Box 5: Lessons from Costa Rica’s Integrated PHC Service Delivery Costa Rica’s successful reorientation of primary healthcare services may offer lessons for similar reforms in Fiji. Following public dissatisfaction with its healthcare system in the 1980s, especially after a major measles outbreak in 1991, there was a phased major reform in the 1990s to change the culture of care delivery. This PHC service delivery shift focused on five key pillars: 1. Integration of public health with primary healthcare: This ensures clear oversight of public health and primary healthcare services to promote efficiency and reduce redundancy of service provision. 2. Multidisciplinary teams integrated within the community: PHC teams, known as an equipo básico de atención integral de salud (or EBAIS), provide comprehensive and coordinated PHC services. EBAIS teams also included CHWs whose role included conducting at least one annual visit per household, to outline stratified risk profiles for communities which then informed a prioritized action plan. Additional support for EBAIS is provided by teams of nutritionists, psychiatrists, and pharmacists within the main PHC organizational unit, called a health area (similar to Fiji’s medical area). 3. Geographic empanelment: All citizens are assigned to an EBAIS team through geographic empanelment, which promotes access and continuity of care. Each team has a target panel size of about 4,000 people. To promote greater equity in access and outcomes, empanelment and the introduction of EBAIS teams began in the country’s most medically under-served rural areas, and then moved to urban areas, including the capital, San José. 4. Measurement and quality improvement at all levels: Quality assurance and improvement are a primary focus for PHC and the health system overall and are supported by robust data feedback mechanisms. EBAIS teams collect comprehensive population data, which are compiled by the health area and sent to the national level. The data are used to assess performance against targets and ensure a high quality of care. 5. Integration of digital technologies at all levels: The country’s electronic health record system facilitates the delivery of comprehensive care to patients. Patient charts function as clinical guides, reminding providers of issues to discuss during their visit, such as hypertension, Type 2 diabetes, and elder care. EBAIS teams use mobile tablets for data collection across urban and rural settings. By 2019, more than 94 percent of the population had been empaneled with an average of one EBAIS team per 4,660 citizens. Following a decade of implementation of these reforms, there has been an 8 percent reduction in infant mortality and a 2 percent reduction in adult mortality. Furthermore, the PHC reforms have notably reduced early deaths, especially among poorer groups. For example, premature mortality decreased overall in Costa Rica between 1980 and 2000 but saw larger declines for the poorest quintile of the population (48 percent reduction) than the richest quintile (39 percent reduction). A 2009 analysis of infant mortality indicated high regional equity: there was no association between geographic region and mortality.105 Costa Rica’s per capita spending on health is below the global average, with households also spending significantly less on health than in neighboring countries. 100. Given the implications of climate change for its health services, Fiji is taking steps to build a resilient health service infrastructure, with improved strategies for energy supply, cooling, heat- ing, and wastewater and waste management as reflected in its infrastructure plan. The govern- ment has also identified digital infrastructure as a bottleneck. In this dynamic environment, the health sector must seek out every opportunity to increase resilience to extreme weather events, in both Fiji and the wider Pacific region, building on a growing evidence base concerning climate-smart health systems and solutions as well as innovation (Box 6). 105. Commonweath Fund. 2021. What Does Community-Oriented Primary Health Care Look Like? Lessons from Costa Rica. https://www.commonwealthfund. org/publications/case-study/2021/mar/community-oriented-primary-care-lessons-costa-rica. Fiji Health Sector Review 69 Box 6: Towards Climate-Smart Health Systems Climate change presents considerable risks, but also opportunities for innovation and efficiencies in Fiji’s health sector. For Fiji, climate-smart health system solutions, critical to ensuring a resilient service delivery continuum, could include the following: • Developing early-warning systems with active surveillance and leveraging community-level care networks for climate-related health promotion and prevention. With urban, coastal, and river floods; cyclones; and extreme heat posing frequent and high-intensity threats to communities, Fiji needs early warning systems to alert communities of foreseen extreme events and risks. Community-level care networks could be engaged to disseminate knowledge about adaptation actions in the face of these events and to build community resilience, especially among vulnerable populations. Strengthened surveillance of climate-sensitive vector-borne diseases is also critical to identify and address their emergence in a timely fashion. • Human resources for health should be equipped with the necessary skills and capacity to address diseases and injury associated with extreme events, as well as the rise in vector-borne diseases and NCDs (e.g., cardiovascular disease related to extreme heat) associated with changing temperature and rainfall patterns, including in urban areas. • Increasing the ability to store and distribute essential medicines by strengthening supply chain management (SCM) and enhancing warehouse capacity, thereby reducing the current practice of splitting up orders of essential medicines into multiple deliveries, which significantly increases costs and carbon emissions, and increases the risk of extreme events disrupting distribution systems. • Enabling digital health across Fiji and, through public or private service arrangements, to other Pacific countries. This would improve access to healthcare advice in remote locations, reduce emissions from healthcare-related travel, and support the effective management of chronic conditions and emergency responses. Digital infrastructure can help to connect facilities, services, information, and people, ensuring quality care delivery to vulnerable communities, bridging gaps in the accessibility and availability of health infrastructure and human resources for health. Tele-enabled digital health could also allow continued service delivery during climate and health emergencies. • Incorporating universal disaster-resilient and energy-efficient solutions into the planning and design of health infrastructure. The spatial planning and design of new health infrastructure and equipment, as well as the retrofitting of existing infrastructure, would benefit from the development of standardized disaster risk management and adaptation principles. Disaster risk management would help limit exposure to natural hazards, and minimize damage in case of extreme events, so facilities can continue to operate during emergencies. Low-cost nature-based solutions such as green cooling should be preferred, promoting well-being and enabling physical activity and better learning. A recent study on green roof development in Southeast Asian countries estimated that implementing green roofs in tropical regions with hot temperatures could reduce energy consumption by up to 50 percent, reducing outdoor surface temperatures by 20°C and indoor temperatures by 14°C.106 Green roofs can also mitigate flooding in countries with high rainfall. 106. Pratama, Sinsiri, and Chapirom. 2023 “Green Roof Development in ASEAN Countries: The Challenges and Perspectives.” Sustainability 2023, 15, 7714. 70 Fiji Health Sector Review Access to Essential Medicines and Consumables 101. Fiji is not providing the diagnostic services its population needs in frontline facilities (health centers and nursing stations). Although they have high ratios of SHWs to patients, and small catchment populations, frontline facilities in remote and rural areas of Fiji are poorly equipped with diagnostic resources. Diagnostic tests play a crucial role at every step of disease management: diagnosis, monitoring, screening, and prognosis, and allow healthcare professionals to initiate timely interventions, saving lives and costs. According to the 2021 Service Availability and Readiness Assessment survey, some diagnostic tests such as for blood glucose (95.7 percent), and urine dipsticks for protein, glucose, and ketones (88 percent), were readily available but meters to test cholesterol levels were only available in 10 percent of Fiji’s healthcare facilities (Figure 29). Rapid diagnostic tests were not available in PHC facilities, requiring patients to visit secondary and tertiary healthcare facilities for some tests. Only 20 percent of health facilities had all the recommended and standard safety precautions or equipment, such as alcohol-based hand rub, water for handwashing, sharps containers, and disinfectant, fully available. Figure 29: Primary and secondary healthcare facilities do not offer the range of diagnostic services needed 67 94 100 100 100 100 100 100 100 93 93 90 80 80 84 85 84 80 70 70 65 60 53 49 50 40 40 35 33 30 20 16 7 10 10 11 7 10 2 6 3 2 0 General Cholestrol Rapid syphillis Rapid HIV Rapid Pregnacy Urine Protein Blood glucose microscopy meter testing testing Haemoglobin test + Glucose + machine Kestone Primary (n=175) Secondary (n=20) Tertiary (n=3) Total (n=198) Source: Service Availability and Readiness Assessment survey 2021. Note: (n) indicates the number of facilities effectively offering that service. 102. Last-mile delivery of essential medicines remains a challenge in primary care. Supply chain management (SCM) challenges related to forecasting and procurement in Fiji have increased the frequency of stock-outs, inadequate access to drugs in government pharmacies, and costly supplementary orders which result in inefficient public sector spending and/or OOP expenditure on medicines. Challenges include outdated policies and standard operating procedures, inadequate forecasting, and emergency procurement which is associated with higher prices and significant shipping costs. Moreover, there is limited access to laboratory capacity to trace and authenticate medicines,107 so Fijians are regularly exposed to counterfeit medicines. These challenges also leave Fiji more vulnerable to stock-outs of essential medicines in the event of a pandemic or extreme weather events. Improving Fiji’s SCM could reduce costs, strengthen resilience during shocks, and improve access to good-quality essential medicines. 107. The Pacific Medicines Testing Program. Fiji Health Sector Review 71 103. The existing storage space for pharmaceuticals, biomedical materials, and consumables and their distribution does not meet current demand. Warehouse capacity is largely centralized at the Fiji Pharmaceutical and Biomedical Services. Storing all essential medicines in a single location presents a security threat. One additional warehouse supported by the World Bank has been completed in Suva and another supported by Japan International Cooperation Agency is underway in Labasa. Once completed, it is estimated these additional warehouses will provide storage capacity for up to six months’ supply of essential pharmaceutical, biomedical materials, and consumables for approximately 80 primary healthcare facilities. The decentralized facility in the Northern Division will allow the MHMS to pre-position supplies for last-mile delivery in the event of health or weather emergencies. 104. Ongoing efforts to improve Fiji’s SCM capacity could be streamlined; in combination with existing efforts to expand storage facilities, improvements in procurement practices will not only reduce stock-outs, but also create efficiency gains. This will also require increasing capacity in Fiji’s Pharmaceutical & Biomedical Services Centre, the Fiji Procurement Office, and associated authorities involved in the forecasting, procurement, and verification of essential medicines. There are opportunities to streamline ongoing capacity building efforts to improve the procurement of specialty medicines and commodities by agencies such as the United Nations Development Programme, United Nation’s Children Fund, and the United Nations Population Fund. These agencies’ considerable experience could also be used to improve forecasts of the need for essential medicines by Fiji’s procurement teams. Ongoing support financed by the European Union and provided by the WHO may also serve to strengthen regulatory capacities, paving the way for dialogue on building regional capacity in the Pacific for laboratory testing and routine control of medicines. A World Bank mission in 2023, conducted in consultation with MHMS, the Fiji Procurement Office, and many development partners108 highlighted opportunities to improve SCM to increase the quality and equity of access to medicines for patients, with possible regional spill- over effects that merit further dialogue with interested PICs, regional partners, and several contributing development partners.109 105. The uptake of pooled procurement has had limited traction in the region. Although other groupings of small island states have produced regional efficiencies through pooled procurements, similar suggestions for the Pacific have been raised and rejected in multiple forums over the years.110 Uptake has been limited to specific areas such as vaccines, reproductive health medicines and commodities, and other specialty drugs, through facilitation by UN partners. 108. Australian Department of Foreign Affairs and Trade (DFAT), the Australian Fiji Program Support Facility, Beyond Essentials Solutions, Cure Kids Fiji, Japan International Cooperation Agency (JICA), New Zealand Ministry of Foreign Affairs and Trade (MFAT), Secretariat of the Pacific Community, UNICEF, United Nations Development Program, United Nations Population Fund, and the World Health Organization (WHO). 109. Aide Memoire, The Health Emergency Preparedness and Response program (HEPR) and the Pacific Health, Nutrition, and Population Programmatic Advisory Services and Analytics (Pacific HNP PASA, P178941), Technical Mission to Fiji: July 27 – August 9, 2023. 110. Discussed at the inaugural Pacific Health Ministers Meeting in Fiji in 1995, it has been raised regularly at subsequent meetings. It was also highlighted as an area for cooperation in the Pacific Plan. 72 Fiji Health Sector Review Photo Credit: Fiji Ministry of Health and Medical Services Fiji Health Sector Review 73 Human Resources Summary: • While MHMS has sufficient allocation of staff positions to meet the World Health Organization’s (WHO) indicative minimum recommended threshold of 4.45 skilled health workers (SHWs) per 1,000 population, the existing high numbers of vacancies mean that Fiji cannot reach the minimum staffing thresholds through the public sector alone. • There was a post-COVID exodus of SHWs from public sector jobs, but many health workers have remained in Fiji, switching to private sector contracts. Considering the health workforce in both public and private sectors, at approximately 4.4 SHWs per 1,000 population, Fiji appears to reach the WHO’s indicative minimum recommended threshold. • Although Fiji’s overall stock of SHWs is close to the WHO minimum threshold, it places the country in the middle of the range observed for its upper middle-income country (UMIC) peers. It will also need to plan for a growing need for allied health workers as its population ages. • Fiji is a medical training hub for the Pacific, especially for Specialist Masters programs, and serves as a medical hub for complex or specialized medical conditions for some neighboring countries. 74 Fiji Health Sector Review 106. Based on registration data, Fiji appears to have an adequate number of doctors, with the number of nurses slightly below WHO recommended thresholds, although these lie roughly in the middle of the range for peer UMICs. Data provided by the Fiji Medical and Dental Secretariat show that 653 government and 443 private sector medical111 licenses were granted in 2023 (Table 11). Using private licenses granted as a proxy for the number of private doctors and adding these to the 860 doctors on MHMS staff, Fiji had 1.4 doctors per 1,000 population (Figure 30). Separately, data from the Fiji Nursing Council show that 2,892 nurses and midwives’ licenses were granted between November 2023 and 31 January 2024. As 2,404 nurses and midwives were working in the public sector (or 87 percent of all licenses), the 488 other registered nurses are assumed to work in the private sector. Overall, Fiji has 3.2 nurses and midwives per 1,000 population, slightly below the WHO-recommended threshold of 3.45 required to achieve Sustainable Development Goal (SDG) outcomes associated with universal health coverage. Relative to WHO thresholds, Fiji does not therefore have a major shortage of SHWs. Fiji‘s SHW numbers lie in the middle of the range for peer UMICs: Jamaica and Tonga have fewer doctors (0.8 and 0.4 per 1,000 population, respectively) and nurses and midwives (2.4 and 2.8 per 1,000 population, respectively), whereas Maldives has almost 3.4 doctors and 4.7 nurses and midwives per 1,000 population.112 Table 11: A significant share of medical registrations in 2023 were for private doctors, but not so for nurses and midwives Medical registrations Public Private Share public Medical interns 286 100 percent GPs 571 339 63 percent Specialists 82 104 44 percent Nurses and midwives113 2,404 488 87 percent Source: Medical registrations issued by the Fiji Medical and Dental Secretariat, Fiji Nursing Council. 111. General practitioners and specialists. 112. Global Burden of Disease 2019 Human Resources for Health Collaborators. 2022. op. cit 113. Licensing data for 2024 do not include disaggregation by public vs private sector. Based on the number of contract switches from public to private practice in the period between 2022 – 2023, it is highly probable that the number and percentage of registered nurses, midwives, and nurse practitioners in the private sector is higher than is estimated in this report. Fiji Health Sector Review 75 Figure 30: Combining the numbers in the public and private sector brings Fiji above the minimum threshold of 1 doctor per 1,000 population 1400 1.6 1.4 1200 1.4 1.2 1000 Number of Registered Doctors 1 Doctors per 1,000 population 800 0.7 0.8 600 0.6 400 0.4 200 0.2 0 0 Public Sector (MHMS) Public (MHMS) and Private Sector Registered doctors Doctors per 1,000 population Source: Fiji Medical and Dental Secretariat (2023) data. 107. There is a shortage of doctors and nurses in government health facilities, as reflected in high rates of unfilled positions. The GoF has 4,337 approved positions for doctors, nurses, and midwives, of which only 3,264 are filled, a vacancy rate of 25 percent overall (30 percent for nurse positions and 4 percent for doctor positions; Figure 31). The number of approved SHW positions has been rising over time (Figure 32) and, if filled, would be adequate not just for the public health facility needs in Fiji, but also to meet the WHO thresholds. The high rate of vacant positions suggests that this may not be possible in the near future. Figure 31: Filling existing positions would meet minimum staffing thresholds for nurses, but additional positions would be needed for doctors Public Sector Staff Analysis 3445 3500 3147 3000 2404 2500 Filled positions 2000 Needed to reach 1 per 1,000 (doc)/ 1500 3.45 per 1,000 (nurses & midwives) 860 912 892 1000 Existing positions 500 0 Doctors Nurses and midwives Source: MHMS administrative employment data (2023) and authors’ calculations. 76 Fiji Health Sector Review Figure 32: The number of allocated positions has risen since 2011 Allocated MHMS Establishment Staff per 1,000 population (2011 and 2023) 8 6.49 6.02 6 4.86 3.75 3.41 3.6 4 3.06 2.46 2 0 Central Western Northern Eastern 2011 2023 Source: 2023 data: MHMS data. 2011 data: Wiseman et al. 2017. “Measuring Inequalities in the Distribution of the Fiji Health Workforce.” International Journal for Equity in Health 16(1):115. 108. In 2022, Fiji experienced an exodus of SHWs, especially nurses, from the public sector, with most staying in Fiji and likely joining the private sector. An estimated 1,655 nurses and 334 doctors left MHMS between 2018 and 2023 (not accounting for rehires), with the greatest numbers leaving in 2022 and 2023. Nine of 10 SHWs who left MHMS resigned.114 Most of those leaving stayed in Fiji and are likely to have joined the private sector (Figure 33). This coincides in part with the Lautoka and Ba PPPs becoming operational which meant previously public sector SHWs would have switched to private sector contracts. Although the exact numbers are not available, anecdotal evidence suggests that, not only did most of the previously publicly managed SHWs stay in post, but that previously vacant positions in these hospitals were rapidly filled by staff from CWM and Labasa Hospitals. This shift from the public to the private health sector has been going on for some time. In 2008 an estimated 25 percent of SHWs were employed in private practice,115 compared to around 30 percent in 2023 based on our estimates.116 114. As opposed to leaving because of compulsory retirement, termination or dismissal, death, or retirement on medical grounds. 115. Negin, Roberts, and Lingam. 2010. ”The Evolution of Primary Health Care in Fiji: Past, Present and Future.” Pacific Health Dialog 16(2) 13-23. 116. Using 2023 MHMS employment administrative data for the public sector; private doctor numbers are estimated by using the number of private medical licenses issued by the Fiji Medical and Dental Secretariat in 2023. Fiji Health Sector Review 77 Figure 33: Large numbers of doctors and nurses have left the public sector in recent years for other jobs Total number of exits Reasons for exiting MHMS 900 100% 821 19% 800 90% 40% 25% 700 80% 3% 9% 2% 600 70% 9% 62% 60% 500 4% 50% 400 47% 340 67% 17% 40% 300 1% 30% 6% 172 200 13% 75 79 90 20% 63 43% 100 11 21 26 54 10% 0 3% 5% 17% 17% 0% 2019 2020 2021 2022 2023 2018 2019 2020 2021 2022 2023 2022 2023 2022 2023 Doctors Nurses Doctors Nurses No Reason Given Pursue Further Studies Personal Other Local Employment Migration/ Other Employment Opportunity Abroad Source: MHMS administrative employment data (2023). 109. Exit data collected by the MHMS suggest that most doctors and nurses who leave Fiji go to New Zealand (NZ), followed by Australia. It is unclear how many of those practice medicine or nursing once they are in NZ or Australia, as anecdotal evidence suggests that a fair share transition into the aged-care industry. Data shared by the Nursing and Medical Councils of New Zealand show a steady rate of Fijian- trained doctors being added to the NZ register annually since 2014, and a significant increase in Fijian- trained nurses in the aftermath of COVID-19 (Figure 34). This upward trend is something to monitor, but numbers are lower than expected given recent media coverage, at least compared to those moving to the private sector within Fiji. Figure 34: An increasing number of nurses left for New Zealand post COVID, while the numbers for doctors have remained constant Fijian-trained nurses added Fijian-trained register toadded nurses to register Fijian-trained Fijian-trained doctors added to register doctors added to register 2023 79 2023 10 2022 32 2022 7 2021 2021 6 16 2020 7 2020 10 2019 8 2019 4 2018 6 2018 7 2017 6 2017 6 2016 9 2016 7 2015 10 2015 3 2014 5 2014 12 2013 2013 13 0 5 10 15 0 20 40 60 80 100 Source: Nursing and Medical Councils of New Zealand data. 78 Fiji Health Sector Review 110. While more experienced health workers are leaving the public sector for private practice, a relatively small but growing cohort of younger staff may be more inclined to go abroad. In 2023, 90 percent of doctors and 74 percent of nurses working in the public sector were below the age of 40, suggesting the public sector is losing seasoned staff. This is confirmed by the large share of specialists registered in the private sector (Table 11). In contrast, the data shared by the Nursing and Medical Councils of New Zealand suggest that relatively young professionals are being attracted by schemes such as the Te Whatu Ora relocation grants to encourage overseas doctors to work in rural primary care. Indeed, 86 percent of Fijian-trained nurses who registered in NZ between 2013 and 2023 were under 35 years old, as were 66 percent of Fijian-trained doctors (96 percent were under 40). 111. Fiji has used salary incentives to improve staff retention in government service, which could be further complemented by non-monetary improvements. In 2017, the GoF adopted a new salary framework for medical doctors employed in the public sector which translated to significant salary increases, ranging from 50 to 80 percent. In 2024 the GoF introduced salary adjustments and the re- introduction of a consolidated allowance to address long-overdue raises for nurses. The salary increases were aimed at addressing the overall attrition of the nursing workforce, while the allowances were intended to attract staff to rural or remote areas.117 It is too early to assess these schemes’ effect on improving retention, but anecdotal evidence during the drafting of this report suggests departures are slowing down. Remuneration may not be the only threat to staff retention; both monetary and non-monetary drivers need to be considered. A qualitative study on nursing job satisfaction in Fiji highlighted unclear promotion pathways as one of the reasons given for leaving MHMS positions. Other reasons included a lack of recognition and devaluation—a feeling of not being heard, unsafe working conditions, stress, burnout and stigma, and complex roles and responsibilities.118 This aligns with trends elsewhere. For example, American SHWs are leaving their jobs in a post-COVID exodus due to chaotic clinical environments exacerbated by COVID-related work conditions, work overload, and perceived lack of organizational support; this has all contributed to “the Great Resignation”.119 Further studies into the motives of recent graduates and more experienced professionals in leaving the public sector would be helpful to strengthen healthcare workforce retention strategies and workforce planning. 112. Women constitute a majority of registered doctors and medical graduates in Fiji. In 2023, 60 percent of registered doctors (both public and private) were women, up from 33 percent in 2007 (it is unclear whether this included private practitioners) (Figure 35). According to data from the Fiji National Univeristy (FNU), 6 out of 10 students who graduated from the Bachelor of Medicine and Surgery between 2015 and 2022 were female (as were half of the Specialist Masters graduates), and 66 percent of interns registered with the Fiji Medical and Dental Secretariat are female. Similarly, 7 out of 10 graduates from the University of Fiji’s health programs in the period between 2018 and 2023, were female. 117. Fiji One News.2023. “Salary for Nurses to Increase.” https://fijionenews.com.fj/salary-for-registered-nurses-will-increase/ 118. Nawaqaliva. 2022. “Exploring Job Satisfaction of Nursing in Fiji.” Auckland University of Technology. 119. Abbasi. 2022. ”Pushed to Their Limits, 1 in 5 Physicians Intends to Leave Practice.” JAMA 327(15):1435-1437. Fiji Health Sector Review 79 Figure 35: Women now make up most of the current and future workforce of doctors Share of female medical Share practitioners, of female medical2007-2003 practitioners, 2007-2003 Intern registration (2023) 66 Specialist registrations (2023) 41 GP registrations (2023) 56 Doctors (2007) 33 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Female Male Source: Fiji Medical and Dental Secretariat data (2023) and WHO Global Health Observatory (2007). Figure 36: Fiji has seen an increase in public sector doctors, nurses, and midwives per population, but stagnating numbers of dentists and pharmacists Medical staff per 10,000 population 50 38 38 40 29 30 26 22 19 19 20 8 9 10 10 4 4 3 2 0 1 1 1 1 1 1 0 Medical doctors Nurses and midwives Dentists Pharmacists 1999 2003 2009 2015 2019 2023 (filled MHMS positions) 2023 (MHMS establishment) Source: Global Health Workforce Statistics (GHWS) Database and MHMS administrative employment data (2023). Note: The GHWS Database aims to collect data on both public and private SHWs. In the case of Fiji, the data are understood to be for the public sector only and include filled positions (rather than approved positions). Based on the MHMS administrative employment data shared by MHMS, there are even fewer filled dentist and pharmacist positions than reflected in the GHWS database. The drop in nursing numbers in 2023 reflects the conversion of contracts from public to private sector as part of the GoF Health PPP. 80 Fiji Health Sector Review 113. Fiji is a training hub for SHWs, with medical and nursing degrees offered at the FNU and the University of Fiji.120 Students with Fijian citizenship made up the bulk of the undergraduate programs for FNU between 2015 and 2022121 with 404 Fijians graduating with Bachelor of Medicine and Bachelor of Surgery (80 percent of the graduates in that period), while 1,810 who graduated with a Bachelor of Nursing were Fijian (90 percent) as were 136 of those receiving their Specialist Masters (65 percent of graduates (Figure 38).122 Fiji also serves as a regional training hub for the PICs. Regional students, predominantly from Kiribati, FSM, Nauru, Palau, Samoa, Solomon Islands, Tonga, Tuvalu, and Vanuatu made up 20 percent of graduates from medicine programs, 10 percent of those from nursing programs and 36 percent of those from masters programs in the period between 2015 and 2022.123 Aside from FNU, two privately operated institutes also provide training for SHWs. The Sangam College for Nursing provides undergraduate nursing training while the University of Fiji has produced medical doctors for over a decade and has more recently also included an undergraduate nursing program from which the first cohort of approximately 70 students graduated in 2023. 114. Despite growing need, the numbers of students completing dietetics and nutrition as well as physiotherapy and rehabilitation training programs have remained relatively unchanged over the years. To address the rising treatment and management needs of people with NCDs, Fiji will need not just growing numbers of doctors and nurses in the future, but also staff with skills in rehabilitation, nutrition, podiatry and so on. However, the data in Figure 37 show the number of students specializing in dietetics and physiotherapy is not rising over time. How these allied health workers are distributed geographically, or between the public and private sectors is not known, although evidence from other countries suggests that they tend to prefer urban settings and assignments in hospitals. Figure 37: The number of students specializing in nutrition and rehabilitation is not keeping pace with growing needs Dietetics and nutrition programs Physiotherapy and rehab programs 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 0 0 2015 2016 2017 2018 2019 2020 2021 2022 2015 2016 2017 2018 2019 2020 2021 2022 Fijians Non Fijians Fijians Non Fijians Source: Fiji National University. 120. Fiji has two medical schools (Fiji National University and University of Fiji) and in addition to the Sangam College of Nursing and Health Care Education, three nursing schools. There were no data available from the Sangam College on Nursing and Health Care Education 121. Nationality data were not available for 2017, and are therefore excluded from this analysis. A total of 64 students graduated with a Bachelor of Medicine or Bachelor of Surgery, 23 from a Specialty Master, and 135 with a Bachelor of Nursing. 122. Specialist Masters include Medicine in Anesthesia, Surgery, Emergency Medicine, Family Medicine, Internal Medicine, Obstetrics and Gynecology, Pathology, Pediatrics, Ophthalmology, Master of Applied Epidemiology, Master of Public Health/ Research, and Master of NCD. 123. Nationality data were missing from the 2017 data shared by FNU. Fiji Health Sector Review 81 Figure 38: Fiji National University and the University of Fiji are producing a steady flow of skilled health workers124 for the Fijian and the broader Pacific workforce 180 100% Share of Non Fijians Graduating 160 90% 140 80% Number of Graduates 70% 120 48% 60% 100 38% 50% 80 37% 34% 30% 40% 60 19% 20% 30% 17% 15% 15% 40 20% 20 10% 0 0% 2018 2019 2020 2021 2022 2018 2019 2020 2021 2022 Bachelor of Medicine and Bachelor of Specialist Masters Surgery Fijians Non Fijians Share of Non Fijians 500 100% 450 90% 400 80% 350 70% 300 60% 250 50% 200 32% 40% 150 30% 100 12% 12% 20% 9% 50 4% 10% 0 0% 2018 2019 2020 2021 2022 Bachelor of Nursing Fijians Non Fijians Share of Non Fijians Source: Fiji National University. Note: Breakdown by nationality was not available for 2017. Totals were 64 graduates with Bachelor of Medicine and Bachelor of Surgery, 23 graduates with Medical Specialist Masters, and 134 graduates with Bachelor of Nursing. 115. Fiji seems able to produce enough SHWs to account for population growth and ‘natural’ attrition.125 Given the projected increase in population and natural attrition of staff, Fiji will need to gradually increase its pool of SHWs. Fiji’s population is projected to increase to 1,066,000 by 2050.126 Assuming successful recruitments to fill current vacancies, the allocation for nurses and midwives’ positions in the MHMS can cater for the projected population growth up until 2032. By 2050, 1,066 doctors and 3,678 nurses and midwives will be needed, up from 912 and 3,147 respectively in 2023 (Figure 39).127 In addition, Fiji would also need to replace the SHWs lost through natural attrition (compulsory retirement, termination or dismissal, death, or retirement on medical grounds) and ongoing migration overseas. Strategies to increase the number of allied staff (nutrition experts, physiotherapists, etc.) and ensure they are equitably distributed across regions and facilities will, however, be needed, including investments in additional training capacity. 124. The analysis excludes data from the TISI School of Nursing and Healthcare Education. 125. Compulsory retirement, termination or dismissal, death, or retirement on medical grounds. 126. Based on 2023 population data shared by MHMS and UN population projections. 127. FNU and University of Fiji collectively train an average of 113 doctors and 288 nurses per year (Fiji nationals). 82 Fiji Health Sector Review Figure 39: Due to population growth, Fiji will need to increase the number of doctors, nurses, and midwives to reach and maintain minimum Sustainable Development Goal thresholds to 2050 4,000 3,427 3,528 3,611 3,678 3,500 3,147 3,315 3,000 2,500 2,000 1,500 912 961 993 1,022 1,047 1,066 1,000 500 - 2023 2030 2035 2040 2045 2050 Number of doctors needed for 1 per 1,000 population Number of nurses and midwives needed for 3.45 per 1,000 population Source: Authors’ calculations based on 2023 population data shared by MHMS and UN population projections. 116. The uneven distribution of SHWs remains an ongoing concern, and Fiji will need to carefully balance the need to cater to growing urban populations, many of whom live in poorly served informal settlements, with the requirements of populations located in remote and rural areas (Figure 40). To improve the distribution of human resources of health across Fiji, MHMS will need to consider the number and geographical distribution of SHWs in both the public and the private sectors. Many urban areas have SHW to population ratios that are relatively low compared to standard thresholds, even when considering only approved positions without accounting for vacancy rates, with most of these positions concentrated in hospitals. A careful consideration of the public-private mix of providers, across facilities and across regions, will be needed to ensure equitable access to services. 117. Health centers and nursing stations that cater to the urban poor lack the staff needed to deliver effective preventive and follow-up care. One notable example is the Suva-Nausori corridor, where the largest concentration of informal settlements can be found.128 Alongside a rising burden of NCDs, inadequate water and drainage in such settlements contribute to diarrheal diseases and infectious disease outbreaks, especially in children.129 Suva’s medical area of Raiwaqa has a catchment population of over 32,500 and 1.05 SHWs per 1,000 people, while Nuffield has a catchment population of over 56,000 and 0.67 SHWs per 1,000 people. At the time of drafting, the Nuffield Health Center had a single public health nurse and three community health workers to cover its entire catchment population. Similarly, Valelevu has a catchment population of over 60,000 and 0.70 SHWs per 1,000 people. They have recently had to halt preventative services due to lack of human resources. 128. Government of Fiji. 2011. The National Housing Policy. Suva, Fiji: Government of Fiji. https://www.housing.gov.fj/housing-policy-development. 129. Barriers for Health Promotion & Disease Prevention in Fiji, 2020. Fiji Health Sector Review 83 Figure 40: Allocated public health positions are unevenly distributed by subdivision and facility level Allocated MHMS staff per subdivision, per 1,000 population. 2023 establishment 14.00 12.02 12.00 9.58 10.00 8.70 8.05 8.05 7.97 8.00 6.75 7.15 6.00 3.53 3.99 3.60 4.00 2.51 2.20 2.09 2.62 2.49 2.97 2.94 2.77 1.51 1.53 2.00 0.00 Rewa Suva* Nadi Lautoka/Yasawa* Tavua Lakeba Nadroga/Navosa Bua Cakaudrove Kadavu Lomaloma Rotuma Tailevu Naitasiri Serua/Namosi Ba Rakiraki Macuata* Taveuni Lomaiviti* CWM including Eastern… Central Western Northern Eastern Allocated MHMS staff per level of facility, per 1,000 people 9.0 8.0 8.0 7.1 7.0 7.1 7.0 6.0 5.5 5.4 4.9 5.3 5.1 5.0 4.0 2.7 2.8 2.9 3.0 2.2 2.0 1.9 1.7 1.6 1.7 1.6 1.4 1.4 1.7 1.4 1.6 0.9 0.9 1.3 1.1 1.1 1.0 1.1 1.2 1.0 1.3 1.5 1.3 1.3 1.0 0.6 0.7 0.8 0.4 0.0 Suva* Tavua Lakeba Lautoka/Yasawa* Nadroga/Navosa Bua Cakaudrove Kadavu Lomaloma Rotuma Naitasiri Tailevu Ba Rakiraki Macuata* Serua/Namosi Taveuni Lomaiviti* Rewa Nadi Central Western Northern Eastern Health center and nursing station staff per 1,000 Hospital staff per 1,000 Source: MHMS administrative data. Note: The population of the Eastern Division has been included when calculating the SHW ratio for the CWM Hospital since it also caters for that division. (*) Unallocated health division staff have been allocated to these subdivisions as the ‘capital’ subdivision of each division. 84 Fiji Health Sector Review 118. CHWs could play a critical role in addressing the health needs of both rural and rapidly growing urban populations, including in formal and informal settlements, and lower the travel and time costs of seeking care in these population groups. The models in rural and urban settings will probably need to differ, so models should be piloted in the medical areas that also pilot PHC upgrades and evaluated. There is a large and growing literature on the contribution of CHWs to addressing noncommunicable diseases and in promoting health more generally.130 In Fiji’s case, where nurses are either moving to the private sector or migrating, CHWs could contribute to addressing NCDs through early identification, prevention, and effective management, including acting as a bridge to healthcare facilities (Annex 8). 119. CHWs have been part of Fiji’s service delivery model since the 1970s but will need additional training and supportive working conditions if their role is to expand. Although their numbers have been declining over time, there are currently 1,640 CHWs131 involved in promoting wellness and addressing public health concerns, including data gathering. The MHMS has provided CHW training since the 1980s,132 but it is unclear how effective these training programs have been, or whether CHWs have been able to apply what they have learned. Currently, CHWs are selected by their communities and expected to receive at least six weeks of standardized training to equip them for their role, covering topics such as maternal and child health, infectious diseases and NCDs, and emergency first aid. As of 2023, only 916 (or 56 percent) of active CHWs had been trained. CHWs are not formally accredited, although there is a growing consensus that establishing accreditation or certification for CHWs could better enable them to function within their defined scope of practice. CHWs are remunerated based on their submission of reports, which results in delays to payments to them, but if their role is expanded to address screening and health promotion for NCDs, they may need additional monetary or non-monetary compensation. Clearly defined, and feasible, roles, as well as opportunities for personal and professional growth are likely to be key to CHW effectiveness. The literature points to strong supervisory support and training opportunities as crucial to CHW engagement.133 120. A more recent introduction into Fiji’s service delivery model are the nursing aides and nursing assistant positions. These were created in 2022 to cover non-clinical work and to increase the amount of time nurses could spend on clinical duties. The Fiji Nursing Association has expressed concern that the employment of nursing aides and assistants could jeopardize patient safety as they are not yet a regulated part of the workforce. Nevertheless, in a situation of HRH shortages, freeing up SHWs from non-clinical work and increasing the time they can spend on skilled care is an important strategy. 130. Jafar et al. 2020. “A Community-based Intervention for Managing Hypertension in Rural South Asia.” The New England Journal of Medicine 382(8):717- 26. 131. Presentation from the MHMS Chief Nursing and Midwifery Officer, 2023. 132. Yoon et al. 2019. “Working Conditions and Perceptions on Incentives of Fiji Community Health Workers:Findings from a Survey with a Convenience Sample in 2016.” Journal of Global Health Science 1(1) : e8. 133. Kok et al. 2015. ”Which Intervention Design Factors Influence Performance of Community Health Workers in Low- and Middle-Income Countries? A Systematic Review.” Health Policy and Planning 30:1207-1227; Colvin, Hodgins, and Perry. 2021. ”Community Health Workers at the Dawn of a New Era: Incentives and Remuneration.” Health Policy Research and Systems 9(S3):106; Mishra et al. 2015. ”Mitigation of Noncommunicable Diseases in Developing Countries with Community Health Workers.” Globalization and Health 11:43. Fiji Health Sector Review 85 86 Credit: Fiji Ministry of Health and Medical Services Photo Fiji Health Sector Review Going Beyond the Health Sector to Improve Population Health Summary: • Fiji needs to implement effective strategies that focus on lifestyle risk factors to delay the age at which people contract NCDs and to reduce the development of multi-morbidity. These require actions that go beyond what the health sector can do by itself, including partnering with the ministries of finance, education, agriculture and waterways, and others, and offer high returns to investment. • Fiji has used taxation to make tobacco products less affordable, but it could do more, as the share of cigarette retail prices accounted for by excise taxes in Fiji is only around half of the shares observed in Australia and the Cook Islands. • Fiji could build on its existing excise tax regime for sugar-sweetened beverages (SSBs), foods with high sugar-content, and unhealthy snacks to include foods containing trans-fats, and tax additional categories of sugar-rich drinks. Value-added tax (VAT) could also be used to target SSBs and unhealthy foods. • Fiji could consider subsidizing healthier options such as fruits, vegetables, and unsweetened water, possibly combined with a strategy of using tax revenues from SSBs and tobacco to make healthier food options more attractive to the consumer. • The revision to the Healthy School Policy by the Ministry of Health and Medical Services (MHMS) and the Ministry of Education, targeting physical activity and healthy eating habits among school children, is welcome. However, effective implementation is key. Fiji Health Sector Review 87 Fiji Health Sector Review 87 121. Fiji needs to implement effective strategies that focus on lifestyle risk factors. The high prevalence of lifestyle-related risk factors in the Fijian population and their significance to the NCD disease burden suggests that policy makers should take measures to reduce lifestyle-related risks that could help delay the age at first onset and slow the transition to multi-morbidity among people with NCDs, thereby reducing the need for healthcare. Although the screening, treatment, and effective management of NCDs—all actions that are in the direct domain of the health sector—are important, additional actions are required that go beyond the health sector. These will likely require engagement with the finance, education, and agriculture and waterways ministries, and other agencies. 122. Fiji has introduced many of the “best buy” population health interventions, but unexploited opportunities remain (Annex 9). The World Health Organization has identified “best buy” (or highly cost- effective) regulatory, tax, and information-based population health interventions focused on promoting healthy diets and physical activity, and lowering alcohol and tobacco use, that have high returns on investment.134 A recent WHO report on tobacco confirms that cigarettes were less affordable in Fiji in 2022 than they were a decade earlier but that tobacco taxes only accounted for 38 percent of the retail price of cigarettes—about the same as in Tuvalu and Vietnam, but considerably lower than Australia (77 percent) and Cook Islands (73 percent).135 Measures to promote healthier diets are also lagging despite these now being the greatest risk factors for NCDs in Fiji. While there have been tax increases on unhealthy foods and drinks, more needs to be done. 123. Eliminating trans-fats in the food chain offers a particularly attractive opportunity for policy action. Trans-fatty acids are created when liquid vegetable oils are partially hydrogenated to make them solid at room temperature. Partially hydrogenated vegetable oils are commonly found in baked and fried foods, snack food products, and cooking oils and spreads (e.g., ghee and margarine). Although removing trans- fats from the production process is straightforward and does not affect product taste, their consumption substantially raises the risk of heart disease, by raising LDL (“bad”) and lowering HDL (“good”) cholesterol levels. Globally (and regionally) many countries (Figure 41) are implementing policies to either ban, or limit, trans-fat content. An early adopter, Denmark, experienced declines in deaths from heart disease, and in adolescent and child obesity following a ban on trans-fats in 2001; by 2016 it was experiencing 30 fewer deaths from heart disease per 100,000 population.136 134. Bloom et al. 2014. “The Macroeconomic Impact of Non-communicable Diseases in China and India: Estimates, Projections, and Comparisons.” Journal of the Economics of Ageing 4(C):100-111; World Health Organization. 2021. Saving Lives, Spending Less: The Case for Investing in Noncommunicable Diseases. Geneva: WHO. 135. World Health Organization. 2023. WHO Report on the Global Tobacco Epidemic, 2023. Geneva: WHO. 136. Spruk and Kovač. 2020. “Does a Ban on Trans Fats Improve Public Health: Synthetic Control Evidence from Denmark.”, Swiss Journal of Economics and Statistics 156(1) 1–32. https://doi.org/10.1186/s41937-019-0048-0. 88 Fiji Health Sector Review Figure 41: Regional policies on eliminating trans-fats Source: WHO TFA Country Score Card; WHO Global Health Observatory; WHO NCD Document Repository; WHO Global Database on the Implementation of Nutrition Action (GINA). 124. Fiji taxes foods that are rich in sugar, SSBs, and unhealthy snacks but the tax system could do more to promote consumption of healthier foods and drinks. Locally produced non-alcoholic beverages containing added sugar or artificial sweeteners, excluding milk-based drinks, are taxed, as are imported SSBs, which are subject to a 15 percent import excise duty. These taxes were extended to four unhealthy food product categories in the 2023–24 budget: ice cream and other edible ice; sweet biscuits, waffles, and wafers; sugar confectioneries (excluding traditional Indian sweets); and snack foods (“obtained by roasting, frying, baking, swelling and the like”).137 However, taxes should also be extended to unsweetened (100 percent) juices and sweetened milk-based drinks, which have similar health risks to SSBs. Excluding unsweetened bottled waters from import excise duty would further support healthier substitution. Fiji could also consider moving to a single excise tax for local and imported SSBs and unhealthy food products, with tiered rates based on nutrient content (see Table 12 for examples of other tax designs within the region). Once established and its effectiveness demonstrated, the tax could be extended to more unhealthy food categories. 137. Excise (Budget Amendment) Act 2023. https://www.frcs.org.fj/wp-content/uploads/2023/07/Act-9-Excise-Budget-Amendment.pdf. Fiji Health Sector Review 89 Table 12: SSB taxes in Fiji and selected neighboring countries Country Structure Exemptions Rates Volume-specific (domestic) Milk-based drinks, 100 per- Domestic: FJ$0.40 Fiji and ad valorem (import) cent juices, unsweetened (US$0.18)/L or kg. Import excise water (domestic only) excise: 15 percent Juices (whether sweetened Samoa Volume-specific excise WS$52.5 (US$19.5) per L or not), unsweetened water Volume-specific excise, Juices (whether sweetened TOP 0.05 (US$0.02)/L up to Tonga tiered based on sugar or not), concentrates, TOP 4 (US$1.7)/L depending content unsweetened water on sugar content Domestic: VUV 50 Volume-specific (domestic Milk-based drinks, juices (US$0.43)/L Imports: 15–75 Vanuatu excise) and ad valorem (domestic only), unsweet- percent depending on bev- (import tax) ened water erage type Source: WHO. 2023. Global Report on the Use of Sugar-Sweetened Beverage Taxes. Geneva: World Health Organization. https://www.who. int/publications/i/item/9789240084995. 125. The VAT system could also be used to incentivize healthier diets. Fiji raised its standard VAT rate from 9 percent to 15 percent in the 2023–24 budget.138 A list of 21 basic consumer goods, including onions and potatoes, but also salt and sugar, remain zero-rated, however. Salt and sugar should be subject to standard VAT, while a limited group of healthy foods—including vegetables, fruits, nuts, and seeds—could be added to the zero-rated list. Fiji could also consider using a portion of the revenue from its SSB and unhealthy food taxes to subsidize healthier alternatives such as fruit and vegetables. Revenue from these taxes is not currently earmarked for any specific purposes. 126. Enforcement, monitoring, and evaluation mechanisms will be critical. Tax rates also need to be regularly reviewed and adjusted to effectively disincentivize consumption. As an illustration, rising incomes can mitigate the budgetary impact of a tax increase. Inflation would erode the impact of per unit taxes on consumer behavior. For this reason, taxes on alcohol, tobacco, and food items need regular updating. Tax impacts on price rises can also be undermined by smuggling and/or tax evasion. Ongoing assessments of the impacts of taxes on prices of alcohol, tobacco, and unhealthy imports, and investment in their enforcement will be necessary. 127. Another promising initiative is the revision of the Healthy School Policy (HSP) by MHMS and the Ministry of Education. This policy targets the high NCD burden among younger Fijians, reaching them at a time when children’s health habits and lifestyles are being formed. Reviews of available evidence suggest that school health interventions aimed at improved diets and physical activity, do in fact increase the consumption of healthy foods and beverages and physical activity.139 This government initiative will introduce health-promoting content into the school curriculum (e.g., physical activity, healthy eating behaviors, and age-appropriate information on healthy choices). It will also provide children from the ages of 3 to 19 with routine, systematic screening in early childhood education, and primary and secondary schools nationwide. This initiative is also intended to promote healthy behavior and enroll students who are malnourished140 for follow-up, treatment, and management according to protocol. The HSP promotes school gardens that make available nutrient-rich vegetables in the school setting, which can serve as a tool to tackle Fiji’s triple burden of malnutrition among Fiji’s children and youth. A Healthy Canteen Policy that will eliminate trans-fats in schools and improve access to healthy food in schools is expected to be approved by Cabinet by May 2025. 138. VAT (Budget Announcement) Bill No. 6 of 2023. 139. Durao et al. 2023. “Effects of Policies or Interventions that Influence the School Food Environment on Children’s Health and Non-health Outcomes: A Systematic Review.” Nutrition Reviews 82(3) :332-60; Robertson-Wilson et al. 2012. “Physical Activity Policies and Legislation in Schools: A Systematic Review.” American Journal of Preventive Medicine 43(6) :643-9; Woods et al. 2021. “The Evidence for the Impact of Policy on Physical Activity Outcomes Within the School Setting: A Systematic Review.” Journal of Sports and Health Sciences 10(3):263-76. 140. Malnutrition, in all its forms, includes undernutrition (wasting, stunting, underweight), inadequate vitamins or minerals, overweight, obesity, and resulting diet-related non-communicable diseases. 90 Fiji Health Sector Review Fiji Health Sector Review 91 Policy Suggestions and Programs 128. In this report, we propose a new vision of healthcare, centered around the expression Mo Bulabula ka Bula Balavu: a Fijian greeting meaning “Wishing you a healthy life, and a long life”. The report provides estimates of the financing required for the country to develop a health system that is fit for the 21st century, meets its population’s expectations, and enhance the economic gains from a healthier working-age population. Leading Fiji toward healthier lifestyles and greater well-being depends not only on expanding access to quality care, but also creating pathways for Fijians to make healthier choices that improve longevity and reduce the risks of chronic disease. 129. Fiji requires a combination of policy reforms, actions, programmatic investment, and service delivery shifts to achieve this vision. This section presents a concise set of costed policy suggestions and programs the government can take now, to improve health outcomes and limit the long-term fiscal burden of the health sector. Table 13 summarizes the priority policy reforms, institutional changes, and programmatic investments required. Table 13: Implementation framework Achieving Fiji’s vision of Mo Bulabula, ka Bula Balavu Priority reform areas are intrinsically linked within and across pillars Vision: Mo Bulabula ka Bula Balavu Wishing you a healthy life and a long life Reform area 1: Reform area 2: Reform area 3: Reform area 4: Redesign health service Spur the adoption of health Build a modern health Strengthen stewardship delivery to meet the health behaviors in the population. workforce for the future. and data for a modern challenges of Fiji and the health system. Pacific. Cross-cutting: Enhance health resilience through health emergency prevention, preparedness, and response, using climate-smart solutions. Improve health equities though pro-poor and gender-sensitive interventions. 92 Fiji Health Sector Review 130. Implementation will be phased over time. The proposed sequencing will enable Fiji to take a stepwise and phased approach to its reform journey given the fiscal cost and the technical and administrative demands of such reforms. Phase I interventions address challenges that affect people’s lives daily. Their implementation, alongside process and impact evaluations will allow the Ministry of Health and Medical Services (MHMS) to “get the basics right” of its health sector transformation, which aligns with the strategic direction of Fiji’s national development plan, particularly its first three years of implementation, and its vision to enhance the provision of foundational policies and services that can tangibly improve the quality of life of ordinary Fijians. Reform area 1: Redesign health service delivery to meet the health challenges of the Pacific 131. Fiji’s health service structure should be ‘flipped’, from a system focused on hospital-based, curative care to one that prioritizes preventive care and disease management at the primary level. Primary healthcare-oriented systems offer the most cost-effective, equitable, and accessible route to improving population health. An effective primary healthcare (PHC) system promotes healthy lifestyles, diagnoses diseases early, and provides effective disease management. Hospitalizations for complications related to diabetes and hypertension can often be avoided if these conditions are managed early and adequately at PHC levels. The goal is to reduce the number of patients with chronic conditions and associated complications who end up seeking care in secondary and tertiary facilities. Box 7 describes how Sri Lanka has met similar challenges. 132. Although the emphasis should be on strengthening PHC, quality tertiary care capacity is still required. It offers protection against the catastrophic economic and health consequences for those who invariably become seriously ill. Although many of Fiji’s secondary and tertiary facilities have benefitted from recent upgrades, this necessitates some investment, for instance to renovate or rebuild the Colonial War Memorial (CWM) hospital. 133. Proposed priority 1: Transform PHC services with the introduction of a credentialed, digitally- enabled, climate-resilient, team-based PHC model. This includes the development of health sector policies detailing a comprehensive package of PHC services that considers Fiji’s evolving disease burdens, demographic and climate change impacts as well as investments in PHC infrastructure and equipment upgrades, including digital infrastructure. The government can implement the PHC reform through a phased approach (Table 14). Given the urgent need to improve population health at scale, some of Fiji’s largest catchment areas—for example, the medical areas of Valelevu, Nuffield, and Raiwaqa, collectively reaching over 150,000 people—could serve as implementation hubs for urban areas, accompanied by evaluations to inform a national roll-out. Similarly, evaluations of phase I implementation interventions in selected rural areas can be used to inform the roll-out and necessary adaptation of health sector policies and programs to rural settings. Fiji Health Sector Review 93 Box 7: The Sri Lanka Health Service Pyramid In the past three decades, Sri Lanka also allocated greater resources to hospital health services than primary healthcare (PHC) facilities (and to tertiary hospitals over secondary ones) with correspondingly higher utilization. This inverted pyramid of health expenditure and healthcare use risked trapping the country into a high-cost system, especially due to its high non-communicable disease (NCD) burden and the increased costs of treating complications in the hospital setting due to conditions not being adequately addressed at the primary or secondary care level. Figure 42: From To Flipping service delivery around, to strengthen the foundation for preventative care and chronic disease management Tertiary Care Tertiary Secondary Care Secondary Care Primary Care Prevention Primary Care and Health Promotion Source: US Public Health Service. 1994. For a Healthy Nation: Return on Investments in Public Health. The intervention to flip the Sri Lankan health service pyramid aimed to increase the utilization and quality of primary healthcare services, with an emphasis on the detection and management of non-communicable diseases in high-risk population groups. Starting in 2018, it envisaged a people-centered and comprehensive model of PHC service provision through three key pillars: (i) reorganizing PHC to meet the country’s epidemiological and demographic needs; (ii) strengthening the health sector through key system improvements; and (iii) enhancing information management to allow service delivery prioritization to change. As one of its key components, the Sri Lankan Ministry of Health addressed the disproportionate number of patients seeking NCD care in secondary and tertiary facilities at the expense of primary care settings. Efforts to accomplish this task include establishment of NCD Clinics, Healthy Lifestyle Centers, and “Happy Villages” to ensure the availability of a comprehensive set of health services at the community level. While outcome indicators are not yet available for the intervention, initial output data show that in the past five years, screening for cervical cancer in PHC has increased by more than 120 percent (from 107,551 to 238,000) and 550 facilities have the required capabilities to provide comprehensive and quality care. Source: World Bank. 2023. Sri Lanka Health Sector Technical Note: Strengthening Sri Lankan Hospitals’ Performance, Efficiency, and Contributions to Universal Health Coverage: A Summary of Evidence and Options, Sri Lanka: Primary Health Care System Strengthening Project, Implementation Status and Results Report, December 2023. 94 Fiji Health Sector Review 134. Proposed priority 2: Unleash the power of digital technology and data for health. Enhanced connectivity and the implementation of Fiji’s 2023–27 digital health strategy will successfully move Fiji toward digital patient records, NCD registries, telehealth services, access to just-in-time consultations with clinicians at higher-level facilities, and online learning, including in the Outer Islands. Fiji’s digital strategy, which was launched by MHMS in 2023, advocates for an enterprise architecture-based approach, ensuring that all digital solutions are compatible, interoperable, and sustainable in the long term. By following this structured framework, the policy will facilitate the integration of new technologies, promote human and financial resource optimization, and enhance the overall efficiency and resilience of the health system. Its successful implementation is contingent on (i) ensuring the basic infrastructure is in place at all levels of care, including in the most remote rural areas, for a digital transformation; (ii) building digital literacy among healthcare workers and the general population to use and trust technology and also be aware of its limitations; and (iii) dedicated human resources with the technical skills to design, maintain, and promote effective use of digital technologies and systems. 135. Proposed priority 3: Transform hospital services by creating a modern, integrated, digitally enabled diagnostic and treatment network for Fijian and Pacific patients. Strengthen the functioning and appropriate utilization of CWM as a tertiary care hospital and training facility, enhance the availability of diagnostics, and management of complex conditions in subdivisional facilities (secondary care), which are critical to improve patient pathways and outcomes. Investments in health facilities across the care spectrum (PHC, secondary, and tertiary) are highly symbiotic. Effective health promotion and a strengthened PHC system are crucial to preventing premature disability and deaths among the working- age population. It reduces the burden on Fiji’s secondary, tertiary, and specialist hospitals. At the other end of the spectrum, increased access to high quality tertiary care and clinical training is necessary to provide advanced health care services that cater to the Fijian demographic profile and disease burden, and by extension, the neighboring countries that rely on Fiji for specialized care. CWM, the Pacific’s largest tertiary and training hospital, serves as Fiji’s national referral hospital, and also cater to patients from other Pacific Island Countries, such as Tuvalu and Kiribati, where access to comprehensive tertiary care is limited. CWM also serves as a training hospital for SHWs from across the Pacific, majority of whom are trained in Fiji. The proposed Priority 3 does not include the recommendations of the CWM infrastructure master plan, which is expected before the end of-2025, generously supported by the Government of Australia. In the meantime, based on a recently concluded architectural and structural assessment conducted as part of the World Bank financed Fiji-COVID Emergency Response Project, engineering recommendations propose enhancing services and safety in three buildings that provide critical services to high patient volumes and provide critical services, the Children’s Ward, the Maternity Ward, and the 1965 building. Therefore, the proposed investments covered in this report is limited to activities required to bring these buildings up to Code in critical areas of seismic and structural stability,141 asbestos,142 and fire safety143 notwithstanding future resourcing needs for CWM per the Master Plan. Additional works have been recommended and costed, including communication, mechanical, electrical upgrades, roofing and base building refurbishments. 141. Review against New Zealand Society for Earthquake Engineering guidelines 142. Various work, health and safety standards, the investigation was only for asbestos-containing materials and excluded other potentially hazardous materials, such as lead paint. 143. In line with local and international building codes including Australia and New Zealand. Fiji Health Sector Review 95 Figure 43: Master plan for CWM Hospital: Identifying three buildings with high patient volumes, the Children’s Ward, the Maternity Ward, and the 1965 building Structural assessments have been developed for three priority buildings, to determine renovations requirements for optimal functionality: • Children’s Ward (Fiji COVID-ERP, WB) • Maternity Ward (Fiji COVID-ERP, WB) • 1965 building (Fiji COVID-ERP, WB) • CWM, a 550+ bed hospital with an average 85% occupancy rate, is symptomatic of an ailing service delivery system. Part of CWM’s overflow must be addressed by improving services at lower levels of care. 96 Fiji Health Sector Review Table 14: Indicative costing and timeline for reform area 1 Implementation of reform area 1: proposed phasing and indicative costing of interventions Phase I over the first 1–3 years Followed by gradual roll-out (phase II) over 4–7 years Define Fiji’s PHC service package and Roll-out PHC upgrades in test if it is fit-for-purpose in urban (e.g., strategically located health centers FJ$61–81m Valelelvu, Nuffield and Raiwaqa) and rural nation-wide. implementation hubs. Implement a digital transformation nation-wide and ensure digital Roll-out PHC upgrades in select FJ$19m foundations are in place at all levels of nursing stations nation-wide. care, including the Outer Islands. Implement highest priority upgrades at Additional operational costs of Fiji’s CWM’s Children’s Ward, Maternity Ward, digital FJ$8m and 1965 building. transformation. TBD by CWM renovation or rebuild. forthcoming Master Plan Phase 1 Phase 2 Total Indicative Cost: FJ$73m Total Indicative Cost: FJ$88m-108m (US$33m) (US$40-49m) Reform area 2: Spur the adoption of healthy behaviors in the population 136. Fiji needs bold action to tackle its NCD crisis, beyond what the health sector can do by itself. It needs to implement effective strategies that focus on lifestyle risk factors to delay the transition to NCDs and the development of multi-morbidity among people with NCDs: this includes partnering with finance, education, agriculture, and other ministries and bodies. In 2014, Fiji endorsed the Pacific NCD Roadmap for policy and regulatory guidance to address the NCD crisis. The Pacific Monitoring Alliance for NCD Action dashboard was endorsed in 2017 and Fiji has since developed a good foundation of NCD related policies and regulations. Although Fiji has introduced policies and interventions that have provided value for money in upper middle-income country (UMIC) settings, they are not ambitious enough to tackle Fiji’s NCD emergency and their implementation is weak. Fijian leaders have an opportunity to take measures that are commensurate with the scale of its NCD crisis. 137. Proposed Priority 4: Improve risk profiling, diagnosis, and management of NCDs through screening programs and use of digital NCD registries. Fiji has already adapted the Package of Essential Non- communicable Diseases services, as recommended by the World Health Organization (WHO), to the Fijian context, in the form of its Cardiovascular Risk Assessment and Management (CRAM). The CRAM system uses a color coding system to show the level of advancement of the disease. Currently, the folders of many of Fiji’s patients enrolled in CRAM are red or deep red, which means they are at an irreversible stage of the disease. Implementing large-scale screening programs for hypertension and diabetes in workplace settings, schools, and PHC facilities will allow for earlier diagnosis and improved disease management and a more robust referral system to secondary and tertiary care for complex cases. Integrated digital patient records and disease registries that can be disaggregated by sex, age, and other sociodemographic determinants are essential to effective risk profiling, screening, diagnosis, and the treatment process. This will be enabled through the implementation of Fiji’s 2023–27 digital health strategy, digital infrastructure (proposed priority 2), and increased capacity for data entry, disaggregation, management, and quality assurance. Fiji Health Sector Review 97 138. Proposed priority 5: Implement and evaluate population-level social and behavior change programs. The Government of Fiji (GoF) has recently adopted promising policies centered around prevention, including the National Policy on Healthy Catering and Sale of Food and Beverages for Government Ministries and Institutions, through which it intends to lead by example. The HSP, approved by Fiji’s Cabinet in April 2024 has significant potential to promote healthy behaviors among children and youth if well implemented and adopted by a majority of Fijians. Progress will be measured by (i) the share of students from early childhood to secondary school who are malnourished144 who are identified and referred for treatment; and (ii) the share of students meeting recommended guidelines for physical activity (disaggregated by boys and girls). Similar indicators could be promoted widely, including in workplace and community settings. The HSP also promotes school gardens that make available nutrient-rich vegetables in the school setting, which can serve as a tool to tackle Fiji’s triple burden of malnutrition among its children and youth. Behavior change interventions that assist health professionals, community leaders, and citizens to engage in health promotion are likely necessary considering high levels of acceptance of NCD risk factors. Evaluations should monitor progress against targets and inform improvements to increase impact. 139. Proposed priority 6: Leverage tax policies to encourage better diets. Policies to encourage healthier diets can offer quick wins that have translated into health gains in other countries. The government can optimize its current value-added tax (VAT) and tax policies to reduce Fijians’ high consumption of salt and sugar, which fuels the diabetes epidemic, and to encourage the consumption of healthier food. Fiji should also seek to eliminate trans-fats from the food chain as their consumption substantially raises the risk of heart disease. The implementation of the forthcoming Healthy Canteen Policy (estimated to be approved by Cabinet by May 2025) offers a good opportunity for this. It can be complemented by efforts to educate the public to choose healthier options through improved labelling in restaurants, shops, and market stands. Many countries have adopted successful policies in this area and their knowledge on how to achieve goals like the elimination of trans-fats can be leveraged and adapted to the Fijian context. Table 15: Indicative costing and timeline for reform area 2 Implementation of reform area 2: proposed phasing and indicative costing of interventions Phase I over the first 1–3 years Followed by gradual roll-out (phase II) over 4–7 years Enhance capacity for risk profiling, screening programs in workplaces, Mainstream tailored risk profiling schools and PHC thanks to digital NCD registries and FJ$7m FJ$9m facilities, with data entered on digital roll-out screening programs in all platform and sex-disaggregated NCD PHC facilities. registries. Enable healthy behavior and tackle risk Implement Healthy Canteen Policy, factors among children and youth in FJ$20m eliminating trans fat from school FJ$2m school (e.g. HSP). canteens. Catalyze population-level behavior FJ$ change (behavior change communication, Operating costs of school-based 16.5m FJ$6m digital prevention. reminders, availability of treatment). Remove salt and sugar from VAT Expanded behavior change exemption Minimal interventions FJ$22m and maximize SSB policies. on NCD prevention and control. Phase 1 Phase 2 Total Indicative Cost: FJ$43.5m Total Indicative Cost: FJ$39m (US$19.5m) (US$18m) 144. Underweight, stunted, wasted, overweight, or obese. 98 Fiji Health Sector Review Reform area 3: Build a modern health workforce for the future 140. A new human resource model combined with PHC upgrades will improve access to care. Meeting Fijians as close to home as possible with health promotion and preventative care will require a combination of solutions that leverage Fiji’s existing human resources in health and that are tailored to the needs of the population they are entrusted to serve. Fiji should consider a combination of actions to change the composition of its healthcare services teams. Recent government changes to salary packages should be carefuly evaluated to assess their impact on SHW career choices. The GoF should also remain responsive to workforce requests for training and to have a voice in health sector decision making. 141. Proposed priority 7: Retain and deploy Fiji’s healthcare workers to meet the needs of the population they serve. Fiji’s density of SHWs is in line with the WHO minimum threshold when both public and private sector SHWs are considered, however, the current allocation of SHWs is uneven (Annex 10). A comprehensive revision of the role delineation policy, which defines the range and services to be delivered at different levels of care, in collaboration with technical partners and a human resource agency specialized in the medical field, can help introduce more flexible staffing models, aligning resources with workloads and needs in both urban and rural settings (Table 16). Given the remoteness of many communities in Fiji, interventions to enhance basic care, including NCD prevention and control will benefit from upskilling the existing cadre of approximately 1,640 community health workers (CHWs). International studies that assessed interventions where appropriately supervised and equipped CHWs were trained to deliver blood pressure monitoring and counselling for lifestyle behavior change found they improved health and were cost effective.145 The updating of CHWs’ roles and responsibilities, supervision, training, and renumeration packages can be informed by global lessons (Box 4), that can be adapted and tested in a phased approach to meet Fiji’s needs, with possible variations in urban versus rural areas. It is also important to monitor the effects of recent salary increases in the public sector, emigration trends, and the impact of phase I interventions on SHW retention. Monitoring non-monetary drivers of resignations such as unclear career paths, poor working conditions, and burnout is an important elemant of an updated and improved human resource strategy. Table 16: Strategies for the reallocaton of public sector skilled health workers • Redeploy SHWs from low utilization hospitals to high utilization PHC facilities and update role delineation • Shift resources toward PHC facilities plan. • Create a roster of weekly PHC outreach activities from hospitals to understaffed health centers. • Enable subdivision administrators (cost center managers) to allocate and rotate staff with greater • Create greater flexibility for rotation of SHWs flexibility between facilities within their subdivision, adding resources to high-utilization facilities. 145. Gamage et al. 2020. “Effectiveness of a Scalable Group-Based Education and Monitoring Program, Delivered by Health Workers, to Improve Control of Hypertension in Rural India: A Cluster Randomized Controlled Trial.” PLOS Medicine 17(1):e1002997; Krishnan et al. 2019. “Cost-effectiveness and Budget Impact of the Community-based Management of Hypertension in Nepal Study (COBIN): A Retrospective Analysis.” Lancet Global Health 7(10):E1367-74; Jafar et al. 2020. “A Community-based Intervention for Managing Hypertension in Rural South Asia.” The New England Journal of Medicine 382(8):717–726. Fiji Health Sector Review 99 142. Proposed priority 8: Build human resource capacity to improve adherence to clinical guidelines and skills to use data for decision making. Fiji will update key clinical guidelines which need to be widely disseminated through in-person and online training. Further skills development to use data for decision making and improved patient outcomes, are worthwhile investments as part of Fiji’s efforts to digitize its health sector. In collaboration with teaching institutions such as the Fiji National University (FNU), the Pacific Community, and an international partner institution, an assessment should be carried out to identify opportunities to improve SHW training materials, including on NCD screening and treatment guidelines. Systems-level training, such as the World Bank Health Systems Flagship programs can also be offered, tailored to the Pacific and building on recent analytical assessments such as the Primary Health Care Performance Initiative (PHCPI) published in January 2024. Considering the high prevalence and tolerance of risk factors such as obesity, an independent study into the attitudes of health workers will help to improve the effectiveness of health promotion and adherence to the CRAM protocol. A broader capacity building program to strengthen the monitoring and evaluation capacity on program performance and cost effectiveness will help create a health systems performance culture that is evidence based and outcomes focused. Table 17: Indicative costing and timeline for reform area 3 Implementation of reform area 3: proposed phasing and indicative costing of interventions Phase I over the first 1–3 years Followed by gradual roll-out (phase II) over 4–7 years Roll-out SHW redeployment and Update the role delineation policy capacity building on data for decision and MHMS HR strategy for improved FJ$5m FJ$2.5m making (building on digitization of the redeployment and guidelines for SHWs. sector). Upskill and strengthen governance of CHW / auxiliary worker scale-up community health workers (CHWs) and FJ$16m programs (training, digital tools, and FJ$37.5m auxiliary worker programs (training, allowances). digital tools and allowances). Strengthening retention of nurse and Stock-taking at current training auxiliary programs (e.g., dieticians) FJ$2.5m programs (initial and continued following training and labor-force FJ$5m education) and testing of new materials. assessments, including on the impact of recent reforms. Phase 1 Phase 2 Total Indicative Cost: FJ$23.5m Total Indicative Cost: FJ$45m (US$10.5m) (US$20m) 100 Fiji Health Sector Review Reform area 4: Strengthen stewardship and data for a modern health system 143. Proposed priority 9: Strengthen public sector capacity for strategic purchasing from the private sector to get better value for money from private providers. The private sector is playing a growing role in Fiji’s health system. With careful planning, it could be leveraged to support health service delivery, including among under-served populations in urban settings. To fully harness the contribution of the private sector, the GoF can strengthen its capacity for strategic purchasing in the Ministry of Finance, Strategic Planning, National Development and Statistics (MFSPNDS) and MHMS through capacity building and learning exchanges to countries that have successful public-private sector engagement models in place. Existing programs to purchase services from the private sector could, if better monitored, be transformed into strategic opportunities for screening, preventive care, and health promotion. The GoF could build on its existing efforts to harness private general practitioners and healthcare providers to expand the reach of its NCD prevention programs. Evaluations to ascertain the cost effectiveness of current and future arrangements are necessary to assess value for money. Fiji Health Sector Review 101 Box 8: PHCorp: Towards Strategic Purchasing in Malaysia Protect Health Corporation Sdn Bhd (or PHCorp) was incorporated as a private limited company in December 2016 by Malaysia’s Ministry of Health (MOH). PHCorp is a fully MOH Malaysia- owned, non-profit enterprise, with its assets and income and/or surplus intended for social purposes. Its goals are to introduce innovative financing mechanisms to ensure improved access to healthcare and health services that are effective and offer value for money; serve as an efficiently functioning competitor (and a benchmark) for other health financing entities in Malaysia that have a similar function; offer an efficient, effective, and equitable financing model for healthcare, to supplement public funding for health; and to enhance access to good quality services by improved utilization of the public-private healthcare ecosystem in Malaysia. The intended functions of PHCorp are quite broad in scope and include the establishment and operation of any kind of medical and health insurance business, to act as an administrator for such businesses (such as a third-party administrator) or as a purchaser of services, and more generally to establish and operate any health financing mechanism that serves public interest, as determined by the Minister of Health. Thus far, PHCorp’s role has been limited to its support for private sector screening services for the poor (the PeKa B40 program) and the purchase of selected consumables and vaccines under relatively rigid funding lines of the Ministry of Finance. Within these operational parameters, PHCorp has negotiated prices with (private) general practitioners to provide free health screening to beneficiaries, and with private laboratories to carry out (specific) blood tests, while meeting pre-specified quality standards. Separately, PHCorp has also negotiated prices for cardiac stents and intraocular lenses from suppliers in the private sector. If the strategic purchasing function is defined as “the capacity to decide on the type of services to be purchased, from whom and the choice of payment mechanism, informed by social priorities, strong information systems and accountability mechanisms,” it is perhaps too early to conclude (as of May 2024), that PHCorp is a strategic purchasing entity. It has, however, served as a major platform for government engagement with private sector providers, learning by doing, and promoting MOH staff capacity in contracting and purchasing operations. Its value to the government is apparent from increased resource allocations to it over time. Source: MOH, Planning Division, and discussions with PHCorp Team. 144. A strategic purchasing entity makes decisions about allocating a given pool of funds among providers, based on population health needs, priorities, and information. The three main types of (inter-related) decisions relate to: • Which services to buy and from where to buy them. • How and how much to pay healthcare providers. • Monitoring and evaluation of how the money is being used. To execute these decisions, strategic purchasing entities must be able to effectively carry out actions such as: • Obtaining information on population health needs, available care, medicines, and diagnostic services in the private sector (and sometimes the public sector) to arrive at purchasing decisions. • Specifying the services to be made available to beneficiaries, selecting the providers, and contracting with providers to meet specific quality standards. • Developing payment mechanisms (e.g., incentives for specific types of services, lump-sum payments or per unit of service, or some combination of these, incentive payments, etc.) and where necessary 102 Fiji Health Sector Review generating cost data to inform pricing negotiations. • Monitoring the volume and quality of services, and the benefits to the population. • Ensuring that payment mechanisms and services covered are aligned to the available budget. Well-functioning strategic purchasing functions also involve establishment of mechanisms that enhance accountability not just to the funder (e.g., GoF) but also to the beneficiaries (e.g., Fiji’s population). 145. Proposed priority 10: Strengthen public financial management capacity. This review has identified three areas of focus to support MHMS in close coordination with MFSPNDS on broader PFM reforms: i). Strengthen budget execution across standard expenditure groups and capital works: This includes creating a cross-sectoral task team (MFSPNDS, the Ministry of Civil Service, local government, and MHMS) and conducting a root-cause analysis of MHMS’ poor budget execution, with guiding actions for each government agency to take. ii). Strengthen MHMS capacity for the roll out of GoF’s updated financial management information system (FMIS) and ensuring interoperability with sector specific tools. This transition will be accompanied by capacity building activities to use the enhanced FMIS systems and data for improved accountability and evidence-based decision making. iii). Strengthen capacity to manage public finances effectively in emergencies. This includes building capacity to enable Fiji to access context-appropriate surge financing instruments, strengthen agile treasury systems (e.g., ensure appropriate procurement rules are triggered during an emergency), and set up systems to prioritize and track climate preparedness spending. Table 18: Indicative costing and timeline for reform area 4 Implementation of reform area 4: proposed phasing and indicative costing of interventions Phase I over the first 1–3 years Followed by gradual roll-out (phase II) over 4–7 years Strengthen capacity for strategic Enhance digital capacity for strategic purchasing of PHC services from the FJ$1m FJ$2.5m purchasing. private sector. Undertake bottleneck analysis and Enhance digital capacity to support recommendations for improved MHMS FJ$0.3m FJ$2.5m PFM at MHMS. budget execution. FJ$1.7m Strengthen PFM capacity at MHMS. Phase 1 Phase 2 Total Indicative Cost: FJ$3m Total Indicative Cost: FJ$5m (US$1.3m) (US$2.3m) Fiji Health Sector Review 103 146. MHMS can lead an update of its results framework to encourage and foster data analysis and accountability for its health system transformation. Table 19: Examples of indicators for improvements in data analysis Type Example Indicators for Improvement in Data Analysis • Share of health centers and nursing with gender-disaggregated patient data captured electronically by disease classification, by region. • The share of population for which patient data are linked across facilities by disease classification. • Implementation and frequency of surveys that capture information on healthcare use and Data generation spending by households by self-reported condition. • Availability and linkage of private sector patient electronic records. • Availability of patient outcomes data indicators (e.g., hospital mortality, re-admissions, diagnosis of NCDs under screening, and whether treated and under control). • Patient satisfaction surveys following healthcare use. • The number of Masters and PhD students engaged in research on health systems issues in the Pacific. • Graduate- and undergraduate-level course offerings and enrolments in health systems financing/healthcare delivery/public health. • Enrolments from the Government of Fiji, researchers, and other stakeholders in Fiji in World Analytical capacity for health Bank Flagship programs. systems research • Research projects undertaken by researchers in FNU, University of the South Pacific, University of Fiji, and other Pacific tertiary teaching institutions, in collaboration with MHMS, that use administrative data on healthcare utilization from health facilities. • Research projects led by Fijian researchers on evaluation of government interventions in the health sector. 147. In a complementary effort, Fiji has obtained a grant from The Pandemic Fund to strengthen its pandemic preparedness through a One Health approach. Fiji’s intersecting climate, antimicrobial resistance (AMR), zoonoses, and infectious disease threats, compounded by increasing human health vulnerabilities related to NCDs and social inequities, demand a One Health approach to all-hazards emergency preparedness and response. The activities align with an overall health systems transformation and center around three areas: (i) health emergency workforce capacity; (ii) community-based surveillance; and (iii) One Health information management and analysis. They represent a paradigm shift from reactive, response-based approaches to a deeper understanding of multi-hazard systemic risk, and investment in pre-emptive and anticipatory measures. The Health and Economic Returns to these Strategies 148. Fiji could achieve significant health gains if it could catch up with the health outcomes of its UMIC peers. The strategies proposed in this report reflect the experiences of UMICs and high-income countries that have had success in addressing morbidity and mortality due to NCDs and other health conditions. If Fiji were able to effectively implement these strategies to achieve the mortality and morbidity outcomes of a typical UMIC by 2050, it would be able to lower its disability-adjusted life years (DALYs), a composite measure of disease burden, by almost 22 percent over the period from 2024 to 2050, and enjoy a boost to life expectancy at birth of almost 10 years relative to its long-term trend if it did not implement such strategies. If it could achieve equivalence faster, by 2040, it would reduce its disease burden by almost one-third (Annex 4). 104 Fiji Health Sector Review 149. These health gains are likely to translate into significant economic gains (Annex 5). This would be both because healthier workers tend to be more productive and because of having more people of working age who are able to take part in the workforce. Under a scenario where Fiji achieves the health outcomes of a typical UMIC by 2040, it would enjoy an extra 0.30 percentage point of growth of real GDP per capita annually during 2024–50 relative to the status quo. Under a scenario where it only catches up by 2050, its growth rate of real GDP per capita would be a still-significant 0.15 percentage points higher annually. These economic gains are separate from any gains in health, which are valuable in and of themselves. When the monetary value of longevity gains is also included, the increase in the average annual rate of growth could amount to an extra 0.90 percentage points annually between 2024 and 2050 relative to the status quo if catches up with a comparator UMIC by 2040; and an 0.55 percentage points extra annually if equivalence is achieved by 2050. 150. The potential economic gains are likely to significantly exceed the costs of the proposed investments and are comparable to returns from high-performing index funds. In Annex 7 we conservatively assess that the annual average real rate of return from investment could range from 7.6 percent to 10.2 percent over the period 2024 to 2050 if Fiji’s health outcomes were to catch up with a typical UMIC by 2040. Even if the catching up took until 2050, the average real annual rate of return would still be quite high, ranging from 4 percent to 8 percent between 2024 and 2050. To put these returns in context, the long-term annual real rate of return from investing in a high-quality index fund such Standard and Poor is approximately 6.4 percent. Fiji Health Sector Review 105 106 Credit: Fiji Ministry of Health and Medical Services Photo Fiji Health Sector Review About the Report: Scope and Methods 151. This work was conducted at the request of the Ministry of Finance, Strategic Planning, National Development and Statistics (MFSPNDS). MFSPNDS requested the World Bank’s support to conduct a systematic review of the health sector, to take stock of its current performance in tackling the country’s major health challenges and propose policy suggestions and programs for improving its performance. A working group with members from MFSPNDS and the Ministry of Health and Medical Services (MHMS) was established to facilitate data collection and oversee progress of the review. It was agreed that the review would cover: (i) health financing, including health expenditure projections to 2030; (ii) service delivery, including a supply-side readiness assessment of services at all levels of the health system, and redevelopment options for Colonial War Memorial Hospital; (iii) human resources for health, including assessment of stock and flow as well as projections of needs and supply to 2030; and (iv) options and models for the Government of Fiji to engage private healthcare providers to improve population health outcomes. 152. This Health Sector Review (HSR) aims to identify transformational investments needed in Fiji’s health sector to respond to its changing epidemiological and demographic needs. It includes an assessment of Fiji’s health system performance and the health financing needs of Fijians going forward. This is followed by an analysis of the proximate drivers of performance, including quality of care in service delivery, efficiency in resource use, regulatory issues, infrastructure and human resource gaps, and behaviors at risk of causing ill health. It also highlights the economic implications of actions to address performance gaps using well-known models of economic growth, and measurement of the economic value of longevity (the annexes provide the detailed methodologies underpinning this work). The review concludes with policy suggestions and programs to guide short- and medium-term health sector investments to help address performance gaps and the resource implications of these investments. It reports the expected returns on investment in terms of health and economic gains under different reform scenarios using standard techniques of cost-benefit analysis, as described in the annexes to this review. MFSPNDS had also expressed an interest in assessing the effectiveness of outsourcing health services to private practitioners, however, access to the data needed for a full assessment of private health service practitioners was not available during the drafting of this report. 153. The review was done through analysis of primary and secondary data from Fiji, global datasets, and key informant interviews. Interviews were conducted with health sector and finance officials mandated with making important budget, policy, and administrative decisions about the Fijian health sector. The HSR has also benefited from an extensive survey of the international literature on healthcare financing, service delivery, human resources for health, health expenditure projection models, modeling of health impacts on economic outcomes, and service delivery costs, as well as assessments of effectiveness of health sector interventions (what worked and did not work in other contexts, and why), alongside health sector reviews previously undertaken for Fiji. 107Health Sector Review Fiji Fiji Health Sector Review 107 154. Fiji’s performance is compared to other countries using an international benchmarking exercise, in line with the 2023 Public Expenditure Review.146 This report draws upon data from domestic and global sources to measure Fiji’s health sector performance and for international comparisons. The benchmarking framework introduced in the Public Expenditure Review chose two sets of comparator countries, based on the following characteristics: (i) population; (ii) gross domestic product per capita (current US$); (iii) tourism receipts; and (iv) island states or continental countries with lower foreign market access. These peer countries, drawn from the Caribbean and other small island developing states, will be used as reference points throughout this report: • Aspirational peer countries: Barbados, Maldives, St. Lucia, and Mauritius. • Structural peer countries: Belize, Jamaica, Samoa, Sri Lanka, and Tonga. • In some cases, Fiji will be compared to upper middle-income countries that have a strong track record of investments in primary care and health promotion. 155. The HSR also builds on, draws on, and complements other important analytical products developed by the World Bank in the Pacific. This includes the Primary Health Care Performance Initiative (PHCPI) which was initiated by Fiji and three Pacific neighbors, Kiribati, Solomon Islands, and the Republic of the Marshall Islands, in December 2022 and concluded in December 2023. Findings from the PHCPI highlight important gaps in the delivery of comprehensive care for various health conditions, particularly non- communicable diseases (NCDs); infectious diseases; and reproductive, maternal, neonatal, adolescent, and child health (RMNCAH). It points to disparities between urban and rural populations in financial barriers to care, with rural areas experiencing higher rates of financial deprivation related to medications, and the need for comprehensive strategies to increase service coverage and reduce disparities in access to essential health care and better health outcomes for all Fijians. The HSR also builds, draws on, and complements previous analytical work completed by the World Bank in Fiji, such as the Fiji and the Nine Pacific Island Countries Public Expenditure Reviews and the Fiji Development Policy Operation with a Catastrophe Deferred Drawdown Option (2024), and the Fiji Country Economic Memorandum (2024). Data and Information Ecosystem for Health Policy 156. In the process of preparing this health sector review, the need to strengthen the availability of data for decision making became apparent. The data gaps in question relate to the unavailability and/ or incompleteness of health sector data, data that are not available in a form suited for policy analysis, and limited human resource capacity for the analysis of the data that are available. W 157. Fiji’s digital health strategy, if fully implemented, will enhance data accessibility, patient experiences of the health system, and patient outcomes. While Fiji has developed and adopted a national digital health strategy, its implementation remains nascent. Many of the major constraints in accessing data that the study team encountered would be overcome if the digital health strategy were fully implemented. The digital health strategy would not only improve the flow and timeliness of data for administrative and policy analysis purposes, but also help to enhance health service delivery and contribute to improving patient experiences and outcomes. The digital health strategy advocates for an enterprise architecture-based approach, ensuring that all digital solutions are compatible, interoperable, and sustainable in the long term. 146. World Bank. 2023. Fiji Public Expenditure Review. Fiji Health Sector Review 108 Fiji Health Sector Review 158. Lack of up-to-date data that link financial flows to the use of health services limits the ability to carry out effective and timely policy assessment and financial risk analysis. Financial flow data for the health sector are typically available on a regular basis from the government, private insurance annual reports, and the household income and expenditure survey conducted by the Bureau of Statistics once every five years. However, the data cannot be easily linked to health service use patterns, since the household survey does not inquire about the health conditions for which expenditure was incurred, or whether a public or private service provider was used, only whether expenditure was for inpatient or outpatient care. Few studies are available on the cost of healthcare provision, and those that exist are now dated, so that allocation of public spending by type of condition is not readily available. In the absence of these data, it is difficult to assess the fiscal consequences of alternative policy interventions, or their implications for financial risks from illness borne by households. Information on costs is also necessary for to help provide price guides in negotiations with private providers. 159. Strengthening data on service quality and health care utilization, especially for NCDs, should be a priority, including using electronic patient records. Fiji gathers considerable information on RMNCAH services, with the most recent data gathered from the Multiple-Indicator Cluster Survey in 2021. The Service Availability and Readiness Assessment also provides additional information on the availability of human resources, basic equipment, and drugs for a set of tracer conditions, including for RMNCAH and NCDs. While availability of information on inputs is obviously relevant, information on the quality of clinical services provided is crucial for identifying training and service delivery gaps and designing interventions to improve access to care, whether in terms of staff training, or outcomes such as the proportion of hypertensive cases which have been diagnosed and are under control. The World Health Organization (WHO) STEPS survey is a useful means of gathering this information, but its frequency is very low, and was last conducted in 2011 in Fiji, with the next round scheduled for this year. Given the high priority assigned to NCD-related actions, Fiji will need to make better use of its facility-level (administrative) data, and to move from paper records to electronic databases in primary healthcare facilities, as its digital infrastructure is strengthened. Good quality administrative data can also be used to generate cost estimates for NCD services at different facility levels and understand the drivers of the cost of care. 160. Fiji will also benefit from strengthening the network of local researchers to investigate policy questions in the health sector. This would include economic analyses and program evaluations, in collaboration with local and regional universities, development partners, and the MHMS. The value of local research teams stems from their deep knowledge of the Fijian health sector, proximity to the policy issues and data, and access to networks within the government and elsewhere. Fiji has the additional advantage in that three leading Pacific universities are based in Suva and Lautoka (Fiji National University, University of Fiji, and University of the South Pacific). These offer a pool of researchers to support a research program with a focus on health sector performance and financing. Mechanisms to expand and strengthen the pool of available researchers could include training opportunities in global health (including leadership programs such as the flagship program of the World Bank), access to data (including administrative data from the government), funding for investigation of research questions of policy interest, and collaboration opportunities with strong international partners (both academic and multilateral). and financing. Mechanisms to expand and strengthen the pool of available researchers could include training opportunities in global health (including leadership programs such as the flagship program of the World Bank), access to data (including administrative data from the government), funding for investigation of research questions of policy interest, and collaboration opportunities with strong international partners (both academic and multilateral). Health Sector Review Report 109 Photo Credit: Fiji Ministry of Health and Medical Services 110 110 Fiji Health Sector Review ANNEXES Annex 1: Forecasting Health Expenditure Requirements for Fiji 2021-2050 Overview Health expenditure projections typically rely on two main approaches.147 The first uses a growth-accounting approach that breaks down the drivers of health spending into three main categories: demographic factors, growth in national income, and a category that encompasses all other factors, sometimes referred to as the “residual”. The residual is intended to capture the impact of drivers of health spending that are often hard to assess, such as changes in productivity, the regulatory environment, and in the prices of health services relative to non-health sector commodities as well as technological change, climate effects, etc. Predicting health expenditure under this approach essentially requires predicting the combined impact of each of these factors on health expenditures. Simplified versions of this approach have also been used to assess the impact of improved health and longevity on future health expenditure profiles.148 The second approach predicts health expenditures through a more detailed assessment of demand and supply-side factors that influence healthcare use and spending and can take several forms depending on data availability. For instance, micro-level (or household-level) data can be used to construct models of demand for health services by econometric methods. Health services demand in this case is typically taken to depend on age, gender, household income, health status, household size and composition, insurance status, etc. Projections of these characteristics can be used to predict the demand for health services of different types in future years. Translating the demand for different types of health services into monetary values requires estimates of the per unit costs of health services of different types, in both the public and private sectors, and may include accounting for wage growth in human resources for health, productivity growth (and thus lower costs), etc. Ultimately though, forecasts under the second approach also require an understanding of future population profiles, income growth, and other factors (residual characteristics) to predict future health spending. Because previous work suggests that forecasts based on a more detailed modelling of health spending drivers and forecasts using the growth-accounting approach turn out to be not too different from each other,149 we opted for the growth-accounting method to forecast Fiji’s health expenditure, given the data requirements are not onerous. 147. Rocha et al. 2021. “Financing Needs, Spending Projection and the Future of Health in Brazil.” Health Economics 30(5):1082-94. 148. See for example Harris. and Sharma. 2018. “Estimating the Future Health and Aged Care Expenditure in Australia with Changes in Morbidity.” PLOS ONE 13(8): e0201697. 149. Rocha et al. 2021. op. cit. Health Sector Review Report 111 Fiji Health Sector Review The Growth-Accounting Approach to Projecting Health Spending: Introduction The approach starts with postulating the following relationship: HE denotes national health expenditure. Y denotes real GDP (or income). P is an indicator of demographic characteristics (population size, and its gender and age composition). R is the residual component. Because our goal is to project health spending, our interest is in understanding the change in health spending going forward. It can be checked that the “change” in HE, denoted as ΔHE, is mathematically given by: The first term on the right-hand side of the equation captures the effect of health spending due to growth (or change) in real national income. The second term captures the impact of changes in demographic characteristics on health expenditure, and the third captures the impact of all other (or residual) factors. Most of the data corresponding to the above terms are available in the literature but in a form that is not captured by the terms in the above equation. Thus, additional manipulations of the above equation were required to reach the following: The terms can be interpreted as follows: 112 Fiji Health Sector Review It turns out that the data for the variables in the right-hand side of the above equation can either be directly obtained from the literature or estimated from available Fijian data. Once the annual growth rate is known, it is straightforward to forecast the level of health expenditure for future years, starting from an initial level of national health spending. Forecasting Fiji’s national health spending requires considering each of the three drivers of health spending in turn and aggregating them. The remainder of this annex describes the approach used to estimate these parameters and the aggregation method for the combined effects of the three main drivers. The main advantage of this method is that it is relatively simple, while still capturing three key features of projections of health expenditures—changes in age structure, population size, and changes in income. Its main weakness is that it does not allow for a straightforward assessment of the implications of alternative courses of action, relative to the status quo (e.g., increased investments in primary care, outsourcing of services to private practitioners, improved efficiency in the provision of services). Although this model may still serve to produce a baseline (and is the workhorse for baseline projections), comparing its projections to the health expenditure implications of alternative courses of action will require the development of variations to the model. Impact of Demographic Structure Changes on Health Expenditure Projections of Fiji’s population broken down by age and gender in 5-yearly intervals from 2019 to 2050 were obtained from the United Nations Population Statistics division portal (https://population.un.org). In order to understand how changes in population size and its age and gender composition affect health spending requires one other piece of information: healthcare spending per capita by age and gender for Fiji. Because of the focus on population age structure on health expenditure projections, these data were only required for a single year, namely 2019, the starting year for the forecasts. Per capita health spending was assumed to remain unchanged in real terms for the entire forecasting period when considering the effects of changes in population size and age composition. Published information on per capita health expenditures by age and gender is unfortunately rare, especially for middle-income and upper middle-income countries (UMIC) in the Asia-Pacific region. For Fiji, information was only available on per capita inpatient spending, from National Health Accounts data150 (Figure A1.1). Because inpatient spending constitutes only a small portion of national health spending in Fiji and liable to produce biased estimates, and gender-specific health expenditure data from Australia and Malaysia were also obtained. Given the relatively similarity of the shares of disease burden across conditions between Malaysia and Fiji, it was decided to use Malaysian data (appropriately calibrated to Fiji’s national health spending in 2019) for forecasting purposes (Figure A1.2). 150. MHMS. 2017. Fiji National Health Accounts: National Health Expenditure 2011-15. Suva: Government of Fiji. Fiji Health Sector Review 113 Figure A1.1: Per capita spending by age in Fiji using inpatient expenditure data, 2019 1400 1200 1000 800 (in FJ$) 600 400 200 0 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70+ Age-Intervals (in Years) Source: Authors’ estimates based on data from MHMS (2015) and World Development Indicators database.. Figure A1.2: Health spending per capita, by age group in Fiji, using age weight from Australia and Malaysia, 2019 1800 1600 1400 Spending per person (FJ$) 1200 1000 800 600 400 200 0 <5 5_9 9_14 15_19 20_24 24_29 30_34 35_39 40_44 45_49 50_54 55_59 60_64 65_69 70_74 75_79 Age Intervals (in years) Malaysia Australia Source: Authors’ estimates, based on data from HPRA, HSR and IHSP (2013), WHO (2020) and the World Development Indicators database. The age-specific per capita expenditure estimates for 2019 were multiplied by the population in the corresponding age group for each year from 2020 to 2050 to generate forecasts of changes in health expenditures that were solely due to changes in population characteristics over time (population size and the share of population in different age intervals). 114 Fiji Health Sector Review Impact of Income Changes on Health Expenditure The impact of national income growth on health spending depends on the magnitude of income growth, and how health spending rises in response to income, all other aspects being equal. Data on the Forecasts of Rate of Growth of National Income in Fiji Until 2050 Forecasts for growth of real GDP are available from the International Monetary Fund for all years until 2032.151 For the years 2033–50, two scenarios were considered. Under the first scenario, real GDP in Fiji was assumed to grow at the annual rates forecasted by International Monetary Fund until 2032. Under the second scenario, real GDP was assumed to grow at the annual average rate experienced by Fiji during the period 2001–19, i.e., the two decades preceding the emergence of COVID-19. Pre-COVID years were considered to avoid the projected rates of growth being influenced by the dramatic economic declines during 2020 and 2021, in Fiji and elsewhere. The forecasts used below are labelled Scenario 1 or Scenario 2 depending on the national income forecast used. Data on the Income Elasticity of Health Expenditure Income elasticity of health expenditure is a “unit-free” measure/parameter of the responsiveness of health spending with respect to changes in national income, indicating the ratio of the percentage change in national health spending to the percentage change in national income. No estimates of income elasticity of health expenditure are available for Fiji but there is a considerable literature estimating this parameter for other countries and in cross-country analyses.152 The consensus in the literature is that the income elasticity of health expenditure is about 1: that is, for a given percentage change in national income, there is an equivalent percentage change in health spending. Notwithstanding this consensus, a single income elasticity of national health expenditure estimate is not well suited to health systems that have differential rates of growth of public and private spending. Indeed, one might expect that public spending, subject as it is to the exigencies of budgetary discussions, might be less responsive to national income growth than say, private spending (whether out-of-pocket, or funded via private insurance). Given that Fiji has experienced a rising share of private spending on health over the last two decades, the approach followed here used separate estimates of income elasticity for public and private health spending, and then added the resulting forecasts for public and private expenditure to arrive at estimates of national health expenditure growth due to changes in national income. Xu and colleagues153 provide a range of estimates of income elasticity of private expenditure from +0.842 to +1.503 based on cross-country analyses from WDI data. Khan and Mahmud154 used WDI data to estimate private expenditure income elasticities for a set of 10 South Asian countries, yielding a range from +0.912 to +1.452. Based on these studies and a careful assessment of the econometric methodologies used and their impacts of elasticity estimates, the projections reported here use an income elasticity of private expenditure of +1.1, which is around the mid-point of the range of estimates in the two studies. The estimate of income elasticity for public spending on health used for projections was +0.7, based on the findings from Xu, Saksena, and Holly whose estimates of this parameter from cross-country analysis for a set of middle- and high-income countries ranged from +0.661 to +0.702. 151. IMF. 2023. “Republic of Fiji: 2023 Article IV Consultation – Press Release and Staff Report.” International Monetary Fund Country Report no. 23/238. Washington, DC: International Monetary Fund. 152. See for example Xu, Saksena, and Holly. 2011. “The Determinants of Health Expenditure: A Country-Level Panel Data Analysis.” Working paper. Geneva: World Health Organization; Khan and Mahmud. 2015. “Is Healthcare a Necessity or Luxury? Empirical Evidence from Public and Private Sector Analyses of Southeast Asian Countries.” Health Economics Review 5:3. 153. Xu., Saksena, and Holly. 2011. op. cit. 154. Khan. and Mahmud. 2015. op. cit. Fiji Health Sector Review 115 Impact of the Residual Component of Health Spending As noted above, this factor is intended to capture hard-to-measure drivers of health spending—health technology, productivity changes, organizational changes, differential medical and general inflation rates, climate effects, etc. The “residual effect” can simply be assumed at different levels as part of a sensitivity analysis. In their study for Brazil, for instance, Rocha and colleagues155 simply assumed the rate of growth of health spending due to the residual factor to range between -0.75 to +0.75 percentage points. Or it can be estimated from past trends in the data on population, health expenditures, and national income. Because the residual factor is likely to vary between public and private spending, the strategy followed for projecting Fiji’s national health spending is to generate estimates the residual factor separately for the two components. Estimates of the growth in the residual factor in health spending were obtained for public spending in health as follows (a similar method was used for estimating the growth in the residual factor for private spending and is not discussed here). First, the projected impact of demographics on public sector health expenditure during 2001-19 was assessed using data on population by 5-yearly age intervals for 2001–19 from the UN Population Statistics Division and age-specific health spending reported in Figure A1.1, after adjusting for the share of public spending in Fiji’s national health expenditures in 2019. Next, using available data on real national income (GDP) for Fiji from the World Development Indicators (WDI) database for 2001–19, (baseline) 2001 data on public health spending for Fiji (also from the WDI database), an estimate of income elasticity of public spending on health of 0.7, a forecasted series of changes in public sector health spending due to income was constructed for the period 2001–19. The resulting projections of public sector health spending due to income and population changes were aggregated and then compared to the data on the actual public spending on health for 2001-19 to estimate the residual. The average annual rate of growth of the residual was then estimated as “r” from the compounding formula: The denominator in the above equation is the projected health expenditure growth if occurring solely due to population and income changes. The rate of growth of public sector health spending due to residual factors was estimated as -0.69 percentage points annually, which is at the lower end of the range used for projecting health expenditures for Brazil, as reported in Rocha et al. In contrast, the residual for private spending on health grew at +3.90 percentage points annually, which lies in the middle of the range for estimates observed for other middle- and upper middle-income countries: less than estimates for Chile, Turkey, and Indonesia, but higher than corresponding estimates for China and India.156 However, the residual rate of growth estimates for China and India are for public and private expenditures combined, and thus likely understate the residual for private spending on health in these countries. 155. Rocha et al. 2021. “Financing Needs, Spending Projection and the Future of Health in Brazil”. Health Economics 30:1082-94. 156. De La Maissoneuve and Oliviera Martins. 2013. “Public Spending on Health and Long-term Care: A New Set of Projections.” Economic Policy Paper no. 6. Paris: OECD. 116 Fiji Health Sector Review Aggregating the combined effects of population and income growth and the residual factor on health expenditures in Fiji As noted above, 2019 was taken as the base year for constructing health expenditure forecasts for Fiji. Starting with 2019, health spending for 2020 was constructed by adding up changes accounted for by population, income growth, and the residual factor. Next, 2020 was treated as the base, and changes in health expenditures in 2021 compared to 2020 were then generated by adding up the health spending due to the three factors under consideration, and so on, until 2050. Our preferred set of estimates (based on the Malaysian data on per capita health expenditure weights) are included in the main report and are also depicted in Figure A1.3. Figure A1.3: Health expenditure forecasts for Fiji: Health expenditure weights from Malaysia 1800 8 Share of Health Expenditure Needs Per Capita Health Spending (in FJ$) 1600 7 1400 6 1200 5 in GDP (%) 1000 4 800 600 3 400 2 200 1 0 0 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 Year HE/GDP Ratio Scenario 2 Per Capita Health Exp Scenario 2 Source: Authors’ estimates. Figure A1.4 reports health projections if per capita health expenditure weights from inpatient services from Fiji were used instead. As indicated above, this is not our preferred forecast; in addition, the projections are similar, given the relatively small share of inpatient spending in Fiji compared to outpatient care. Figure A1.4: Health expenditure forecasts for Fiji: Health expenditure weights for inpatient services in Fiji 1800.00 8.00 1600.00 7.00 1400.00 6.00 As Share of GDP (%) 1200.00 5.00 (in FJ$) 1000.00 4.00 800.00 3.00 600.00 2.00 400.00 200.00 1.00 0.00 0.00 2019 2021 2023 2025 2027 2029 2031 2033 2035 2037 2039 2041 2043 2045 2047 2049 Year Health Expenditure/GDP Ratio Scenario 1 Health Expenditure/GDP Ratio Scenario 2 Per Capita Health Exp Scenario 1 Per Capita Health Exp Scenario 2 Source: Authors’ estimates. Fiji Health Sector Review 117 How to Interpret these Projections of Health Spending for Fiji? The projections obtained above are best thought of as providing the expected outcomes of a “business as usual” scenario, since they are constructed around past patterns in the use of health services and changes in health status, public expenditure patterns, and growth of private spending, conditional on available income growth projections. Should circumstances change, such as a major reallocation of public resources towards (or away) from the health sector, that would have implications for private spending and overall health spending. These implications cannot be predicted by the existing model, which is essentially driven by past trends, with many of the key effects captured by the “residual factor”. Nor would these projections capture the effects of major new investment in preventive care or health promotion, as these will impact not just health outcomes, but also economic growth. Integrating the effects of such investments into health expenditure forecasts requires a careful assessment of their impacts on economic growth and follow-on implications for public and private health spending. Annex 2: Definition of Health Expenditure Aggregates and Exchange Rate Total health expenditure includes current health expenditure (CHE) and capital expenditure, and both private and public expenditure on health. Public expenditure on health includes government financing and compulsory financing schemes such as social health insurance. Government financing includes the domestic contribution of the government raised through general consolidated revenue collection and on-budget external financing—both on and off system.157 Private health expenditure can also be split between domestic and external financing, and includes out-of- pocket payments, private external expenditure, and private health insurance. Source: World Development Indicators DataBank, World Bank, https://datatopics.worldbank.org/world-development-indicators/. 157. On Budget means that the development partner funding provided for specific purposes is transparent and reflected in the government’s budgets, and On System means that where possible, these funds flow through the government’s own financial management information system. 118 Fiji Health Sector Review Exchange rates 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 $USD to 1.59 1.96 1.92 1.79 1.79 1.84 1.89 2.10 2.09 2.07 2.09 2.16 2.17 2.07 2.20 $FJD $FJD to 0.63 0.51 0.52 0.56 0.56 0.54 0.53 0.48 0.48 0.48 0.48 0.46 0.46 0.48 0.45 $USD Annex 3: Assessing Efficiency in Resource Use in Public Hospitals: Pabon-Lasso Approach With more than three-quarters of Ministry of the Health and Medical Services (MHMS) expenditure going to hospitals, the effectiveness with which public sector hospitals use scarce resources has implications for health sector outcomes in Fiji. However, assessing this effectiveness can be complex. Outcomes data, such as am- bulatory care sensitive conditions (ACSCs)158 or hospital readmission rates, were unavailable for Fiji. Reliance on simple one-dimensional measures of hospital outputs (such as the number of inpatient stays or bed occu- pancy rates) can be problematic due to differences in patient mix or strategic purposes (e.g., specialist hospi- tals, hospitals that prioritize disadvantaged or remotely located populations even if bed occupancy rates are low, or the need for surge capacity in beds), and because hospitals produce multiple outputs, such as inpatient and outpatient services. Traditional approaches to assessing hospital efficiency compare the hospitals that produce the highest ratios of output relative to inputs to those with lower ratios, to determine the magnitude of inefficiency in resource use. This can sometimes result in spurious identification of efficiency differentials. Therefore, this section uses a different method to analyze the functioning of public hospitals in Fiji. The Pabon-Lasso method for assessing hospital functioning combines three commonly used indicators to assess the likelihood of a hospital functioning less than optimally. Bed occupancy rates (BOR) in public hospitals can reflect hospital type (e.g., referral hospitals versus secondary hospitals), weak primary care services resulting in avoidable hospitalizations, availability of doctors/specialists, availability of private care, quality of care, etc. They only provide a partial picture unless they are very low (suggesting unused capital investments) or very high (which may point to problems with quality of care—overburdened staff, readmissions, and/or early discharge). The average length of stay (ALOS) of inpatients at a hospital refers to the time from admission to discharge. A high ALOS may result from the mix of patients (older and sicker), and type of hospital (e.g., rehabilitation facility), but also ‘social admissions’ (lengthy stays arising from informal providers seeking respite), delayed discharges reflecting weaknesses in community or transition care arrangements, or hospital- acquired infections (an indicator of healthcare quality). The bed turnover rate (or BTR) indicates the annual number of patients using an average hospital bed. Because it depends on the ratio of BOR to ALOS, the BTR helps to separate out cases where process improvements are driving increased bed use from those where rising BORs result from higher ALOS (due to hospital-acquired infections, for example). We use the Pabon Lasso159 classification that combines these three indicators to develop a typology of perfor- mance for hospitals. This typology classifies hospitals into four groups (Figure A3.1). The dividing (green and red) dotted lines are somewhat subjective MHMS hospitals while allowing for some variations in the boundar- ies of the regions depicted. 158. ACSCs, first described in 1993, are health conditions or diagnoses for which timely and effective outpatient care can reduce the risks of hospitalization by either (i) preventing the onset of an illness or condition; (ii) controlling an acute episodic illness or condition; or (iii) managing a chronic disease 159. Pabon Lasso 1986. ”Evaluating Hospital Performance Through Simultaneous Application of Several Indicators.” Bulletin of the Pan American Health Organization, 20: 341–57. Fiji Health Sector Review 119 Figure A3.1: Pabon-Lasso Typology of Hospital Performance Region II Region III Low Bed Occupancy • High Bed Occupance • High Bed Turnover • High Bed Turnover • Low ALOS • ALOS: not too high or low • Indicative of maternity • Considered well functioning • Bed Turnover Rate hospitals, inadequate care Region II Region III Low Bed Occupancy • High Bed Occupancy • Low Bed Turnover • Low Bed Turnover • ALOS: not too high or low • High ALOS • Regiona of concern - low • Indicative of social care • quality, or limited services functions; psychiatric hospitals; poor quality Bed Occupancy Rate (%) Source: Pabon Lasso. 1986. ”Evaluating Hospital Performance Through Simultaneous Application of Several Indicators.” The four regions of the Pabon-Lasso diagram can be used to make approximate inferences about the functioning of MHMS hospitals. Hospitals in Region I are likely to involve significant inefficiencies in resource use that need attention and are likely to be characterized by low quality of care or a limited range of services. Hospitals in Region II have short stays and a high bed turnover rate, which are characteristic of hospitals predominantly providing maternity services, but could also indicate excess bed capacity, admissions for ‘observation’, or unnecessary admissions reflecting problems with primary care. They may have limited specialist care and high transfer rates to divisional hospitals. Hospitals in Region III are in some respects functioning at the most effective level, with high bed turnover and occupancy rates and with lengths of stay are neither “too high” nor “too low”. Region IV includes hospitals that have high bed occupancy rates and long stays, but a low bed turnover rate. This is characteristic of specialized hospitals that provide rehabilitation services and psychiatric care, but also hospitals that treat more severely ill patients or chronic conditions. Inefficient resource use among hospitals in Region IV may result from a mix of poor quality of care (e.g., hospital-acquired infections) and social admissions. Examining three output indicators together—BOR, the bed turnover rate (BTR) which indicates the average number of patients per bed in any given year, and the average length of hospital stay of an inpatient (ALOS)— suggests that about six hospitals in Fiji could be considered well-functioning. The top-right quadrant location is typical of well-functioning hospitals. However, a large subset of hospitals is in the bottom-left quadrant, indicating low outputs (Figure A3.2). The three divisional hospitals are all in the top-right quadrant, the group of better performers. Of the other three hospitals in this group, two are in the Central Division (Navua and Wainibokasi) and one in the Western Division (Nadi). All the hospitals in the Eastern and Northern Division, except the divisional Labasa hospital, lie in the bottom-left quadrant indicating the worst performers. Unlike other countries with a significant private hospital sector, these findings for the Eastern and Northern division do not reflect a superior private sector option available to people living there. There are no private hospitals and few private practitioners in those divisions. Instead, the more likely explanation is some combination of input shortfalls e.g., the lack of adequate numbers of staff, drugs, and consumables, indicating a lack of capacity to deliver services of adequate quality, or excess beds (relative to population needs). 120 Fiji Health Sector Review Figure A3.2: Performance varies greatly between hospitals in Fiji. 2019 Source: Authors’ calculations based on MHMS 2019 data. Annex 4: The Potential Health Gains from Investing in Preventive Care and Health Promotion in Fiji How might Fiji benefit from investing more in health promotion, preventive services, and effective management of non-communicable conditions? In a resource constrained environment, the tendency is to focus on any cost savings that might result (e.g., due to reduced numbers of hospitalizations). However, given that a key health sector goal is to improve health, it is perhaps even more important to understand the magnitude of health improvements that might result from such investments. A popular approach to estimating the health impacts of investments is to rely on well known evaluation techniques that compare outcomes in areas where they are implemented to where they are not and tracking performance of ‘treatment’ and ‘control’ areas over several years. When interventions are relatively simple, such as conditional cash transfers for child nutrition, or the use of peer groups to address behavior at risk for acquiring non-communicable diseases (NCDs), evaluation techniques can be quite sophisticated, including relying upon experimental methods.160 Larger scale multi-dimensional interventions, such as those involved in the Primary Healthcare Performance Initiative (PHCPI) or those proposed in this review, are more difficult to assess in this manner. Complex interventions tend to be evaluated as before-after studies, qualitative assessments, or, in rarer cases, experimental and quasi-experimental studies. A famous example of a mix of these approaches is the long-running evaluation of interventions undertaken in North Karelia (Finland) to reduce the risk of cardiovascular disease in the local population.161 The challenges of evaluating the impact of interventions become even more formidable when, for reasons of need or other factors, implementation decisions about complex multi-dimensional programs must be taken in short order and estimates of potential returns to investment calculated. One useful way to assess potential returns in such cases is by comparing Fiji’s health outcomes with those of aspirational comparators that have implemented programs like those under contemplation in Fiji. For Fiji, we considered a peer country that has health outcomes that lie in the middle of the range for upper middle-income countries (UMICs). However, in 160. Fernald, Gertler, and Neufeld. 2008. “Role of Cash in Conditional Cash Transfer Programmes for Child Health, Growth and Development: An Analysis of Mexico’s Oportunidades.” The Lancet 371(9615): P828-37; Thankappan et al. 2018. “A Peer-support Lifestyle Intervention in Preventing Type 2 Diabetes in India: A cluster randomized controlled trial of the Kerala Diabetes Prevention Program” PLOS Medicine, 15:6. 161. See for example Vartiainen. 2018. “The North Karelia Project: Cardiovascular Disease Prevention in Finland.” Global Cardiology Science and Practice 2:13 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6062761/. Fiji Health Sector Review 121 doing so we caution that Fiji does differ from many other countries in its large indigenous population, so that the analysis that follows should be considered indicative rather than prescriptive. The Questions To assess the potential health impact of a substantial investment in health promotion, prevention, and primary care to address non-communicable conditions, we asked the following three questions: • How much would Fiji’s population gain in terms of improved health if it were to achieve health outcomes equivalent to those achieved by an average UMIC? • Does the speed at which such equivalence is achieved (i.e., priority given to the effective implementation of the proposed interventions) matter in terms of health outcome gains achieved, and by how much? • How feasible is it for Fiji to catch up with a typical UMIC by 2030, 2040, and 2050, respectively? Methods and Data We considered two types of outcomes: • Life expectancy at birth (an indicator that summarizes mortality risk at different ages). • Disability-adjusted life years (DALYs), an indicator that estimates health losses by combining the implications of disease for morbidity and mortality.162 Data on the variables necessary to construct the measures presented here were obtained from the Institute for Health Metrics and Evaluation (IHME) for Fiji and a collection of UMICs for 1990 to 2019 (2020 was excluded owing to the impacts of COVID-19). Population data for the same period (by age group) were obtained from the United Nations Population Statistics database. A country was chosen that reflected life expectancy at birth and DALYs lost per capita that reflected the mid-point of the range of health accomplishments for UMICs, for comparison with Fiji. Figure A4.1 compares differences in current and projected life expectancy at birth between Fiji and the UMIC under consideration. The data show that a Fijian born today is likely to live almost 9.1 years less than their counterparts in the peer UMIC, and moreover this gap is likely to increase to almost 10.5 years by 2050. 162. Murray, Lopez, and Jamison. 1994. “The Global Burden of Disease in 1990: Summary Results, Sensitivity Analyses and Future Directions.” Bulletin of the World Health Organization 72(3):495–509. 122 Fiji Health Sector Review Figure A4.1: Fiji’s life expectancy is forecast to lag a typical UMIC (2022-50) Source: Authors’ estimates using information from IHME, and (for life expectancy) compared to United Nations forecasts for validity checks. A similar situation can be observed for comparisons of DALYs lost that capture both mortality and morbidity (absence of perfect health) in the Fijian population, relative to a typical UMIC (Figure A4.2). Figure A4. 2: Fiji’s population has a much higher disease burden than a typical UMIC (2019) 3 2.5 DALYs Lost Per Capita 2 1.5 1 0.5 0 4 9 4 9 4 9 4 9 4 9 4 9 4 9 4 9 4 9 4 + 0- 5- -1 -1 -2 -2 -3 -3 -4 -4 -5 -5 -6 -6 -7 -7 -8 -8 -9 95 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 Age Groups Fiji UMIC Source: Authors’ estimates using information from IHME. Fiji Health Sector Review 123 Two sets of three scenarios were considered when assessing potential health gains to Fiji from investing in health promotion, preventive and primary care services to lower the burden of ill health. Under the first set of scenarios, we evaluated the health gains resulting from: • Scenario 1 (Challenging): Fiji catches up with the comparator UMIC in terms of DALYs lost per capita among people 30 years and older by 2030 (7 years). This is extremely challenging since the mortality gains involved are likely unprecedented. • Scenario 2 (Ambitious—achievable with immediate action): Fiji catches up with the comparator UMIC in terms of DALYs lost per capita among people 30 years and older by 2040 (17 years). • Scenario 3 (Moderate—achievable with moderate effort): Fiji catches up with the comparator UMIC in terms of DALYs lost per capita among people 30 years and older by 2050 (27 years). Because mortality reductions are a key component of health gains, we also considered another set of scenario comparisons, with a focus on mortality reduction (captured by life expectancy at birth) : • Scenario 1 (Challenging): Fiji catches up with the comparator UMIC in terms of life expectancy at birth by 2030 (7 years). This is extremely challenging since the mortality gains involved are likely unprecedented. • Scenario 2 (Ambitious—achievable with immediate action): Fiji catches up with the comparator UMIC in terms of life expectancy at birth by 2040 (17 years). • Scenario 3 (Moderate—achievable with moderate effort): Fiji catches up with the comparator UMIC in terms of life expectancy at birth by 2050 (27 years). For the analysis on disease burden, it is possible, indeed very likely, that people younger than 30 years would also benefit from these interventions. This suggests that any estimated health gains would constitute a lower bound to the full gains in terms of DALYs. It is also necessary to explain why this analysis considers DALYs lost from all causes and not just NCDs (which this report recommends as a priority concern). First, it is likely that the proposed interventions will affect morbidity and mortality from all causes. Crucially, however, even if the interventions were focused on NCDs alone, the analysis would have to account for competing disease risks when assessing health gains. Although the alternative strategy of focusing on health gains from NCDs (averted losses in DALYs from NCDs) has the advantage of being more directly associated with the proposed interventions, it would overestimate the returns to the investments recommended in this review. The notion of “catching up” must also be clarified. Both Fiji and the comparator UMIC will see some improvements in health outcomes over time (i.e., in life expectancy at birth or DALYs lost per capita by age-group) even under the status quo of no action. Therefore, the assumption is that “catching up” under Scenario 1, and for a given age group, means Fiji’s DALYs lost and mortality rates reach the same level as those of the comparator UMIC as they are forecast to be in 2030, when they are predicted to be lower than in 2023, due to underlying trends. Thereafter (i.e., after 2030), it will be assumed that the per capita DALYs lost in Fiji (or the age-specific death rate; ASDR) would progress at the trend rate for the UMIC comparator. The reduction in Fiji’s DALYs lost per capita (or ASDR) required to achieve parity with the comparator UMIC by 2030 is assumed to occur at a constant annual rate between 2023 and 2030. Similar adjustments were required to catch up with the comparator UMIC by 2040 (Scenario 2) and 2050 (Scenario 3). Note though that the annual reduction in the disease burden per capita (or age-specific death rate) that is required to achieve parity under Scenarios 2 and 3 is typically lower than that required under Scenario 1 as the catching up occurs over a long period. For Fiji and for the comparator UMIC, an initial trend rate was projected from 2023 to 2050 under the status quo for each age category over 30 years, in 5-year age intervals (with a top category of 95+ years). This trend rate of decline (mostly) was estimated using regression methods for data from IHME for the period 1990–2019. 124 Fiji Health Sector Review To assess health gains, DALYs lost under the status quo trend were compared to DALYs lost under the three scenarios to highlight their respective health gains. Life expectancy levels in Fiji were also compared to those of the UMIC under the three scenarios. Results Two sets of results are presented, one for savings in DALYs and the other for life expectancy gains, under each of the three scenarios. Savings in DALYs Lost from All Causes Under the Three Scenarios Figure A4.3 reports the findings for the gains in reduced disease burden (DALYs) for Fiji under the three scenarios. Figure A4.3: DALYS lost from all causes under the status quo and three scenarios to 2050 Source: World Bank estimates using data from the IHME and United Nations Population Statistics Division. Fiji Health Sector Review 125 These simulations indicate that were Fiji able to reach the level of DALYs lost per capita of the comparator UMIC in 2030 (the challenging scenario), it could benefit from enormous gains in lowered morbidity and mortality, with an almost 38.0 percent reduction in the disease burden relative to the status quo of no action. However, even if the progress towards attaining parity with the UMIC was slower, the gains are potentially still quite high. Catching up by 2040 (ambitious scenario) would mean a 30.5 percent reduction in DALYs lost compared to the status quo, while doing so by 2050 (moderate scenario) would mean a 21.8 percent reduction in DALYs. Life Expectancy Gains Under the Three Scenarios Figure A4.4 reports our findings for catching up measured as achieving parity in life expectancy at birth with the typical UMIC, under different scenarios. To achieve parity by 2030, Fiji would need to achieve extremely rapid increases in life expectancy at birth, with an estimated required gain of 10.5 years, which is unprecedented. For instance, Thailand took 27 years (1985–2012) to achieve a similar increase in life expectancy from levels equal to those of Fiji in 1985; and the Republic of Korea took 21 years (1983–2004) to achieve an equivalent increase from the level of Fiji in 1983. Figure A4.4: Life expectancy under the status quo and three scenarios to 2050 We believe that immediate investments are required for Fiji to be able to achieve somewhere close to parity with UMICs, even ones below the “typical” case considered here. Although Fiji has an unfinished agenda related to improvements in reproductive, maternal, neonatal, child, and adolescent health (RMNCAH), rapid reductions in infant and child mortality, even if they reached the levels observed in a UMIC by 2030), would yield at most 0.7-0.8 years in life expectancy gains. The greatest returns are likely to come from investments in non-communicable conditions. 126 Fiji Health Sector Review Annex 5: The Potential Economic Gains from Investing in Health Improvements in Fiji Fiji can also benefit economically from investing in health promotion, preventive services, and effective management of chronic non-communicable diseases (NCDs). Healthier workers are more economically productive and healthier children have superior cognitive skills and better educational outcomes. But it is not just the direct improvements in human capital that positively impact national output. Improvements in health are also associated with a window of economic opportunity (demographic dividend) in the form of an increased number of people in working age groups, and their share of the total population, reflecting a combination of longevity, and a decline in fertility.163 A large literature on economic growth has documented the role of gains in health in improving macroeconomic outcomes, including GDP, GDP per capita, and the rate of growth of real GDP per capita.164 Bloom and colleagues demonstrated that a 1-year increase in life expectancy at birth is associated with a 4 percent increase in real GDP.165 In their later analyses of the Chinese and Indian growth experience, Bloom and colleagues show that changes in the population shares of people of working ages (15-64 years) people and life expectancy (a proxy of past investments in health) were key drivers of the rate of growth of real GDP per capita, along with a range of other economic variables.166 However, they also acknowledge that, especially when increases in life expectancy arose from improvements in child health, the associated impacts could occur with a lag, reflecting (for instance) the time when these children grow up to become part of the workforce. As Fiji considers investments in health promotion, preventive care, and primary care, it is useful to ask: based on what we know about the implications of health and demographics for economic growth, what are the potential economic gains, if any, from such investments? Methods and Data For this analysis, we assessed the possible economic gains that would accrue to Fiji if it were able to achieve declines in mortality risks to the levels achieved by an upper middle-income country (UMIC) with a strong track record of investment in primary care and health promotion. Morbidity could not be directly captured owing to the reliance of most existing empirical models on measures of mortality risk (including the one used to derive the estimates for this analysis). However, it is well known that measures of life expectancy are highly correlated with disability-adjusted life years (DALYs), the indicator commonly used to capture morbidity and mortality in a population, in cross-country analyses. Fiji’s projected growth of real GDP per capita in a business-as-usual scenario between 2023 and 2050 was compared with the growth of real GDP per capita under two scenarios: • Scenario 1: Fiji’s age-specific mortality risks fall to equal the comparator UMIC by 2040 and remain equal to the comparator UMIC’s mortality risks thereafter. • Scenario 2: Fiji’s age-specific mortality risks fall to equal the comparator UMIC by 2050. Fiji’s life expectancy of approximately 67.9 years in 2019 (which is derived from age-specific mortality data) was well below that of the comparator UMIC (76.0 years in 2019) and expected to rise to almost 80 years by 2040, and 81.6 years by 2050 (United Nations Population Statistics Division). To achieve the comparator UMIC’s life expectancy by 2040 would require extremely rapid reductions in mortality risks among Fijians. For instance, Mongolia achieved gains in life expectancy of around 8.4 years between 2002 and 2019 (World 163. Bloom and Williamson. 1998. “Demographic Transitions and Economic Miracles in Emerging Asia” World Bank Economic Review 12(3):419-55. 164. Cervellati and Sunde. 2015. “The Economic and Demographic Transition, Mortality, and Comparative Development.” American Economic Journal: Macroeconomics 7(3):189-225; Lorentzen, McMillan, and Wacziarg. 2008. “Death and Development.” Journal of Economic Growth 13:81-124. 165. Bloom, Canning, and Sevilla. 2004. “The Effect of Health on Economic Growth: A Production Function Approach” World Development 32(1):1-13. 166. Bloom et al. 2010. “The Contribution of Population Health and Demographic Change to Economic Growth in China and India” Journal of Comparative Economics 38:17-33. Fiji Health Sector Review 127 Development Indicators database), and the Republic of Korea achieved life expectancy gains of 13.3 years (67.6 years to 80.9 years) over 27 years between 1985 and 2012. Importantly, Fiji’s life expectancy is well below what might be expected at its level of GDP per capita, suggesting that there may be an opportunity for particularly rapid mortality risk reductions from its current position. The Overall Approach Our assessment of the implications of health improvements for macroeconomic performance relies on the results of Bloom et al.167 who estimated the association between the 5-yearly rate of growth of real GDP per capita, and a wide range of explanatory variables, in a cross-country panel regression model. In addition to the standard explanatory models used in growth regressions (e.g., log of initial GDP per capita, ratio of investment to GDP, average years of school, etc.), the model also included life expectancy at birth at the beginning of each 5-year period, the log of the share of working-age population, and the growth of the share of working- age population as explanatory variables (Specifications 2 and 3 of Table 5 in Bloom et al.), and it is these that are of interest to us. We estimated the impact on the 5-yearly rate of real GDP growth under the two different scenarios, using coefficient estimates for these three variables from Bloom and colleagues, and projections for the three variables that we constructed for the two scenarios. It is possible that investments/improvements in health also work through other channels (variables) that influence GDP per capita but these were excluded from the analysis. Projecting Life Expectancy in Fiji and the Comparator UMIC in the Business- as-Usual Scenario: 2023 to 2050 Age-specific death rate (ASDR) data for 1990–2019 for Fiji and the comparator UMIC were obtained from the IHME for 5-year age intervals for ages above 5, and for 0-1 years and 1-4 years. Next, regression models of log-odds of ASDR with time as an explanatory variable for each age group were estimated for 1990–2019 and for 2001–19. Other models were also considered, including adding GDP per capita as an explanatory variable. Including GDP per capita also meant, however, that projecting ASDR to 2050 would require future projections of GDP per capita, adding a lot of complexity to the modelling. Instead, coefficients of the regression model with a trend term were used to project ASDR forward for Fiji (after suitable calibration). Information on projected ASDR was used to generate life expectancy at birth forecasts for Fiji for 2023 to 2050 under the business-as- usual scenario. A similar method was used to forecast ASDR and life expectancy at birth for the comparator UMIC from 2023 to 2050. Forecasts of life expectancy at birth under the business-as-usual scenario were constructed (i.e., the choice of coefficients from among the regression models referred to above) to be as close as possible to the medium scenario forecasts of the United Nations Population Division for 2050 for the two countries. In the case of the comparator UMIC, the model’s predictions matched the 2050 life expectancy at birth exactly. In Fiji, our projected life expectancy was 71.2 years in 2050, compared to UN forecasts of 71.7 years, but given the difference of 0.4 years in the baseline year (which is 2023 in our calculations), the change in life expectancy between 2023 and 2050 was very similar. 167. Bloom et al. 2010. op. cit. 128 Fiji Health Sector Review Projecting Life Expectancy for Fiji under Scenarios 1 and 2: 2023 to 2050 Next, we forecasted life expectancy for Fiji under the scenarios that it would catch up in ASDR with the comparator UMIC by 2040 (Scenario 1), and by 2050 (Scenario 2), respectively (we also considered 2030 as a catch-up year, but that would require an almost 10-year gain in life expectancy over 7 years, which appears challenging to achieve). For this purpose, the following assumptions were used for catching up by 2040 and 2050: • Scenario 1: For each age group, Fiji’s ASDR declines at a constant rate from its level in 2023 so as to reach the comparator UMIC’s ASDR by 2040, and then remains at the forecasted levels of ASDR for the UMIC until 2050. • Scenario 2: For each age group, Fiji’s ASDR declines at a constant rate from its level in 2023 so as to reach the comparator UMIC’s ASDR by 2050. Life expectancy at birth was calculated for each year for Fiji from 2023 to 2050 under projected ASDRs for Scenarios 1 and 2. Projecting Fiji’s Population under Business as Usual and Scenarios 1 and 2 An initial set of population projections was constructed for Fiji assuming zero net migration. This is obviously unrealistic for Fiji (where it is mostly negative, with estimates varying from -4 per 1,000 to -20 per 1,000 annually), so an adjustment was made to account for forecasted net migration as indicated in the next sub- section. Fiji’s population from 2023 to 2050 was forecasted with the help of projections on ASDR (under the business- as-usual scenario and Scenarios 1 and 2 above) and projections on the crude birth rate (live births per 1,000 population). The projections on the crude birth rate for Fiji were developed as follows: Step 1: a linear relationship was estimated from UN data on crude birth dates (dependent variable), the total fertility rate and trend (explanatory variables) between 1990–2019 for Fiji and the comparator UMIC. The fit was extremely good with the coefficient of determination exceeding 0.96. Step 2: UN projections on the total fertility rate between 2023 and 2050 (along with the time trend variable) were used along with the coefficient estimates from Step 1 to project crude birth rates for Fiji (under the business-as-usual scenario) and the comparator UMIC between 2023 and 2050. Step 3: We constructed two projected scenarios for crude birth rates in Fiji: one where crude birth rates in Fiji decline steadily to equal the comparator UMIC by 2040 (Scenario 1) and one where crude birth rates in Fiji decline steadily to equal the comparator UMIC’s by 2050 (Scenario 2). It is also possible to assume that crude birth rates remain at their current projected levels, but most of the literature suggests that mortality reductions are accompanied by fertility reductions, albeit with a time lag.168 Step 4: The crude birth rate projections were used to generate the number of (live) births in any given year. With this, and ASDR projections, it was possible to project Fiji’s population by age under the different scenarios: business as usual, Scenario 1, and Scenario 2. Accounting for Net Migration Because Fiji experiences high levels of outmigration (i.e., negative net migration), projecting population totals under a closed population assumption will result in upwardly biased population projections. To address this, we compared our business-as-usual population projections for Fiji between 2023 and 2050 to Fiji’s population projections by the UN Population Division (medium scenario). The resulting difference was regressed on time to derive a projected trend for net migration. This was subtracted from our population projections (under the closed population assumption) to obtain the final population projections under Scenarios 1 and 2. 168. See for example Lorentzen, McMillan, and Wacziarg. 2008. “Death and Development” Journal of Economic Growth 13:81-124. Fiji Health Sector Review 129 Assessing Growth Implications Bloom et al.169 estimated the following relationship in a cross-country panel regression model: The estimates of obtained by Bloom and colleagues in Table 5 (Columns 2 and 3) of their paper were used to generate the impacts of changes in life expectancy and the share of working-age population on Fiji’s economic growth in Scenarios 1 and 2. Because the effect of child health improvements on the future work force is likely to occur with a lag (after they have reached working ages), we assumed that any economic gains from improvements in life expectancy in the above equation occur after 10 years, so as to reflect a combination of gains in adult and child mortality assumed for Fiji. We also considered the case where any impact on economic growth from life expectancy gains only happened after 30 years (that is, not during 2023–50), the “long lag” scenario. Economic Growth and GDP Impacts Under Alternative Policy Scenarios Under Scenario 1 (which is essentially a very rapid reduction in mortality risks), we estimate that real GDP per capita would grow faster annually, relative to the business-as-usual scenario, by 0.03 percentage points by 2028, by 0.11 percentage points by 2038, and by 0.64 percentage points by 2048. If instead, Fiji achieved the outcomes envisaged in Scenario 2, real GDP per capita would grow 0.01 percentage points faster annually than the status quo by 2028, rising to 0.36 percentage points faster by 2048 (see Figure A5.1). The benefits in terms of economic growth would be smaller if life expectancy gains operated with a very long lag (say 30 years), shown in the figure as the “long lag impact”. Our GDP per capita growth impact projections do not directly capture the implications of changes in women’s labor force participation, an area that is of interest to countries in the region concerned about unlocking opportunities for economic growth. But they do so indirectly, through the health indicator (life expectancy at birth) since that is likely to capture the myriad channels through which health influences growth in the Bloom et al (2010) paper, such as reduced caregiving responsibilities. 169. Bloom. et al. 2010. “The Contribution of Population Health and Demographic Change to Economic Growth in China and India.” Journal of Comparative Economics 38:17-33. 130 Fiji Health Sector Review Figure A5.1: Annual rate of growth implications of health gains in Fiji, 2023–50 0.7 Percent point differnce from 0.6 business-as-usual 0.5 0.4 0.3 0.2 0.1 0 2028 2033 2043 2048 Year Scenario 1 Scenario 2 Long Lag Impact Our projections under the two scenarios can also be translated into average annual growth rates for the entire period from 2024 to 2050. Specially, if Fiji’s mortality outcomes were to equal that of the comparator UMIC by 2040, it would enjoy an average annual growth of real GDP per capita that is 0.30 percentage points higher than the business-as-usual scenario. If it was to catch up by 2050, then the average annual growth of real GDP per capita would be 0.15 percentage points. Annex 6: The Value of Longevity and Income Growth (Full Income) from Investments in Health in Fiji We are interested in the question: is it possible to provide a monetary value to improvements in mortality risks alongside income growth? In Fiji’s case, this is especially important since many of the income gains arising from health investments proposed in the report are not likely to occur until almost 20-25 years into the future. To get the monetary value of mortality risk reduction, we used the methodology of Becker et al.170 170. Becker, Philipson, and Soares. 2005. “The Quantity and Quality of Llife and the Evolution of World Inequality.” American Economic Review 95(1):277-91. Fiji Health Sector Review 131 Here the term outside the brackets on the right-hand side is the inverse of the elasticity of the instantaneous utility function with respect to income, evaluated at This expression highlights that the monetary value of longevity gains depends on gains in mortality reduction (the term in parentheses) and the intertemporal elasticity of substitution of consumption. The latter term captures the extent to which consumption can be substituted across time periods—or, put simply, can be used to support a longer life due to mortality risk reduction. To estimate E, we need to know the value of the expression on the right hand side of the above expression. Because we already know the age-specific death rates (ASDR) for Fiji under various scenarios, we can generate survival functions, and thus the term inside the brackets is relatively straightforward to calculate. This leaves the elasticity term. In fact, we need more information than just the elasticity parameter because the above expression is a simplification of the full formula (see below). To facilitate this, we need to work with a specific instantaneous utility function. Becker et al. (2005)171 recommend that: 171. Becker, Philipson, and Soares. 2005. op. cit. 132 Fiji Health Sector Review We project the (per capita) value of monetary health gains to Fiji’s under four scenarios: • Business-as-usual, projecting the value of GDP per capita and the survival function in 2023 with the value of GDP per capita and the survival function in all years until 2050. • Scenario 1, projecting the value of GDP per capita and the survival function in 2023. with the value of GDP per capita and the survival function in all years until 2050, when catching up with the comparator UMIC (see Annex 4) by 2040. • Scenario 2, projecting the value of GDP per capita and the survival function in 2023 with the value of GDP per capita and the survival function in all years until 2050, when catching up with the comparator UMIC by 2050. • Long lag scenario, projecting the value of GDP per capita and the survival function in 2023 with the value of GDP per capita and the survival function in all years until 2050, when catching up with the comparator UMIC by 2050, while assuming the effects of increased life expectancy only kick in after 2050. Results Figure A6.1 summarizes the main findings of this note on the value. Under the business-as-usual scenario, the the monetary value of the gains from lowered mortality in 2040 will be an income flow of FJ$585 (US$245) per capita in real terms (or about 5.2 percent of estimated income per capita in 2023). In contrast, under Scenario 1, the monetary value of gains from lower mortality are estimated to be FJ$4,735 (US$2,130), or about half of income per capita in 2023. Under Scenario 2 and the long lag scenario, the gains are slightly more than 20 percent of income per capita in 2040. By 2050, under all scenarios except business-as-usual, the monetary value of gains in mortality reduction are estimated to exceed 40 percent of the income per capita in 2023; under business-as-usual, the gains in mortality reduction are only 11 percent of the income per capita in 2023. Figure A6.1: The monetary value per capita of potential longevity gains in Fiji The Monetary Value Per Capita of Potential Longevity Gains in Fiji, 2040 and 2050 (2019 FJ$) 7000 6000 5000 4000 3000 2000 1000 0 Status Quo Long-Lag Scenario 2 Scenario 1 2040 2050 Source: Authors’ estimates. Fiji Health Sector Review 133 Annex 7: Returns from Investments in Infrastructure, Health Promotion, Prevention, Management, and Treatment of Non-Communicable Diseases (NCDs) The general approach used to assess the rate of return to investments in health promotion, prevention, treatment, and management of chronic conditions was the following: Then, the real rate of return (or the “internal rate of return”) is given by the r that solves the equation specified below. Note that multiple solutions are possible when there are irregularities in the stream of benefits and costs. However, that was not the case in this analysis. The rest of this note explains how the benefit and cost streams were calculated for the proposed health sector investments in Fiji. Assessment of Benefit Streams Benefits from investing in health essentially fall into three categories (i) health gains; (ii) gains in national economic output; and (iii) gains in efficiency in health spending (e.g., addressing gaps in key economic inputs such as human resources, drugs, or equipment so that health services are delivered at lower average and marginal costs; improved ways of providing health services, such as through an expanded role for community health workers (CHWs), or screening that helps in early identification and thus lower costs of interventions/ outpatient visits/hospital stays, etc.). We first focus on monetary values of benefit streams from the first two categories, since any efficiency gains will ultimately translate into improved health outcomes, provided overall expenditures are unchanged. Improved health contributes to better national economic outcomes because healthier workers tend to be more productive and are more likely to participate in productive work. Declines in mortality rates also contribute to improved national economic outcomes by increasing the numbers of people of working age, if they can be gainfully employed. Declines in mortality (whether at younger or older ages) also pave the way for declines in fertility; these also raises the share of people of working age in the economy which, in turn, creates opportunities for increased income per capita (this is sometimes referred to as the “demographic dividend”). There are other implications of declining mortality that also contribute to national economic growth, such as increased investments in education, and increased savings that also promote growth in real income per capita. In addition, improvements in health are valuable in and of themselves. To translate health gains into monetary gains, economists typically rely on the “compensating variation” method which essentially involves answering the following question: how much extra would an individual have needed in income (money) to forgo the health gains that have enjoyed? (This was the subject of Annex 6). 134 Fiji Health Sector Review Separate technical notes prepared for this review (included in the annexes to the review) estimate both the growth in real income per capita, as well as the value of potential gains in health that could result from health sector interventions proposed in this review. However, note that the monetary value of any gains attributable to the investments is the difference between the gains arising from the interventions and the gains under the status quo of no intervention. The gains were assessed under two scenarios that have been the focus of this review: • The proposed investments enable Fiji to catch up to the health outcomes of an average UMIC by 2040. • The proposed investments enable Fiji to catch up to the health outcomes of an average UMIC by 2050. Considering real income per capita alone, the difference between the first scenario, where Fiji catches up with the health outcomes of an average UMIC by 2040, and the no-intervention (or business-as-usual) scenario was approximately FJ$1,769 (US$796) per person by the last year of the comparison (2050). The difference between the second scenario, where Fiji catches up with the health outcomes of an average UMIC by 2050, and the no-intervention scenario was approximately FJ$959 (US$431) per person by the last year of comparison (2050). With a population of slightly above 1.1 million projected for 2050 under both the intervention scenarios, these amount to gains of between FJ$0.96 billion (US$432 million) and FJ$1.77 billion (US$796.5 million) per year by 2050, with most of the gains concentrated in later years. If we also consider the value of gains in longevity, the added benefits are considerably larger in 2050, about FJ$4,735 (US$2,131 per person) under the first scenario (relative to no-intervention). In contrast, the gains are slightly lower, or about FJ$4,516 (US$1,870) per person in 2050 under the second scenario (relative to no- intervention). Note that for longevity, the monetary value of gains is not added for individual years but is taken as the cumulative gain from 2024 to 2050 and will (effectively) be discounted when calculating the internal rates of return for the proposed investments. Assessment of Cost Streams Baseline (No Intervention) Utilization of Health Services per Capita in 2023, and Unit Costs of Health Services in Fiji in 2023 As a first step, the project envisages investments in infrastructure (health centers, nursing stations, hospitals), training, equipment, staff costs (especially community health workers), and related costs of various health promotion programs. These costs are substantial, and we estimate these to be at least about FJ$327 million (US$144 million) over the next 10 years, excluding investments in Colonial War Memorial hospital (CWM). However, these are not likely to be the most significant cost-element arising from the proposed investments. We believe that the most significant costs of the intervention are likely to come from increased use of health services (e.g., outpatient visits and hospitalizations as people with previously undiagnosed conditions are identified). This required estimating the added/differential utilization (relative to the no-intervention status quo) under the two scenarios outlined above, and the costs associated with the differential use of health services. To assess what these costs of extra utilization resulting from the investments might look like, we compared Fiji’s inpatient and outpatient utilization patterns with an average UMIC (note that health services are not only outpatient and inpatient services, but may involve diagnostics, screening, etc. However, these were tacked on to the costs of inpatient and outpatient care as outlined below). Because our post-investment scenarios can be expected to have important effects on the age distribution of the population (and age is known to be associated with healthcare use), we looked for age-specific utilization statistics for inpatient and outpatient care. For Fiji these data were not readily available. We did locate a study, Fiji Health Sector Review 135 however, that reported data on total inpatient stays/discharges for people over 55 years (and people under 55) over a 4-year period (2014–17).172 The study also provided information on outpatient visits in hospitals for one year for people over 55 years and people under 55 years. Data from this study show that roughly 21 percent of all inpatient discharges in Fiji hospitals were accounted for by people aged 55 years and over. This age group also accounted for 25 percent of hospital outpatient visits. Data on health centers were only available from two facilities, Nuffield and Nausori, for two months in 2023 for the former, and for the full year of 2010 for the latter.173 These visits were converted into annualized numbers and information on their estimated catchment populations was used to arrive at per capita health center visits in the Fijian population. This per capita estimate was then used to estimate aggregate outpatient visits to health centers throughout Fiji, and the share of these visits was allocated to people over and under-55 based on the shares described above for Fijian hospitals. However, outpatient visits/home visits also occur in the context of services by nursing stations. Data on the quantity of services provided are not available for nursing stations. However, some idea of the number of healthcare visits can be obtained from research undertaken at various Australian and New Zealand universities, which mention that nursing staff at the stations have busy days and may be seeing as many as 30-40 patients or more per day.174 Assuming a full day of work and 15 minutes per patient, we arrive at 30 visits per day, working for 22 days a month and 12 months in a year (or about 7,920 visits per nurse). Since nursing stations are staffed by one nurse, this was the number of visits assumed per nursing station. Assuming roughly 100 nursing stations yielded an estimate of 792,000 nursing station outpatient visits/home visits per year throughout Fiji (and these were divided into people over 55 and under-55 based on the shares described above). Collectively, these numbers (hospitals, health centers, and nursing stations) yielded about 3.75 outpatient visits per capita across all facilities (we assumed that private outpatient visits are negligible), with higher numbers for people over 55, and lower for those below 55. Estimates of inpatient stays (discharges) for 2019 are available from the Ministry of Health and Medical Services (MHMS) using 2019 as the base since COVID-19 likely impacted healthcare use in ways that may not accurately capture future trends. These amount to about 0.08 hospital discharges per capita (about 0.074 for people less than 55, and 0.117 for people over 55). Unit Costs of Health Services There was very limited costing data available for Fiji. We could find only one study from 2010 – that had some information for three facilities: two hospitals and one well-equipped health center.175 For our analysis we adjusted unit cost data (outpatient visits and inpatient stays at hospitals; and outpatient visits at the health center) for inflation between 2010 and 2023 to get to a 2023 estimate. This yielded the following unit costs for 2023: FJ$70 (US$32) per hospital outpatient visit; and FJ$31 (US$14) for a health center outpatient visit. The cost per hospital discharge was estimated as FJ$755 (US$340). We assumed a cost of FJ$7 (US$3) per visit to a nursing station. This was assessed by dividing the daily equivalent of a nurse’s salary + allowances and dividing it by 30 and adding the cost of drugs, utilities, travel and other incidentals, and overheads (assumed to be the same as the cost of nurse time). 172. Palagyi et al. 2022. Health System Responses to Population Ageing in Fiji: Identifying Policy Program and Service Priorities. Sydney: George Institute for Global Health. 173. Irava, Pellny, and Khan. 2012. Costing Study of Selected Health Facilities in Fiji. Suva: Fiji Ministry of Health; MHMS. 2024. Central Health Services: Sub- divisional monthly reports (November and December 2023). Suva: MHMS. 174. Nawaqaliva. 2022. Exploring Job Satisfaction of Nurses in Fiji. Doctoral dissertation. Auckland: Auckland University of Technology. 175. Irava et al. 2012. op.cit. 136 Fiji Health Sector Review Estimating Healthcare Use and Spending under Alternative Scenarios Status quo (no intervention) scenario: We used UN (United Nations) population forecasts to project Fiji’s population by age (over- and under-55 years only) and used the per capita utilization of inpatient and outpatient care by the respective age groups to project future utilization of health services (outpatient visits to nursing stations, health centers, and hospitals) and inpatient discharges (hospitals). This was combined with the estimates of unit cost for 2023 to predict health expenditures on inpatient and outpatient care in 2023 prices. Scenario 1: Investment scenario where Fiji achieves equivalence in health outcomes by 2040: We used our own population projections (by age) that have been discussed in other annexes. We assumed that per capita healthcare use rates for inpatient and outpatient care (by the two age groups of 55 and over, and below 55) would be the same as for Malaysia (the only other UMIC country in the region for which age-specific utilization data were available for a recent year and by age). We assumed that the Fiji would increase its healthcare utilization rates steadily to attain the Malaysian per capita health service utilization rates by 2033. Obviously, any such change in utilization patterns may take even longer but estimating utilization in this manner will result in a conservative estimate of the rate of return (by effectively assuming higher costs at an earlier period). Unit costs of health services were assumed to be the same as above and helped yield the time profile of healthcare costs for inpatient and outpatient care under this scenario. However, we believe that better screening and early identification of health conditions may result in less costly (early) interventions, so one might expect inpatient and outpatient visits at hospitals to be somewhat lower and to be less costly on average. The data for Malaysia show that the average number of outpatient visits per Malaysian under 55 years is approximately 4.0 per year, compared to 6.2 visits per year for those over 55. Hospital discharges are 0.11 for people under 55, and 0.22 over those aged over 55 years (Institute for Health Systems Research 2013). Scenario 2: Investment Scenario where Fiji achieves equivalence in health outcomes by 2050: We used our own population projections (by age) that have been discussed in other annexes. We assumed that per capita healthcare use rates for inpatient and outpatient care (by the 2 age-groups 55 and over, and below-55) would be the same as for Malaysia (the only other UMIC country in the region for which age-specific utilization data were available for a recent year and by age). We assumed that the Fiji would increase its healthcare utilization rates steadily to attain the Malaysian per capita health service utilization rates by 2040. A change in utilization patterns to reach Malaysian levels may take even longer, as discussed above. Unit costs of health services were assumed to be the same as above and helped yield the time profile of healthcare costs for inpatient and outpatient care under this scenario. However, we suspect that better screening and early identification of health conditions may result in less costly interventions, so one might expect inpatient and outpatient visits at hospitals to be somewhat lower. Calibrating our Health Expenditure Estimates to Aggregate Projections under the Review Under the status quo projections our estimated spending on inpatient and outpatient care came out to be less than half of our projected spending. Because it is likely that other services (lab tests, X-rays, etc.) are conducted for patients and not fully accounted for, and because the unit costs may themselves have been underestimates, we calibrated our projected inpatient and outpatient expenditure spending to equal to the projections for health expenditure undertaken as part of the review. Interestingly, the “inflationary factor” did not vary much from year to year. The same inflationary factor was used to scale up the inpatient and outpatient expenditure projections under Scenarios 1 and 2, although in practice we would expect the inflationary factor to be somewhat lower as health services are likely to cheaper on a per unit basis as conditions are identified earlier and treated early and at lower cost. Fiji Health Sector Review 137 Estimated Rates of Return If Fiji were to catch up with a UMIC’s health outcomes by 2040: • Benefits measured in gains to income per capita: annual rate of return: 7.6 percent. • Benefits measured in gains in income per capita + gains in longevity: annual rate of return: 10.2 percent. If Fiji were to catch up with a UMIC’s health outcomes by 2050: • Benefits measured in gains to income per capita: annual rate of return: 3.6 percent. • Benefits measured in gains in income per capita + gains in longevity: annual rate of return: 8.2 percent. Annex 8: Cost of Recommended Actions: Methods and their Rationale Introduction The policy suggestions and programs emerging out of the Health Sector Review for Fiji fall into two broad areas: • Supply side interventions • Demand side interventions On the supply side, the proposed policy suggestions and programs focus on (i) strengthening primary care services for the prevention, and the diagnosis, management, and treatment of chronic conditions while maintaining other services (e.g., maternal and child health); (ii) outreach programs in school and communities; (iii) investments in community health workers (CHWs); and (iv) investments that can strengthen hospitals with low utilization rates to achieve improved population health outcomes. Costs on the supply side: • (S.1) Upgrading infrastructure and equipment in selected health centers and nursing stations. • (S.2) Strengthening of digital infrastructure. • (S.3) Filling gaps in human resources for health (community health workers, ancillary health workforce, training). • (S.4) Interventions to change behavior of healthcare workers (incentives for health promotion, diagnosis, and management of chronic conditions, and effective referrals). • (S.5) Interventions to incentivize private providers to enhance their efforts related to screening, treatment, and management of chronic conditions. Costs on the demand side: • (D.1) Conditional cash transfers for patients that promote lifestyle changes, annual health check-ups, and better self-management of chronic conditions. • (D.2) Peer-based efforts to promote lifestyle changes. • (D.3) School health programs and parental outreach via these programs. • (D.4) Incentivizing consumers/schools/business to promote healthy behavior and treatment/management of chronic conditions (e.g., traffic-light food labelling). • (D.5) Health taxes. 138 Fiji Health Sector Review Cost of Supply Side Interventions: (S.1) Infrastructure and Equipment Upgradation of Health Centers and Nursing Stations Fiji has 81 health centers spread across four divisions (Central, Western, Northern, and Eastern). The catchment population served by these health centers varies significantly across regions. Figure A8.1 shows that the population in the catchment area of each health center is significantly smaller in the Eastern region, compared to the others. Figure A8.1: Health centers in Fiji, number and average catchment population (2023) Health Centre in Fiji, 2023: Number and Average Catchment Population 18000 30 16000 Population per Health Centre 25 14000 Number of Health Centres 12000 20 10000 15 8000 6000 10 4000 5 2000 0 0 Central Western Northern Eastern Division Number POP/Facility Source: Fiji Bureau of Statistics, with population adjusted for growth since the 2017 census; data on number of health centers from MHMS. MHMS assessments show that most of the infrastructure requirements at health centers related to (i) electrical and plumbing work; (ii) generators and/or solar panels; (iii) equipment and infrastructure maintenance; and (iv) signboards. While the Ministry of Health and Medical Services (MHMS) identified some health centers as needing major works, such as being relocated, or being “converted from old” to new, these were relatively few, with seven facilities designated as needing relocation (four in Central Division, one in Northern Division, and two in Western Division) (Figure A8.2). Fiji Health Sector Review 139 Figure A8.2: Infrastructure support needs at health centers in Fiji (2024) 90.0 80.0 70.0 Share Facilities in Need 60.0 50.0 40.0 30.0 20.0 10.0 0.0 Central Western Northern Eastern Division Maintain Elec/Plumb Gen/Solar Source: MHMS. Note: for generators/solar energy, the higher number was chosen. Infrastructure and equipment investments to upgrade these facilities will obviously vary by the size of the facility. For instance, the size of the catchment population per health center in each division suggests that the typical health center in the Western and Central Divisions will be larger than their counterparts in the Northern and Eastern Divisions, especially for some health centers located in or near major urban areas. Available estimates of the cost of construction (or re-construction) suggest this. For instance, after accounting for inflation, the cost of constructing and equipping the Nakasi Health Center in Central Division (FJ$6.7 million in 2018) would be roughly FJ$8.3 million in 2024.176 In comparison, Koro Health Center was estimated to cost FJ$634,000 in 2018; accounting for inflation it could be expected to cost FJ$800,000 in 2024.177 The cost of reconstructing the Qamea Health Center on Qamea Island (in the Taveuni Subdivision)—estimated at FJ$1.5 million in 2017, or about FJ$2.0 million in 2024—is consistent with the observation that its catchment population is larger than a typical health center in the Eastern Division, and much smaller than a corresponding health center in the Western and Central Divisions. Costs of Upgrading Health Centers How much might an upgrading of existing health centers cost in Fiji? A recent estimate by the United Nations Office for Project Services (UNOPS)178 for three health centers serving large catchment populations in the Central Division for infrastructure and equipment upgrades suggests estimated renovation costs of approximately: • Nuffield Health Center: US$2.355 million (approximately FJ$5.3 million). • Raiwaqa Health Center: US$1.533 million (approximately FJ$3.48 million). • Valelevu Health Center: US$4.36 million (approximately FJ$9.90 million). The UNOPS report shows that the catchment population served by these health facilities is considerably 176. China.org. 2018. “New Health Facility to Raise Bar in Medical Treatment in Fiji”. http://www.china.org.cn/world/Off_the_Wire/2018-11/01/ content_69508228.htm. 177. Fiji Times. 2018. “Newly Upgraded Health Centre Opens on Koro.” https://www.fijitimes.com.fj/newly-upgraded-health-centre-opens-on-koro/. 178. UNOPS. 2024. Assessment and Scoping Mission of Three Primary Healthcare Centers in Suva, Fiji for the World Bank. 140 Fiji Health Sector Review larger than a “typical” health center in Fiji (35,000 for Raiwaqa, 56,000 for Nuffield), and therefore the cost of upgrading for a typical health center in Fiji is likely to be lower than estimates for the three health centers the report was concerned with. Given that the infrastructure needs (at least as specified in MHMS assessments) are broadly similar across health centers, resource requirements for upgrading can be projected as a percentage of the construction cost of a new facility. Here, the expenditure incurred to construct/equip the Nakasi health center offers a natural benchmark. Using the UNOPS Nuffield estimates to indicate the renovation costs involved, this would suggest renovation costs equivalent to 66 percent of the cost of a totally new facility. If the Raiwaqa estimates were used instead, renovation costs would be 42 percent of a totally new facility. These can be used to generate a range of estimates for the cost of upgradation of infrastructure and equipment across Fiji. Specifically, if all facilities had to be renovated, the upgrading costs would be: • Eastern Division (14 centers): FJ$4.7–5.3 million. • Northern Division (20 centers): FJ$6.8–26.6 million. • Western/Central (47 centers): FJ$163.1–251.3 million (or FJ$94 million using an alterntive method; see below). The above estimates notwithstanding, our preferred strategy for estimating scaled-up costs of upgradation in the Central and Western Divisions (which we think is more appropriate for an average health center in these two divisions, given that the three health centers we are considering are relatively large even in within this group), is to rely on the estimates of renovation costs for another health center, Nausori Health Center, at roughly FJ$2 million.179 If this number is used as a proxy for upgrading the facilities in the Western and Central Divisions, the costs would be FJ$94 million for all 47 facilities in these two divisions. Upgrading the Infrastructure and Equipment of Nursing Stations We classified upgrading into two main categories: major works (relocation, conversion from old to new, and upgrading) and “other upgradation”. Almost half of the nursing stations were categorized as needing major works in the Central and Western Divisions, and 12 percent of nursing stations required major works in the Northern and Eastern Divisions. Together this amounts to slightly above one-quarter of the total number of nursing stations. Several nursing stations were identified as needing electricity or plumbing works, generators or solar energy, and maintenance works. We assumed that the cost of major works would be equivalent to two-thirds of the cost of construction of a new facility, and the rest would require about 40 percent of the cost. The cost of constructing and equipping a new nursing station was estimated to be FJ$500,000 and assumed to be the same across all divisions. With these assumptions, the estimated cost of upgrading nursing stations was estimated to be FJ$9.63 million in the Central and Western Division, FJ$4.05 million in the Northern Division, and FJ$4.83 million in the Eastern Division. • Total cost of upgrading nursing stations: FJ$18.51 million. Cost of Screening Strategy for Diabetes and Hypertension We assumed that screening will be undertaken annually as follows. Community health workers will administer a questionnaire to all individuals aged 30 years and over to measure their diabetes risk score and will measure their blood pressure (BP). The share of population aged 30 years and over in Fiji is 56 percent (approximately 510,000). The International Diabetes Federation estimates the share of people of people aged 20–79 years 179. FBC News. 2023. “Peters Tours Nausouri Health Center’s $2m Renovation”. https://www.fbcnews.com.fj/news/peters-tours-nausori-health-centers- 2m-renovation/. Fiji Health Sector Review 141 with diabetes in Fiji to be 17.7 percent.180 This prevalence rate is equivalent to about 90,250 people with diabetes in this age group. The prevalence data also suggest an additional 72,250 people with impaired (fasting) glucose tolerance (IGT). We assume that the number of people with a diabetes risk score above the risk threshold is twice the number of people with diabetes and IGT (i.e., 325,000). For screening purposes, it is also assumed that these individuals would be administered a rapid diagnostic test for diabetes (random glucose test). Based on a study by Kaur et al. for India,181 and accounting for inflation, the cost of administering these tests is approximately US$0.50 per person, leading to a total cost of US$162,500. We also assume that there is no measurement error (sensitivity and specificity) associated with these tests. Thus, exactly 90,250 people will be identified with diabetes, subject to a confirmation test (assumed to be HbA1c). The cost of an HbA1c test was estimated to be USD2.50 per person, or an estimated US$207,500. In total, this translates into US$370,000 in the first year for both screening tests combined, plus an additional small amount for the development and administration of the risk-score questionnaire (which we ignore in our calculations with the major cost being CHW time which is already accounted for under a separate cost category). In future years, screening costs may go up somewhat as individuals diagnosed with diabetes will require tests at more frequent intervals, and new individuals will be identified. Estimates based on Kaur et al. suggest a cost of BP measurement of approximately US$1.30 per test (accounting for inflation).180 These costs include the costs of human resource time. Because we have accounted for CHW costs elsewhere, we assume the costs of measurement to be one-third this, i.e., US$0.43 per measurement (accounting for travel, etc.). Because all individuals over 30 years of age are assumed to be tested (slightly over 500,000), this will cost US$220,000 annually. As in the case of diabetes, we would expect these costs to increase over time. Overall, we estimate the costs of screening for the entire Fijian population aged 30 years and over for diabetes and hypertension to be US$590,000 in the initial year, rising in future years for the reasons outlined above. This should be budgeted for something closer to US$1 million annually in future years. (S.2) Cost of Strengthening Digital Infrastructure Table A8.1 covers the estimated costs of implementing major elements of the proposed digital health strategy, focusing on the non-infrastructure elements (MHMS Digital Health Strategy 2023–27). We have also added the cost of harmonizing health-related data sources and upgrades to interoperable digital systems, divisional and subdivisional workforce capacity for clinical coding, and data entry, disaggregation, management, quality assurance, and cloud storage. 180. International Diabetes Federation Diabetes Atlas – Fiji. https://diabetesatlas.org/data/en/country/69/fj.html. 181. Kaur et al. (2022). “Cost-effectiveness of Population-based Screening for Diabetes and Hypertension in India: An Economic Modelling Study.” Lancet Public Health 7: e65-e73. 142 Fiji Health Sector Review Table A8.1: Costs of Implementing the Digital Health Strategy (2023–27) 133 facilities are yet to be connected to the internet: 33 health centers, 98 nursing stations, the Military Hospital, and Naiserelagi Maternity Satellite/ structure cabing internet Starlink ~US$200,000 (standard) or “US$460,000 (high performance) Intial investment * US$1,500 per facility (standard kits) or ~US$600,000 (excludes hardware like laptops, com- US$3,500 (high performance kits). *US$4,500 per facility pute, monitors etc) * excludes shipping and any other instal- lation cost of fixing on roof, extra cabling etc. ~US$320,000-400,000 annually ~US$570,000 annually *US$355 per US$200-250 per facility per month for the Ongoing connection facility per month (include telephone and global connection (likely necessary for (for an additional 133 facilities) internet) Fiji because Starlink connection would be from AUS or NZ). Total Cost FJ$2.8m FJ$1.6m Lower estimate Upper estimate Component Recurrent (FJ$) (FJ$) (FJ$) Maintenance and licensing of existing systems $600,000 PATISPlus replacement, patient administration If open source, then no fixed recurrent $5 million $12 million systems and EMR all sites cost Reporting system replacement of CMRIS with If open source, then no fixed recurrent $1.5 million $2.5 million full deployment and GIS interaction cost System integration Including with Aspen PPP $320,000 $380,000 $70,000 If open source, then no fixed recurrent Immunisation registry $240,000 $340,000 cost Business analysis and system design $190,000 $270,000 Organisational and other analysis $170,000 $220,000 Architecture and standards $350,000 $450,000 Fiji Health Sector Review 143 (S.3) Costs of Expanding CHWs’ Roles into Health Promotion and the Effective Diagnosis, Management, and Referral of Chronic Conditions There are an estimated 1,640 CHWs in Fiji. CHWs constitute a key component of the frontline HRH that link its population to health services, promote health and wellness in the community, and help manage the health issues faced by the community, including maternal and child health conditions and NCDs, and administer first aid. The MHMS has provided CHW training since the 1980s.182 The 2013 training manual for CHWs in Fiji identifies the following core competencies to support child health, safe motherhood, and wellness: • Engage the community to produce healthy outcomes. • Implement health promotion programs. • Provide emergency first-aid response. Training components include skills to engage effectively with the community, understanding and implementing community health promotion strategies. These include accessing and providing IEC materials, how to refer sick individuals to health services, promoting healthy lifestyles, identifying and managing NCDs, environmental health, and promoting maternal and child health (family planning, antenatal visits, transportation). It is unclear how effective these training programs have been in developing CHW competencies in Fiji, or whether the CHWs have been able to apply these competencies. Yoon et al.183 provide some insights about CHW working conditions based on a 2016 survey of CHWs. The authors noted that CHW numbers reached a peak of around 3,000 in the 1990s but there has been considerable attrition since then. As of end-2015 there were 1,805 registered CHWs, and clearly there has been further attrition since then (we estimate attrition rates at 1.1 percent per year). The survey found that: • Almost one-third of the CHWs were over 50 years of age (mean age 45 years). • Almost 87 percent were recruited by the community, and about 11 percent by the health centers. • CHWS worked for about 10-12 days/month with a median work duration of 2-2.5 hours/day. • 80 percent had a commute to the nearest health facility of 60 minutes or less. • Most CHWs identified their core activities as the ones identified in the core competencies manual. With respect to training: • The vast majority (>95 percent) of CHW respondents reported attending at least one training program in the previous 12 months, with a mean of three per respondent. • The respondents requested longer and better quality of training, suggesting that existing training programs may need strengthening, and underlined the need for refresher training. • Many felt that travel costs for training needed to be covered. 182. Yoon et al. 2019. “Working Conditions and Perceptions on Incentives of Fiji Community Health Workers:Findings from a Survey with a Convenience Sample in 2016.” Journal of Global Health Science 1(1) : e8. 183. Yoon et al. 2019. op. cit. 144 Fiji Health Sector Review Costs of Additional CHW-related Investments in Fiji’s Health Sector There is a large literature on the contribution of CHWs to addressing non-communicable diseases and in promoting health more generally.184 In Fiji’s case, where nurses are either moving to the private sector or migrating, the role of CHWs will be crucial in addressing health services gaps, especially regarding NCDs, through early identification, prevention, and effective management. From the standpoint of additional costs, there are four elements to consider: • Are there the right number of CHWs? • Is their current pay adequate? • Will there be additional costs of managing them? • Will there be any equipment requirements? Are existing numbers sufficient or are additional CHWs needed? The current number of CHWs is about 1 CHW per 550 population, which is relatively high (In Bangladesh, the ratio is about 1,320, and in India it is 1 per 1,400 for instance). An alternative strategy is to consider how many CHWs are likely to be required given Fiji’s needs related to NCDs. One recent paper185 proposes that CHWs should aim to visit households 2-3 times a year to check blood pressure, administer risk-score questionnaires etc. to identify new cases, and for those already identified, assist in relevant referrals and management. The expected time in any given household would be around 40 minutes, suggesting about nine households can be covered per day. With Fiji’s population of about 920,000 and household size of 4.1, this amounts to about 220,000 households. If these households are visited 2.5 times per year on average, we would need about 278 CHWs working full time on just the NCD interventions. According to Yoon et al.186 the average time a CHW works is about 3 hrs/day for about 10–12 days, or about 20-25 percent of a full-time employee on average. Even assuming that some of their work time is spent on activities unlinked to NCDs, this translates into an “excess capacity” of 1,238 CHWs. If we take the position that CHWs are working longer hours than reported by Yoon et al., and that other NCD-related programs (e.g., palliative care, respiratory conditions, etc.) also require their attention, the current CHW numbers in Fiji appear to be adequate. However, if the data are incorrect, and all the current CHWs are overworked, then the above calculations suggest that at most 300 new CHWs will be needed. Additional costs may be incurred if there is attrition due to retirement, since in that case onboarding (presumably additional training costs) will be needed. This last scenario will not be considered in the cost estimates. Will the CHWs need to be paid more than they are currently being paid? If so, by how much more? In the survey by Yoon and colleagues, the CHWs requested incentive payments of around FJ$100/month. However, part of the reason for attrition since 2015 may be their low economic returns from working. Currently, it is assessed that CHWs receive around FJ$200/month. If the government decided to supplement this income so that the CHWs received the current minimum wage of FJ$4/hour, a full-time CHW would earn FJ$650/ month. If this were to rise to FJ$6/hour, a full-time CHW would earn FJ$975/month. To the extent that the government may wish to limit its fiscal liability (and seeks to ensure that CHWs are 184. e.g. Jafar et al. 2020. “A Community-based Intervention for Managing Hypertension in Rural South Asia.” The New England Journal of Medicine 382 (8): 717-26. 185. Finkelstein et al 2021. op.cit 186. Finkelstein et al. 2021 op cit. Fiji Health Sector Review 145 viewed as belonging to the community and not government staff), a working assumption could be that they are paid at 75 percent of the proposed full-time employee salary (i.e., between FJ$340 and FJ$730 per month, depending on the minimum wage). Some/most of this extra payment could also be justified as covering travel costs for training and for providing an expanded range of services. That is, compared to their current compensation of FJ$200 per month, they could get an extra amount ranging from FJ$140 to FJ$530 monthly. This would imply an extra expenditure of FJ$2.77–10.49 million, with a midpoint of FJ$6.63 million annually. Additional training costs above the levels that the government currently incurs Given that CHWs are already undergoing three rounds of training per year (on average) according to the Yoon et al. study, it is unclear whether more is needed. Perhaps training content needs to be different, focusing more intensely on NCD-related health promotion, screening (BP measurement, blood sugar measurement, weight, and height assessments, etc,), and the management of NCDs (facilitating drug supply, adherence messages, tracking control of BP, blood sugar, etc.) One estimate of such training costs is available from a study in Bangladesh, Sri Lanka, and Pakistan, which assessed the added NCD-related costs to be approximately US$181–365 per CHW in the first year of training (Finkelstein et al. 2021).187 Taking this range of numbers as a guide, we end up with US$300,000–US$600,000 for one round of training, or US$0.9–1.8 million for three rounds of training in the first year that could cover a range of other topics. Based on a comparison of MHMS data from Fiji, where just the “training allowances” are estimated to be FJ$3 million (US$1.32 million), the last set of numbers appear to be more realistic. Studies have also assessed the annual costs of (refresher) training to be roughly half of the initial year training costs in subsequent years, or about US$450,000-US$900,000. To this we can add a small amount for the costs of new training (1.1 percent) to cover the costs of attrition. Overall, therefore, the training costs for all 1,650 workers are likely to be: • FJ$2.1 million–4.1 million plus travel allowances in the first year. • FJ$1.1 million–2.1 million plus travel allowances annually in the years following. Note that although allowances have already been accounted for in the training costs calculated in the cited studies, these could be higher in Fiji given that it has a higher income per capita and there may be extra travel costs. Timeline for training, and training costs Training the approximately 1,650 CHWs may be difficult to arrange in a single year initially, especially since new course content will need to be developed. Instead, we assume that about 450 CHWs are trained in the first year, in the areas where the pilot project is envisaged. Refresher training is assumed to occur each year thereafter, at about 50 percent of the costs of the initial training. In each year thereafter, we assume an additional 300 CHWs are provided their initial training in NCD-related health promotion, screening and management. Table A8.2 outlines the resulting envisaged training schedule for the first seven years. Using the midpoint of the range of estimates above, we estimate the training costs (plus allowances) to be FJ$4.1 million in phase 1 (2-3) years; and FJ$8.5 million from years 4 to 7. 187. Krishnan et al. 2019. “Cost-effectiveness and Budget Impact of the Community-based Management of Hypertension in Nepal Study (COBIN): A Retrospective Analysis.” Lancet Global Health 7 (10): E1367-74. 146 Fiji Health Sector Review Table A8.2: Training schedule for community health workers Number of CHWs trained in NCD-related interventions Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Initial 450 300 300 300 300 0 0 Refresher 0 450 750 1050 1350 1650 1650 Will there be additional costs of managing CHWs? The CHWs in Fiji are already being supervised by district nurses. But perhaps one has also to consider that they will now be working longer hours, and therefore will need additional supervision. Plus, the literature suggests that CHW performance is dependent on close supervision and clarity of their role. Assuming that supervision hours will increase as CHW work hours increase we added an additional 10 percent of additional CHW salary costs to capture increased costs of supervision, or about FJ$600,000/year. Overall, the costs of training, supervision, and allowances will amount to FJ$16.1 million in the first 2-3 years, and then FJ$37.6 million in years 4-7. Will there be any equipment requirements? Once trained, the effectiveness of CHWs at managing chronic conditions will also depend on the equipment they have available with them (BP machine, random blood sugar testing kits, weighing scales, height measurement equipment and others, as appropriate). Equipment can be expected to depreciate/get damaged over time. Previous work suggests that the costs of this equipment are likely to be around US$150 per worker, with about one-third of the equipment failing each year. If so, we can expect the equipment costs to be US$50 per worker in future years. These costs amount to: • US$248,000 (1,650*US$150) or FJ$562,000 in the first year. • US$83,000 or FJ$187,000 in subsequent years. MHMS leadership suggested providing tablets to CHWs as an incentive. We think that this is critical equipment that will enable CHWs to deliver the extended range of services envisaged and to help link them better to the digital infrastructure that is proposed in this report. While MHMS leadership was supportive of the idea of added funding for incentive payments, non-financial incentives were also mentioned. Thus, three scenarios could be considered: (i) no extra payment + training + equipment; (ii) extra payments at FJ$4/hour + training + equipment; and (iii) extra payments at FJ$6/hour + training + equipment. Our estimates above (based on 75 percent of the minimum wage/hour) could be considered as a reasonable mid-point of these different scenarios. (S.4) Changing Behaviors of Healthcare Workers (Incentives for Health Promotion, Diagnosis and Management of Chronic Conditions, and Effective Referrals) The literature on behavior-change interventions suggests that health professionals can (opportunistically) help improve lifestyle and drug adherence behaviors among their patients.188 A recent review suggests that behavior change interventions to assist health professionals in health promotion cost about US$112.13 (about FJ$253) per person per training program. If we assume that these professionals are provided training once every two years, the cost would be roughly FJ$1 million every two years. Training costs for CHWs are captured separately (see above). 188. Ball et al. 2023. ” Effectiveness, Feasibility and Acceptability of Behavior Change Tools Used by Family Doctors: A Global Systematic Review.” British Journal of General Practice 73(731): e451-9; Patnode et al. 2017. “Behavioral Counselling to Promote a Healthful Diet and Physical Activity for Cardiovascular Disease Prevention in Adults Without Known Cardiovascular Disease Risk Factors: Updated evidence report and systematic review for the US Preventive Services Task Force.” Journal of the American Medical Association 318(2):175-93. Fiji Health Sector Review 147 (S.5) Incentivizing Private Providers to Enhance their Efforts Related to Screening, Treatment, and Management of Chronic Conditions We are unclear to what extent the government is already doing this. Any costing of this class of interventions should be based on a careful evaluation of the ongoing experience of government funding for private providers for providing health services to the poor. Some additional expenditure may be needed to strengthen strategic purchasing capacity within the government such as contracting or priority setting. It is possible that some of the functions of a strategic purchaser (such as analysis of costs, quality of care, and evaluation) could be undertaken by Fijian academic institutions with the government providing funding for this purpose. Costs of Demand-Side Interventions (D.1) Conditional Cash Transfers for Lifestyle Changes The evidence for these interventions is relatively limited. Thus, we did not cost their implementation. It may be that in a pilot, these interventions may be adopted. In any event, we expect the estimated investments to be relatively small. (D.2) Peer-based Interventions for Lifestyle Behavior Change We reviewed estimates of costs of peer-group interventions implemented in other countries that were found to be effective and cost-effective. The cost estimates of these studies were then translated to Fiji’s case to arrive at costs. Sathish et al. (2020):189 This was a study of a peer-group intervention for the prevention of diabetes in a rural sub-district of Kerala (India) among 30-60 year-olds. The content of the peer support program was “culturally adapted” from evidence-based peer support interventions in high-income countries (e.g., Finland, Australia, and the United States). The intervention consisted of 15 group sessions conducted over 12 months, conducted in local neighborhoods in community buildings on weekends. Project staff delivered an initial 60–90 minute session to introduce group participants to the program and its mentoring style. Experts in nutrition, physical activity, and diabetes delivered two half-day sessions on prevention and management strategies for diabetes. Peer leaders were identified from within the group (10–20 people in one group) and delivered 12 sessions at one-month intervals for 60-90 minutes. These peer leaders were trained and provided with training materials, and in ways to run peer-group sessions. Lifestyle interventions focused on bodyweight, healthy eating habits, adequate sleep, physical activity, tobacco cessation, and alcohol consumption. Participants were encouraged to maintain contact with peer leaders outside the group sessions. The estimated cost of the training and sessions (human resources, overheads, etc) were US$24.20 per participant in peer groups deemed to be at high risk of moving to diabetes. The additional time costs of attending group sessions was equivalent to US$6.20 per participant. The intervention costs assume that screening has already been done (thus screening costs would be extra but would obviously apply only to a subset of the population that was deemed at risk). Shearer et al. (2021):190 reported the results from a peer-support intervention intended to bring about lifestyle behavior change in the people aged 5–40 years in urban Colombo (Sri Lanka). Participants deemed to be at risk based on lifestyle factors were given an “intensive peer-educator delivered tri-monthly lifestyle modification” (four sessions annually) for roughly three years. Trained peer educators offered individualized lifestyle modification advice. The cost was US$33 per participant in the group sessions annually on average (with higher costs in the first year), but this excluded the cost of training peer educators. Johanssen et al. (2016):191 reported the results of a peer education intervention in Austria for patients with diabetes. The intervention consisted of regular group meetings facilitated by trained peer supporters. The group exercised together, discussed diabetes-related topics, and received support from professionals. This resulted in shorter hospital stays. The cost of intervention was estimated to be EUR 211 per participant per year (including peer supporter training, physical activity meetings, professional support, travel expenses, peer 189. Satish et al. 2020. “Cost-effectiveness of a Lifestyle Intervention in High-risk individuals for Diabetes in a Low- and Middle-Income Setting: Trial-Based Analysis of the Kerala Diabetes Prevention Program.” BMC Medicine 18:251. 190. Shearer et al. 2021. “Cost‑Efectiveness of Peer‑Educator–Delivered Lifestyle Modifcation for Type 2 Diabetes Prevention in a Young Healthy Population in Sri Lanka: A Trial‑Based Economic Evaluation and Economic Model.” Pharmacoecon Open 5(4): 693–700. 191. Johanssen et al. 2016 “Effectiveness of a Peer Support Program versus Usual Care in Disease Management of Diabetes Mellitus Type 2 Regarding Improvement of Metabolic Control: A Cluster-randomized Controlled Trial.” Journal of Diabetes Research. doi: 10.1155/2016/3248547. 148 Fiji Health Sector Review supporter reimbursement, and print materials). Given that Fiji’s income per capita is about one-fourth that of Austria’s in PPP terms, this amounts to about EUR 42 per person enrolled into the peer groups per year (if assessed in the Fijian context), or about US$46 per person. In summary, a reasonable estimate of the range of costs of peer group interventions is about US$30–46 per participant in group sessions. If the targeted age group was 30–65 year-olds, this would constitute about 40 percent of the Fijian population. If the pilot intervention was run in Suva, as proposed, with an estimated population of around 100,000, this translates into a target population of 40,000. If one-tenth of this group participate (about 4,000), the cost of the peer-group intervention could range from US$1 million to US$2 million (FJ$2.3– 4.5 million). (D.3) School Health/Garden Programs This section provides estimates for multi-intervention garden programs consisting of the following: • School gardens • Education about agriculture, nutrition, and water, sanitation, and hygiene education and practices • Community outreach in schools to improve food security and nutrition (involving parents as part of community involvement). Such programs can help to promote environmental education, nutritional behavior change, and community engagement. The estimated costs per school (for a three-year period) from similar projects are US$5,900 (in Bhutan), US$2,952 (in Burkina Faso), US$1,950 (in Indonesia), and Nepal US$1,350 (in Nepal).192 Indonesia might be a useful comparator given the large number of islands, although school sizes are likely smaller. Table A8.3: School distribution and total number of students Division Primary Children/Primary Secondary Children/Secondary Central 200 64 Western 254 57 Northern 166 37 Eastern 116 14 TOTAL 736 116,500 172 89,900 We could not find comparable numbers for Indonesia and Fiji for number of children per school. For the moment, if we assume the school program to cost the same as in Indonesia and that this is done separately for secondary and primary schools, we end up with a three-yearly cost of FJ$4.10 million, or about FJ$1.37 million per year. We also extrapolate from the proposed implementation plan for Fiji’s revised Healthy School Policy underway as this report is being drafted. (D.4) Cost of Traffic-Light Food Labelling An Australian study by Sacks et al. (2011)193 classified the costs of such interventions into three areas: • Legislation (implementing, administering and enforcement) = AUD 18 million • Cost of social marketing campaign = AUD 2.8 million • Cost of changing food labels (borne by industry) = AUD 62 million. 192. World Vegetable Center. School Gardens for Nutrition and Health. https://avrdc.org/download/publications/policy-briefs/REV1_VGTS-Policy- Paper_6pagesA4_in-sequence.pdf. 193. Sacks et al. (2003) “Traffic-light Nutrition Labelling and ‘Junk Food Tax’: A Modelled Comparison of Cost-effectiveness for Obesity Prevention.” International Journal of Obesity. DOI: 10.1038/ijo.2010.228. Fiji Health Sector Review 149 Adjusted for inflation, this would be 70.2 percent higher compared to 2003 (Reserve Bank of Australia). On the other hand, technology could have reduced the cost of changing labels, etc. We assumed a 50 percent increase in costs over time. The total costs would therefore be approximately AUD 125 million. Fiji is a much smaller economy than Australia, with much lower food consumption. For instance, food consumption in Australia was estimated to be $118.5 billion in 2021–22 at current prices (about 10 percent of all consumption expenditure, but Australia is richer than Fiji so the share of food consumption can be expected to be lower). In contrast, in Fiji (total household consumption expenditure, food and non-food) was FJ$7.72 billion, or AUD 5.1 billion at 2022 prices. The Reserve Bank of Fiji puts a weight of 28 percent on food in its consumer price index and presumably one could assume that 28 percent of all household spending is on food; the HIES data for Fiji estimates suggest 17 percent (this needs to be further confirmed). With this as the range, food spending in Fiji can be assumed to range from AUD 0.87 billion to AUD 1.43 billion. • We assume that the cost of food labelling is directly proportional to food spending, all else the same. For Fiji this would be approximately, AUD 0.68 million to AUD 1.12 million in 2022, or about FJ$1.04 million to FJ$1.70 million in 2022. • The cost of the social-marketing campaign will depend on a variety of factors – population size, difficulty of accessing population sub-groups, etc. Again, adjusting for inflation, in Australia the expenditures in 2023 would be AUD 4.8 million (FJ$7.3 million). If we assume that the costs are directly proportional to population size, the corresponding 2023 number for Fiji would be (recognizing that the Australian population in 2003 was 19.72 million), FJ$0.34 million. Assuming 15 percent extra costs to account for the remote location of many Fijian population sub-groups, we get FJ$0.40 million. • Legislation costs (administration, implementation and so on) in 2022 current prices in Australia would be FJ$46.5 million. Presumably it would be much lower in Fiji – so let us assume a population-ratio based cost (relative to Australia) of about FJ$2.1 million. • The total estimated costs would therefore be FJ$3.5–4.2 million. (D.5) Costs of Health Taxes We do not expect health tax increases to have significant intervention costs. That is because there is likely already a set up in place to implement such taxes. Perhaps there may be some costs related to communication of these changes, relevant legislation, and perhaps the costs of undertaking feasibility studies to determine how taxes are to be levied (there is already a substantial international literature on this topic), including assessing any impacts on employment and incomes. Thus, the costs of implementing increased taxes under a health tax regime were ignored in our estimates. There may be costs of combating smuggling and other illegal activities arising from the imposition of tax increases. We suspect though that the bigger impacts will likely be on revenue, and the evidence suggests that revenues generally go up after tax increases. (D.6) Behavior-change Interventions Among Patients with NCDs There is emerging evidence of effectiveness and cost-effectiveness of digital interventions. A recent study used SMS messages to encourage diabetes patients to adopt lifestyle changes, with one message sent daily for 6 months. The cost of the intervention was roughly US$4 per person, with an additional US$4 per person for costs (including human resource time) arising from increased use of healthcare among the targeted group.194 A review of the literature suggests that SMS-based interventions (e.g., reminders for healthcare appointments) suggests that the costs for such interventions average around US$3 per person reached, with a range of US$0.11–11.63.195 If we assume that similar programs are addressed to the estimated 162,500 people with diabetes (or impaired glucose tolerance), this will likely amount to about US$650,000 (or FJ$1.5 million) annually in program costs. Presumably an additional 33 percent of the 90,250 patients estimated to have diabetes will seek care, or about 30,000 (at a cost of about US$15 per person year for medicines) or about FJ$1 million per year. Thus, the total cost would be FJ$2.5 million annually. If all adults over 30 with high blood pressure w also targeted (over and above any who have diabetes), we expect the program to reach about 150,000 people, or approximately FJ$1.5 million. Thus, the total costs of SMS messages plus treatment 194. Islam et al. 2020. ”Cost-effectiveness of a Mobile-Phone Text Messaging Intervention on Type-2 Diabetes: A Randomized Controlled Trial.” Health Policy and Technology 9:79-85; Nkhoma et al. 2021 ”Digital Interventions Self-management Intervention for Type 1 and Type 2 Diabetes: A systematic review and meta-analysis” Computer Methods and Programs in Biomedicine 210:107370. 195. Avenir Health. 2021. ”Documenting the Costs of Social Behavior Change Interventions for Health in Low- and Middle-Income Countries.” Breakthrough RESEARCH Technical Report. Washington, DC: Population Council. 150 Fiji Health Sector Review costs would amount to FJ$4.0 million annually, for people at risk of diabetes and hypertension. Note that this estimate would effectively also account for increased use of healthcare due to screening. Annex 9: A Scorecard of Fiji’s Efforts Against NCDs The Pacific Monitoring Alliance for NCD Action (MANA) dashboard uses a “traffic light” rating system to track countries’ progress on policies and legislation and, if available, the strength of related action and implementation, across eight thematic areas, covering 31 indicators. Fiji’s progress in the eight thematic areas is outlined below (see also Box A9.1): Leadership and Governance: Since 2019, policies and legislation have been available on multi sectoral taskforce (low action or implementation), national strategy (no action or implementation), with explicit indicators and targets (no action or implementation). Tobacco Preventative Policy: Fiji has performed strongly across this thematic area whereby policies and legislation have mostly been accompanied by medium or high action or implementation. Alcohol Preventative Policy: Apart from alcohol advertising, other areas under this thematic area have policies and legislation in place. There is mixed intensity of action or implementation across policies and legislation around alcohol licensing to restrict sales, taxation, and drink driving. Food Preventative Policy: Unhealthy food marketing to children is the only policy or legislation that is currently not in place. Of the remaining five policies or legislation, there has been no to low action or implementation. Physical Activity Preventative Policy has medium implementation or action concentrated around compulsory physical education in the school curriculum. Enforcement: While there is an existing policy and legislation on enforcing laws and regulations related to NCD risk factors, there has been no action or implementation. Health Systems Response Programs: All policies covered under this thematic area are present with low to medium action or implementation. Monitoring: Two policies are in place with moderate to high action or implementation while youth and adult population risk factor surveys have been delayed. Fiji Health Sector Review 151 Box A9.1: Pacific Monitorig Alliance for NCD Action (MANA Dashboard) 152 Fiji Health Sector Review Annex 10: Current Allocation of Skilled Healthcare Workers (Doctors, Nurses, and Midwives) Figure A10.1: Allocation of public sector skilled healthcare workers (SHW/1,000 population) at different levels of care in the Central Division Figure A10.2: Allocation of public sector skilled healthcare workers (SHW/1,000 population) at different levels of care in the Eastern Division Fiji Health Sector Review 153 Figure A10.3: Allocation of public sector skilled healthcare workers (SHW/1,000 population) at different levels of care in the Western Division Figure A10.4: Allocation of public sector skilled healthcare workers (SHW/1,000 population) at different levels of care in the Northern Division 154 Fiji Health Sector Review