RECOMMENDATION DESIGN REPORT HEALTH SYSTEM TECHNICAL SUPPORT TO GOVERNMENT OF GHANA: ASSESSING NETWORKS OF PRACTICE THROUGH PATIENT PATHWAY AND NETWORK ANALYTICS Table of Contents List of Acronyms 1 Executive Summary 2 Introduction to the Research Study 4 Recommendation 1 – Configure New NoPs by Using a Comprehensive Mapping Approach, which Includes Health Facility Profiles, Perspectives of Healthcare Providers and Users, © 2024 International Bank for and Geolocation 11 Reconstruction and Development / The World Bank Recommendation 2 – Prioritize the Further Upgrading of Network 1818 H Street NW Hubs to Reorient Patient Pathways, Decongest Hospitals Washington DC 20433 and Strengthen the Primary Care Levels 12 Telephone: 202-473-1000 Internet: www.worldbank.org Recommendation 3 – Invest in the Human Resources Available at This work is a product of the staff Primary Care Facilities to Improve Patient-Provider Contacts and of The World Bank with external Facility Capacity to Deliver the Full PHC Service Package 13 contributions. The findings, interpretations, and conclusions expressed in this work do not Recommendation 4 – Strengthen the Quality of Care Especially in necessarily reflect the views of The the Network Hubs Given that Quality Drives Care-Seeking Behavior 14 World Bank, its Board of Executive Directors, or the governments they represent. Recommendation 5 – Improve NoP Maintenance Efforts 15 The World Bank does not guarantee the accuracy of the data included in Recommendation 6 – Build Capacity on Patient Referral to Make this work. The boundaries, colors, the System Work Better for Patients and Providers 16 denominations, and other information shown on any map in this work do not imply any judgment on the part of The Recommendation 7 – Promote Health Equity through Increasing World Bank concerning the legal status Awareness, Community Engagement, and Financial Protection 17 of any territory or the endorsement or acceptance of such boundaries. Recommendation 8 – Design Effective Urban Networks of Practice 18 Rights and Permissions The material in this work is subject to copyright. Because The World Conclusion 19 Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@ worldbank.org. Photo credits All photos courtesy of School of Public Health, University of Ghana. RECOMMENDATION DESIGN REPORT List of Acronyms Acronym Definition Acronym Definition ANC Antenatal Care MNH Maternal and Newborn Health CEmONC Comprehensive Emergency Obstetric and NAS National Ambulance Service Newborn Care NGO Non-Governmental Organization CETS Community Emergency Transport System NHIA National Health Insurance Authority CHAG Christian Health Association of Ghana NHIS National Health Insurance Scheme CHMC Community Health Management Committee NoP Network of Practice CHPS Community-Based Health Planning and OOPE Out-Of-Pocket Expenditure Services PHC Primary Health Care GhiLMIS Ghana Integrated Logistics Management PNC Postnatal Care Information System PPA Patient Pathway Analysis GHS Ghana Health Service PPP Public-Private Partnership GHSERC Ghana Health Service Ethics Review SNA Summative Network Analysis Committee SPH School of Public Health GIS Geographic Information System TBA Traditional Birth Attendants HCW Health Care Worker UHC Universal Health Coverage HMIS Health Management Information Systems UNICEF United Nations Children’s Fund HRIMS Human Resource Information Management USAID United States Agency for International System Development HTN Hypertension WBG World Bank Group 1 RECOMMENDATION DESIGN REPORT Executive Summary This report provides comprehensive recommendations selecting appropriate hub locations and ensuring timely designed to strengthen primary health care (PHC) in Ghana referral services, especially for emergency cases like through its growing Networks of Practice (NoP) Program, maternal and newborn care. Collaboration with district based on research in six selected districts. It focuses on assemblies, communities, National Health Insurance addressing the challenges identified in how the current Authority (NHIA) district levels, and Health Development health system works for patients and healthcare workers Partners is essential to optimize NoP effectiveness and (HCWs) by offering actionable steps for improvement. The participation. research was designed to provide information on how Prioritize the further upgrading of the network hubs to patients navigate the local health system, especially for reorient patient pathways, decongest hospitals, and conditions that require repeated health contacts, continuity strengthen the primary care levels of care, and retention (such as the maternal care continuum and hypertension care). It also assessed how networks ■ Rationale: Health centers need to be capable of function and what obstacles HCWs face in implementing the managing conditions at the level of their competency NoP vision. The study, therefore, informs NoP design and and to make them a preferred point of primary contact strengthening so that the NoP objectives, such as improved for healthcare users. This will help reduce the current referrals, better provider collaboration, and enhanced burden on district hospitals due to the bypassing of knowledge sharing, come to fruition. health centers for higher-level care. In the NoP Program, telemedicine and teleconsultations are considered The key recommendations presented here span NoP potential components (“spokes”) of the network. configuration, health facility infrastructure, human However, telemedicine services were available only in resources for health, use of technology, and community Hohoe at the District and Zonal Hospitals and Hubs, participation. They are accompanied by considerations of reportedly improving consultation ease in some cases. the “why” and the “how” and will be carried forward with Community members still favored face-to-face local health teams through knowledge translation support consultations and appointment-based services. by the Ghana Health Service. ■ Actions to implement recommendation: To enhance health centers’ capacity, prioritize achieving maturity KEY RECOMMENDATIONS level 3 as per NoP Implementation Guidelines, ensuring they have level-specific infrastructure and Configure new NoPs by using a comprehensive mapping laboratories for diagnostics. Maintain adequate approach, which includes health facility profiles, equipment and drug availability aligned with national perspectives of healthcare providers and users, and norms, strengthen the drug supply chain using geolocation Ghana Integrated Logistics Management Information ■ Rationale: Patients strongly prefer higher-tier health System (GhILMIS), and expand telemedicine across all facilities, leading to overburdened district hospitals. facilities to improve efficiency, reduce hospital referrals, Strengthening health centers with appropriate and manage patient load while promoting sustainability resources will improve primary care services and in the health system. alleviate this burden. Invest in the human resources available at primary care ■ Actions to implement recommendation: Implementing facilities to improve patient-provider contacts and facility NoPs requires detailed stakeholder mapping to capacity to fully deliver the PHC service package include all relevant formal and informal care providers, such as pharmacies, clinics, and other health posts, to ■ Rationale: One of the keys to good patient-provider improve collaboration and trust in the formal healthcare interaction is effective, courteous communication and system, especially in urban areas. Utilizing geographic respectful staff behavior. A recurring concern raised information system (GIS) technology to assess by respondents was poor conduct by the health staff, transportation availability and road safety is crucial for which also influenced the patients’ preferences 2   RECOMMENDATION DESIGN REPORT regarding health facilities. Patient observations also awareness among CHAG facilities, highlight the need suggested poor listening by health providers. HR for continuous training, proper handovers, and digital shortages, staff attrition, suboptimal handovers and health tools to ensure complete data entry and effective skills sets affected the service delivery in some health program monitoring. facilities. ■ Actions to implement recommendation: Appoint NoP ■ Actions to implement recommendation: Implement coordinators at each health facility, enhance capacity communication, soft skills, and leadership training for building through regular workshops and supportive healthcare providers to enhance patient interactions. supervision, and strengthen monitoring and evaluation Increase recruitment of healthcare staff, particularly by instituting key performance indicators, conducting physician assistants, community health nurses, and regular review meetings, and setting up a reward midwives, to improve care continuity and reduce facility system. Adapt DHIMS for network-level data capture, waiting times. Adopt workforce retention strategies increase community involvement with scorecards, such as incentives, performance bonuses, and equip health centers with digital devices for data recognition programs to attract and retain healthcare management and record-keeping, and ensure professionals, supported by capacity-building and adequate funding to support network activities. refresher training programs, while ensuring HRIMS is effectively utilized for HR allocation planning. Build capacity on patient referral to make the system work better for patients and providers Strengthen the quality of care, especially in the ■ Rationale: Weak referral gatekeeping in six study network hubs given that quality drives care-seeking districts of Ghana led to most Community-Based behavior Health Planning and Services (CHPS) compounds ■ Rationale: Quality of care was the primary driver for referring directly to district hospitals instead of health health facility selection among many respondents, centers. Overlooked factors like transportation and with 46% of HTN respondents citing it during initial family support, lack of patient involvement in referral contact and 55% during monitoring visits of HTN. This decisions, delays in emergencies, financial burdens, led patients to prioritize better perceived quality over and rare follow-ups by health workers also led to patient cost and travel to distant facilities. For NoPs and dissatisfaction. universal health coverage initiatives, ensuring high- ■ Actions to implement recommendation: Educate quality care and positive health staff attitudes is health staff on proper referral mechanisms and imperative. services available at various health system levels and ■ Actions to implement recommendation: Strengthen disseminate information about insurance-covered supportive supervision, on-the-job training, and services to minimize out-of-pocket expenditure (OOPE). mentorship, and sensitize healthcare workers to Strengthen collaboration with district hospitals for adopt positive attitudes towards patients to enhance training, expand the National Ambulance Service (NAS) care quality. Empower communities through the for better availability and response times, and enhance Community Score Card initiative, institute facility cooperation between the NAS and Community modifications for better patient experience, and Emergency Transport System (CETS) for efficient provide mandatory refresher training and continuing patient transfers. education for staff. Promote health equity through increasing awareness, Improve NoP maintenance efforts community engagement, and financial protection ■ Rationale: NoPs in the study districts are largely ■ Rationale: Achieving health equity in Ghana requires successful but require enhanced operational empowering communities with knowledge, involving partnerships such as with CHAG and private facilities, them in decision-making, and ensuring financial improved monitoring, and better coordination. protection to support equitable access to healthcare. Issues in Dormaa Central, such as fewer review While some community members are aware of NoPs, meetings, insufficient supervision, and lack of general awareness, and utilization remain suboptimal. 3 RECOMMENDATION DESIGN REPORT The strong preference for higher-tier health facilities care. Patients in urban areas have easier access to is likely due to perceived better service quality, facilities but higher OOPE compared to rural areas. This highlighting the need to build public trust in primary highlights the need for a distinct urban NoP design that care facilities while addressing challenges such as is different from the rural model. high OOPE. ■ Actions to implement recommendation: Map and ■ Actions to implement recommendation: Appoint a assess diverse health providers for inclusion in urban dedicated focal person at the district level and NoPs, upgrade public sector facilities to enhance their coordinators at each spoke to enhance public service range and reduce OOPE, establish formal awareness and advocacy. Regularly train healthcare collaborations with the private sector, implement staff on NoP interventions, leverage Community Health comprehensive training for private facility staff, and Management Committees (CHMC) for community provide HMIS access for effective monitoring and awareness and engagement, strengthen collaboration review. with district hospitals for clinical training, optimize NHIS policies to reduce out-of-pocket expenses, establish NHIA-credentialed health centers in areas CONCLUSION with critical deficits in care access, consider a The recommendations outlined in this report aim to inform strategy of upgrading of specific CHPS compounds the evolution of the NoP Program in already covered districts to health centers, and include selected private/NGO and new areas in rural and urban Ghana. By focusing on sector facilities under NHIS for broader financial improving infrastructure, human resources, technology, and protection especially in urban areas. community engagement, the report provides a roadmap for enhancing healthcare delivery and outcomes through a Design effective urban networks of practice successful NoP Program. These measures, when implemented, ■ Rationale: Urban residents in Ghana prefer hospitals and will not only alleviate the current challenges but also pave private clinics over health centers for MNH and HTN the way for a more resilient health system in Ghana. Introduction to the Research Study BACKGROUND TO THE RESEARCH collaborative operations. The NoP model employs a hub-and- spokes structure, with health centers (hubs) providing technical The Networks of Practice (NoP) initiative is a strategic effort and operational support to connected health delivery points by Ghana to achieve Universal Health Coverage (UHC) at the (spokes), which include health posts, clinics, maternity homes, primary care level, emphasizing quality health services, pharmacies, and other PHC providers. The hubs are partnership, and innovation. Initially piloted from 2017 to strengthened to become “Model Health Centres” to ensure 2019 in two rural districts by the Ghana Health Service (GHS) widespread access to basic healthcare services locally. and the Ministry of Health, with support from USAID’s Systems NoP implementation is based on nine key guiding principles1 for Health Program, the approach showed positive results in envisaged by the GHS (Table 1). facility collaboration and service delivery. Consequently, it was scaled up to 10 additional rural districts in 2020 and The main objectives of the NoP initiative are to ensure 2021 to develop a sustainable model for equitable, efficient, universal access to quality healthcare, reduce maternal, and high-quality primary health care (PHC). adolescent, and child mortality and disabilities, and increase access to emergency services. The initiative is accompanied The NoP initiative aims to improve the coverage and quality of by analytical work and knowledge translation to monitor, various health services by organizing and supporting learn from and evaluate it. This study, initiated when community and sub-district health services within a formal district health system. It connects public and private health service sites through an administrative and clinical 1 Implementation Guidelines for Networks of Practice. Ghana Health management model that fosters client-centered, efficient, and Service. 2024. 4 RECOMMENDATION DESIGN REPORT TABLE 1. CORE VALUES AND GUIDING PRINCIPLES ENSHRINED UNDER THE NoP IMPLEMENTATION GUIDELINES Core values and Guiding Principles enshrined under the NoP Implementation Guidelines Equity Focus on essential health services package, the poor and vulnerable with emphasis on pregnant women and children, adolescents, and the elderly. Patient centeredness Patients shall be handled with respect and dignity and included in decision- making about their health. Quality of Care Adherence to the highest standards of care Strengthened referral systems Work in line with the district health system to maintain the gatekeeper system such that clients receive timely and effective care at the most appropriate level of the health delivery system. Internal collaboration and relationship building Foster teamwork among the key players in direct service delivery, management and resource utilization. Partnership at the operational level Include faith-based/ private health facilities/providers, pharmacies and chemical sellers in networks to ensure access to PHC services. Partner with the district assemblies, communities, district National Health Insurance Scheme (NHIS) and other partners for NoP scale-up. Commitment to working together and Inclusiveness, transparency, and fairness supporting each other Community engagement Understand community health needs and promote patronage of NoPs by the community members. Continuous Learning and Adaptation Monitoring and documentation of key lessons, challenges to inform course adjustment, tools development, and dissemination of best practices. approximately 15 rural districts implemented NoPs, was STUDY DESIGN supported by the World Bank, IQVIA, the University of Ghana School of Public Health, and NTT Data and carried out under The study was designed and conducted in 2023-24 across GHS oversight. six districts in four regions of Ghana (Table 2). Four districts were rural, with two implementing the NoP program (7 official NoPs included) and two not (9 potential networks defined for PATIENT PATHWAY ANALYSIS AND study purposes). The two rural non-NoP districts were SUMMATIVE NETWORK ANALYSIS selected from the same regions as the NoP districts, ensuring they were matched across socioeconomic indicators and The study aimed to evaluate how clients seeking services for health outcomes using Multidimensional Poverty Indicators either maternal and newborn health (MNH) or for hypertension and health scores from the 2021 District League Table published (a tracer for chronic non-communicable disease) navigate the by the National Development Planning Commission and local health system and how care delivery and health system UNICEF. Additionally, two urban districts were included functions are organized to meet their needs in both NoP and (7  potential networks defined for study purposes) to help non-NoP settings. To achieve this, the team employed two inform future NoP rollout in Ghana’s urban areas where 59% novel methodologies: Patient Pathway Analysis (PPA) and of the country’s population live. Summative Network Analysis (SNA), which in combination assess the interactions and dynamics between health system An initial review of key documents on NoP implementation in users and providers. The study objectives included expanding Ghana was followed by a stakeholder mapping activity at the the knowledge base on NoPs, understanding the structural district and sub-district levels. In consultation with GHS and and relational aspects of health facilities, and validating based on the reviewed district data, the study districts were research results with local stakeholders to identify solutions. chosen to represent areas of lower socio-economic status. 5   RECOMMENDATION DESIGN REPORT TABLE 2. PPA/SNA STUDY DISTRICTS AND NETWORKS TABLE 3. DATA COLLECTION MODALITIES District 1 District 3 Data collection modality No. of respondents NoP Hohoe West Dormaa Central Health system users (50% MNH, 50%HTN) (Volta Region) (Bono Region) Exit interviews 780 Rural ■ 3 official NoPs ■ 4 official NoPs district Community based interviews 351 District 2 District 4 pairs Shadowing of care visits 156 No Ketu North Tain NoP (Volta Region) (Bono Region) Health care providers ■ 6 potential ■ 3 potential Key Informant Interviews – KIIs 210 NoPs NoPs Focus Group Discussions – FGDs 12 District 5 District 6 Poor health Fairer health outcomes outcomes DATA COLLECTION AND ANALYSIS No Urban Ayawaso Central Atwima- NoP Secondary data collection: Existing routine data was (Greater Accra Nwabiagya collected in two rounds, first at the district level, and then at Region) (Ashanti Region) the facility level. The facility-level health indicator data ■ 3 potential ■ 4 potential concerned service delivery, budget levels, infrastructure, NoPs NoPs and performance to understand the utilization and performance of the health facilities in a network. District- level data was collected for all six districts. Focal persons The following health indicators were relevant to district from the district-level offices of GHS (district health teams) selection (2021 District League Table data and scores): were identified from each district. ■ Percentage of mothers with four antenatal care (ANC) Primary data collection: The study team conducted district visits engagement visits over three weeks to obtain information ■ Percentage of mothers/newborns receiving postnatal on private practitioners/clinics, traditional healers, traditional care (PNC) within 48 hours of delivery birth attendants (TBAs), drug sellers, and pharmacies, and map out facility selection for primary data collection. The ■ Percentage of fully immunized children under one data collection instruments, modelled on similar research year (using Penta-3 as a proxy) studies, were adapted to the needs of this study. They were ■ Percentage of skilled birth attendance coded and administered in the field using REDCap. A three-day comprehensive training session was conducted to improve The study was, therefore, designed to make comparisons understanding of protocols, tools, and data collection between rural ‘NoP’ and ‘non-NoP’ districts and between modalities, and the ethical principles of data collection. rural and urban districts. The study considered the patient Once the study protocols and tools were finalized, ethical pathways and experiences of two types of healthcare clients: clearance was sought from the Ghana Health Service Ethics ■ Pregnant and recently delivered women accessing Review Committee (GHS-ERC). The ethical approval was antenatal, delivery and post-natal, and ancillary services. obtained on 13 June 2023 (GHS-ERC 022/04/23). The data collection tools were piloted in Madina La Nkwantan District. ■ Patients accessing care for hypertension. Data collection was completed over one and a half Quantitative and qualitative data collection modalities were months, followed by transcription of the qualitative data used, and sample sizes were large enough to enable analysis and cleaning of the quantitative data according to best by NoP status and rural/urban setting. A study protocol, ethical research practices. In the analysis of pathways, the study procedures of patient consent, and appropriate data team took into consideration Ghana’s essential health collection tools were developed and carefully reviewed services package and where services are expected to be against the study objectives. provided along the continuum of care for the conditions 6 RECOMMENDATION DESIGN REPORT assessed. Beyond applying this normative standard, the regular ANCs (after the 1st ANCs), the onset of labor, delivery reporting also tried to reflect how some more contemporary of the child, and PNC services. The main observations are: trends in the concept of ‘right-place care’ could be taken ■ There was a strong preference for hospital-level up in the context of Ghana, including devolving basic NCD services across all maternal care touchpoints, care to lower levels of the health system and ensuring including regular ANC and PNC. availability of care for MNH complications in higher-level ■ With the advancing maternity journey, there was a facilities within a safe timeframe. gradual increase in the utilization of hospital services, The project team conducted analysis using tools under with a slight drop in the utilization of health centres. Microsoft Excel, Atlas.ti, QGIS and Python. During data analysis, ■ Maximum hospital utilization (72%) was seen for the findings from different data collection sources were delivery. Out of the 71 women who delivered at triangulated. Following this, a ground-truthing of the key hospitals, only four were referred by providers, and the findings from the PPA and SNA was conducted. remaining 67 ‘preferred’ hospitals for delivery. ■ Postnatal care was more decentralized, with a higher COMMUNITY ENGAGEMENT FOR utilization of lower-tier health facilities. VALIDATION AND CONSULTATION HTN care pathways The community engagement aimed to validate and refine the Figure 2 shows the flow of 420 HTN respondents through the PPA and SNA findings through “ground truthing”, delve deeper health system seeking care for their HTN. Touchpoints with into the identified issues, and listen to suggestions from health the health system depicted are 1st  contact with the health system users and providers for improving healthcare delivery. system regarding any symptom linked to HTN, diagnosis of Ground truthing, as described by Adams et al. (2015), serves as HTN, treatment initiation and maintenance, and monitoring a crucial step in health research, mitigating biases and services. The main observations from the HTN pathway are: enhancing the validity of findings by comparing them with independently gathered qualitative data. ■ The HTN pathway was hospital-centric across all The engagement involved returning to the communities touchpoints even though HTN is an ambulatory care- where the initial data was collected, ensuring that the study’s sensitive condition. conclusions were not artifacts of study design, data collector ■ Most HTN respondents (71%) had entered the care bias, or analytic retrofit. The study team held 12 discussions pathway via a blood pressure check during any hospital with diverse community groups and healthcare providers to visit, leading to circumstantial detection of possible HTN. capture a wide range of perspectives. A total of 25 community Another 24% of the respondents had been detected members per session were selected from varied groups when presenting with HTN-related symptoms, 4% during such as Drivers’ Unions, Hairdressers and Dressmakers’ an acute emergency, and 1% in preventative screening. Associations, Pensioners’ Associations, and Disabled ■ The pathway results suggest that health centers and Persons’ groups, ensuring representation from different health posts provide HTN monitoring and that some demographic segments, including the elderly and those with patients are down-referred to continue their treatment disabilities. Each session also included approximately and monitoring at lower-level facilities. There was less 15 healthcare providers, both clinical and administrative, to evidence of decentralization of diagnosis and treatment validate health system perspectives from the primary data. initiation for HTN to the health center level, for which Moderators familiar with the original data collection were patients continue to use hospitals. trained to facilitate unbiased discussions using carefully ■ In an NoP, the hub and spokes are expected to work designed guides. Key findings are outlined below. together on long-term HTN management, with the spokes being utilized for regular BP monitoring and KEY PRIMARY FINDINGS the hubs being the point for prescribing/review. However, the study found that prescriptions for HTN MNH care pathways (postpartum respondents only) treatment were not always taken at the hubs (health Figure 1 shows the journeys of 136 MNH respondents centers) and patients showed a tendency to circumvent as they navigated the health system seeking healthcare. the network hubs and go directly to the district hospital Five care touchpoints were defined and tracked: 1st ANC, for HTN care, including diagnosis and prescription. 7 RECOMMENDATION DESIGN REPORT FIGURE 1. MNH CARE PATHWAYS Care at Onset of First ANC Regular ANCs Labour Delivery Regular PNC Hospital (H) H 54 H 44 H 65 H 71 H 56 Health Centre (HC) HC 39 HC 34 HC 30 HC 27 HC 37 Clinic (CI) CI 18 CI 18 CI 23 CI 25 CI 12 Health Post (HP) HP 15 HP 12 HP 12 HP 9 HP 14 Maternity Home (MH) MH 2 MH 1 MH 1 MH 1 MH 1 Traditional (TH) TH 1 31 5 Unassisted Don’t delivery at 4 Know home Did not visit a health 2 1 No PNC provider (NP) NP 1 No regular NP 10 Delivered on ANCs the way to the facility 4 Assisted delivery at home FIGURE 2. HTN CARE PATHWAYS Treatment Treatment Monitoring First contact Diagnosis initiation maintenance facility type Hospital (H) H 219 H 228 H 232 H 210 H 165 Health Centre (HC) HC 88 HC 88 HC 89 HC 92 HC 116 Clinic (CI) CI 63 C 65 CI 67 CI 56 CI 64 Health Post (HP) HP 34 HP 19 HP 8 HP 46 HP 30 Maternity Home (MH) MH 6 MH 3 MH 3 MH 3 MH 6 Pharmacy (Ph) Ph 5 Ph 3 Ph 4 Ph 40 Ph 11 Home based non-physician health NHW 4 NHW 1 NHW 1 NHW 2 worker (NHW) Traditional (TH) TH 1 TH 1 TH 3 TH 2 TH 2 Community Health Worker (CHW) CHW 1 1 2 9 7 No diagnosis No initiation No maintenance No of treatment of treatment monitoring 8 RECOMMENDATION DESIGN REPORT Care pathways involving NoP hubs or health centres mission hospitals across most touchpoints, sometimes being If NoP hubs work well, MNH and HTN care pathways can be preferred even more than public hospitals. In rural districts expected to pivot towards health centres. Results from health centres were usually the second most preferred pathway interviews suggested that in NoP districts, there health facilities after hospitals. For example, to seek 1st ANC, was higher utilization of health centers across almost all hospitals were the most visited facility in rural areas (34%), MNH and HTN care touchpoints, than in their paired non- followed closely by health centres (33%), whereas, in urban NoP districts. Pathway data showed, for instance: areas, the most preferred facilities were private sector facilitates (42%) followed by public sector hospitals (32%). ■ A 19-percentage point greater use of health centres Health centres had minimal utilization in urban areas, for first ANC in Hohoe’s NoPs vs. Ketu North District sometimes even lesser than health posts/ CHPS compounds. ■ A 15 percentage points greater use of health centres Overall, the roles played by clinics/private facilities and for the onset of labor in Dormaa Central’s NoPs vs. health centres were reversed in the two urban districts Tain District compared to the rural districts in the sample. ■ A 21-percentage point greater use of health centres Network analysis showed that most districts have two or for HTN treatment monitoring in Hohoe’s NoPs vs. more NoP hubs in close proximity to each other, and Ketu North District some instances of spokes being closer to the district ■ A 19-percentage point greater use of health centres hospital than to their respective hubs. This may lead to the for HTN treatment initiation in Dormaa Central’s NoPs bypassing of hubs by primary care patients going directly vs. Tain District to the district hospital. A positive finding was that most spokes were within 30 minutes of travel time for their The pivot towards NoP health centres was primarily an respective hubs- a factor crucial for obstetric emergency effect of decreased care-seeking at lower-level health care. Referral patterns were not always aligned with the facilities, i.e. patients coming directly to the health centre hub and spoke model. Some patients preferred visiting hub instead of the spoke. In Dormaa Central, there was also health facilities that were easier to reach through safe a small shift away from seeking care at the hospital level for roads and cheaper transport. The unavailability of MNH. However, the picture was mixed and the sample ambulances and free transport was found to be a deterrent numbers were too small to reliably detect shifts in patterns towards referral completion from spokes to hubs. of by-passing spokes vs. pivoting away from hospital care. One health provider made the important comment that From the health system user surveys, it emerged that the top prior to the establishment of NoPs, facilities focused on 5 factors that influenced user decisions on which facility to meeting their individual targets; however, with NoPs being seek health care from were perceived quality of care, proximity implemented, the focus has shifted from individual facilities to the health facility from home, waiting time at the health to the networks and even to the district as a whole. facility, previous good experience at the health facility, and trust in the provider. NoP districts (Dormaa Central and Hohoe) mostly adhered to NoP guiding principles, such as ensuring equity in Affordability did not emerge as a strong determinant of healthcare, providing good quality of care, collaboration, patient choices, notwithstanding reported concern of and communication, than the non-NoP districts. Use of elevated OOPE by numerous respondents, particularly in Community Scorecards to measure the performance of urban districts. This underscores the patient’s inclination for the health facilities was commonly reported by Hohoe prioritizing quality care over cost considerations, revealing compared to other districts. Hohoe reported regular a willingness to make trade-offs in favor of better quality meetings and workshops at the network facilities, which of care. enabled knowledge sharing. Visits by doctors from higher Urban areas showed a strong preference for hospital-level facilities also supported continuous learning- a feature services, similar to the utilization pattern noted in rural areas. that was not present before the NoP implementation. A significant departure from rural districts was the high (More detailed information can be found in the technical preference for private health facilities (clinics, hospitals) and findings report). 9 RECOMMENDATION DESIGN REPORT KEY FINDINGS FROM THE COMMUNITY ■ Weak referral gatekeeping at lower-tier facilities and a ENGAGEMENT tendency by healthcare providers, including health posts, to directly refer to district hospitals even if the The community engagement exercise was a way to establish health center is strengthened. that the conclusions of the PPA/SNA study are meaningful. ■ Referral processes include using referral notes, calling A return to the communities in which the primary data was referral facility to check the availability of services, collected, helped to delve deeper and identify challenges and, in some emergencies, accompanying the to the implementation of the NoP interventions from patients to the referral facility. additional perspectives, and to surface recommendation to these challenges. This process therefore helped validate, Primary findings not fully validated: invalidate or refine the initial research outputs and generate additional ones. ■ Partial validation of the primary finding that referrals made by health providers are patient-centred and take Major primary PPA/SNA findings which could be confirmed: into account factors such as availability of transport, services at referral facility, etc. It was found that referrals ■ Closeness to home is a significant factor for health were less determined by patient choice and more by facility selection. The additional importance of quality of service capacity considerations of referrer. care and trust in the provider is also corroborated. ■ Could not validate primary finding of high patient ■ The patients prefer higher-tier facilities in specific satisfaction with services. Community engagement circumstances including urgency, need for laboratory/ provided a more mixed picture with less satisfied diagnostics, NHIS cover. care users due to quality and staff behavior concerns ■ Collaboration exists between health facilities in (care shadowing also suggested the mixed quality of service provision. care provided). ■ More awareness about NoP is required in the community. Major new insights gained from the community ■ The main perceived benefits of NoPs are resource engagement: sharing (drugs, other commodities, HR) and collaboration ■ Insurance coverage of services at a health facility between health facilities. drives health facility selection by patients. ■ The long distance/ time taken to reach the facility and ■ Suggestions on how to address health system the long waiting time after reaching the facility are challenges, such as the need for affordable drugs major challenges in the rural setting. and diagnostics closer to the patient, better behavior by ■ Dominant role of private health facilities in urban the health providers, suitable spaces to rest and areas, and urban populations’ preference for ‘clinics’ for care for the patients and accompanying caregivers, healthcare, including MNH care. more information on NoPs shared with the community, and regular training for healthcare staff. Based on the research findings and community suggestions for improvements, the study team formulated a series of key recommendations while also considering Ghana’s existing policies and NoP plans. The recommendations are tailored to address specific challenges and capitalize on opportunities identified. They are crafted to foster meaningful change and promote sustainable development of the NoP program. By incorporating diverse perspectives and evidence-based strategies, the following recommendations can inform future actions and policies. 10 RECOMMENDATION DESIGN REPORT Recommendation 1 – Configure New NoPs by Using a Comprehensive Mapping Approach, which Includes Health Facility Profiles, Perspectives of Healthcare Providers and Users, and Geolocation Rationale The Ghana Health Service aims to form NoPs that encompass the various healthcare providers - and their users - in any given area. NoPs are expected to include both GHS and non-GHS facilities, including faith-based organizations, private maternity homes, community pharmacies, chemical sellers, and other private providers to ensure broad access to an essential package of PHC services. While this collaborative configuration of NoPs is a recent addition to GHS’s vision of NoPs in Ghana, this has not been fully implemented in the districts studied. Pharmacies and chemical sellers were not included in the NoP formation process and there were instances where CHAG facilities were officially NoP spokes but the staff at the CHAG facility was not aware of their role in the NoP or even their membership status. This is reflective of insufficient collaboration at the operational level. Careful network configuration will be especially important in urban areas where healthcare is often accessed via clinics and private health facilities. In Ayawaso Central, more than half (52%) of the MNH respondents chose to visit clinics for their 1st ANC, and 45% for regular ANCs. Across all touchpoints in the HTN pathway, findings were similar- with the two urban districts showing a higher utilization of clinics compared to health centres, unlike in the rural districts, pointing to the need to configure networks differently in urban areas compared to rural districts. Patient-centric care is at the heart of PHC, hence patient experience and health outcomes must be considered in the NoP development process in a district, as well as distances, transport options and safety of roads – which can also be strengthened in line with suitable hubs. Specific actions ■ Conduct detailed stakeholder mapping in sub-districts targeted for NoP implementation, especially to implement in urban areas, to identify formal and informal care providers who should become members of an recommendation NoP, e.g., community pharmacies and chemical shops, smaller clinics and health posts functioning under non-health government departments and ministries, such as school clinics, to foster better collaboration, and promote trust in the formal health system. ■ Use information on transport options, travel times, and distance to inform hub selection, as these are found to be critical in the uptake of referral services. GIS technology can be leveraged to spatially map facility locations, distance, and average travel times. ■ Where more than one health center is present in a sub-district, select hubs with 30 minutes or less travel time to the district hospital to improve outcomes for emergency referrals, especially for mothers and newborns, who would have access to Comprehensive Emergency Obstetric and Newborn Care (CEmONC) services within a reasonably safe timeline2. Follow through the steps proposed by the GHS to collaborate with the district assemblies, communities, district level of the National Health Insurance Authority (NHIA), and at the national level with health development partners, for network configurations3, to enhance participation, interest, unity of purpose and effectiveness of NoPs. Other factors to ■ Patients’ decision-making regarding health facility use is also dependent on the non-health public be considered sector infrastructure, such as the availability of all-weather roads in good condition, and availability of reasonably fared transportation. These should be advocated with concerned ministries in the government and gradually implemented. 2 Roder-DeWan S, Nimako K, Twum-Danso NAY, Amatya A, Langer A, Kruk M. Health system redesign for maternal and newborn survival: rethinking care models to ,close the global equity gap. BMJ Glob Health 2020;5:e002539. doi:10.1136/bmjgh-2020-002539 3 Implementation Guidelines for Networks of Practice. Ghana Health Service. 2024. 11 RECOMMENDATION DESIGN REPORT Recommendation 2 – Prioritize the Further Upgrading of Network Hubs to Reorient Patient Pathways, Decongest Hospitals and Strengthen the Primary Care Levels Rationale The hospital level remains overburdened with patients bypassing the primary care providers. There was a strong preference among patients for higher-tier health facilities in the six study districts. Across most touchpoints of MNH and HTN care, patients preferred seeking care at hospitals rather than at health centers. This was also still essentially the case in NoP sub-districts. Even for ANC and delivery, the MNH respondents preferred visiting more distant hospitals than health centres that were closer. Around 36% of all 1st ANC visits and 52% of the deliveries were at hospitals. Likewise, for HTN, 50% of the patientvws were diagnosed and initiated on treatment in hospitals. CHPS compounds continue to refer to the district hospital directly in many instances, even when the condition can be managed at a health center. Health centers need to be resourced and improved to make them capable of managing conditions at the level of their competency and to make them a preferred point of primary contact for healthcare users. This will help reduce the current burden on district hospitals due to the bypassing of health centers for higher- level care. In the NoP Program, telemedicine and teleconsultations are considered potential components (“spokes”) of the network. Among the six study districts, digital health innovations were limited. Telemedicine services were available only in Hohoe at the District and Zonal Hospitals and Hubs, reportedly improving consultation ease in some cases. Despite its promise, community members expressed skepticism due to unreliable networks and fraud concerns, favoring face-to-face consultations and appointment-based services. Specific actions ■ Equip health centers with a level-specific infrastructure to enhance care delivery and improve health to implement outcomes. Achieving maturity level 3 for health centers, as described by the NoP Implementation recommendation Guidelines, should be prioritized to enhance their patronage and capacity to deliver appropriate care. ■ Establish laboratories in health centres, in line with the NoP implementation guidelines. The unavailability of diagnostics and laboratory testing was reported to be one of the challenges faced by patients and could contribute to patients opting to visit hospitals for primary health care. ■ Ensure equipment and drug availability as per national drug policy and norms, to manage cases effectively. For this, strengthen the drug supply chain, maintain drug consumption records to enable quantification of requirements, and keep buffer stock of essential medicines as per national norms. Ensuring that the integrated logistics management system, GhILMIS, is operational in these facilities would be crucially beneficial for this purpose. ■ Expand telemedicine to all health facilities to provide secondary and tertiary care at the primary care level, saving time and expenses, improving management efficiency, reducing hospital referrals, and decreasing patient load at district hospitals, while contributing to health system decarbonization. Other factors to be ■ Non-health public sector infrastructure enhancements, such as improved communication networks, considered road repairs, should be advocated with concerned ministries in the government and gradually implemented, prioritizing access to NoP hubs. ■ Explore multi-sector coordination through resources available with non-Health departments and Ministries to make transportation available. ■ Telemedicine availability and use are subject to good internet connectivity and broader digitization of the health system, the public sector, and the country as a whole. 12 RECOMMENDATION DESIGN REPORT Recommendation 3 – Invest in the Human Resources Available at Primary Care Facilities to Improve Patient-Provider Contacts and Facility Capacity to Deliver the Full PHC Service Package Rationale One of the keys to good patient-provider interaction is effective, courteous communication and respectful staff behavior. A recurring concern raised by respondents was poor conduct by the health staff, which also influenced the patients’ preferences regarding health facilities. It was observed that nearly half of 156 MNH and HTN patients shadowed in the six study districts encountered instances where the health provider did not listen closely, and was distracted while the patient was speaking, and had no other staff to assist, leading to suboptimal treatment during the consultation. Health providers interviewed at the selected facilities reported a constant HR shortage leading to overburdening of the available staff. This led to many responsibilities being held by one person, and in some instances, a reduced number of services delivered by the facility. For example, outreach work cannot be conducted if the sole nurse or midwife needs to be present at the health facility to serve out-patients. Staff attrition, combined with incomplete handing over and lack of frequent training to sensitize the new staff, could have led to suboptimal performance of some NoPs. The health centres (NoP hubs) can only fulfill their mandate if they have the necessary HR numbers and skill sets within their networks. All healthcare workers need to understand the NoP operations and referral practices. Specific actions ■ Deliver communication, soft skills, and leadership training for all healthcare providers and workers to to implement ensure a better patient experience and improved patient-provider communication. recommendation ■ Recruit more health staff to fill up vacancies as per the National staffing norms, especially physician assistants, community health nurses, and midwives, to support care continuity and reduce hospital waiting times. ■ Adopt strategies and policies to attract and retain workforce in the health system, such as providing incentives for healthcare workers in rural/remote areas, providing financial support in the form of reimbursement of transportation fare, performance-linked bonuses, and recognition (“Star Performer of the Month”) of high-performing workers. ■ Institute capacity building and refresher training programs for all health care providers and workers to deliver their core skills and emergency management. Strengthen collaboration with district hospitals on clinical training for physician assistants at health centres to increase their capacity, in line with GHS recommendation to conduct in-service training activities to build the capacity and competence of NoP staff. ■ Ensure that the Human Resource Information Management System (HRIMS) is regularly updated and utilized for HR allocation planning. Other factors to be ■ A short training module on NoP functioning at the health center level should be developed and considered disseminated through trainings conducted in the districts. ■ Sufficient funding to ensure full complement of well-trained staff. 13 RECOMMENDATION DESIGN REPORT Recommendation 4 – Strengthen the Quality of Care Especially in the Network Hubs Given that Quality Drives Care-Seeking Behavior Rationale Quality of care featured prominently across the study districts as one of the primary drivers for health facility selection by patients. For most of the care touchpoints, HTN respondents gave “quality of service” as the most important reason for choosing a health facility (46% during initial contact, 55% during HTN monitoring visits). The study noted patient inclination to prioritize quality care over cost considerations, revealing a willingness to make trade-offs in favour of better-perceived quality of care. This also explains why patients bypassed health centres which were closer to their homes (but scored lower in perceived quality), and preferred travelling to distant district hospitals while incurring more OOPE. For NoPs and other initiatives in support of universal health coverage, ensuring good quality of care becomes imperative - both in terms of the technical expertise available and the attitude of the health staff towards the patients. Specific actions ■ Strengthen supportive supervision to provide on-the-job skills training and mentorship to staff in PHC to implement facilities in line with GHS guidance to enhance the maturity of health centres and cultivate a quality recommendation culture. ■ Sensitize healthcare workers to adopt positive attitudes towards patients, promoting kindness, passion, and commitment to delivering high-quality healthcare services for all, especially at the health centres, to attract more clients. ■ Empower communities through the Community Score Card initiative to demand quality care from healthcare providers. ■ Institute facility modifications to improve patients’ care experience. This could include making facilities disability-friendly and prioritizing services for persons with disabilities. ■ Provide refresher training and continuing education to staff and mandate the completion of specific high-priority quality-related topics for renewal of facility and individual licenses. Other factors to be A quality culture encompasses a broad set of values, attitudes, behaviors, and organizational norms considered prioritizing continuous improvement, patient safety, and excellence in care. Availability of adequate resources is essential for the necessary inputs for quality services to be in place. 14 RECOMMENDATION DESIGN REPORT Recommendation 5 – Improve NoP Maintenance Efforts Rationale It was found that Networks of Practice implemented in the study districts have generally proven successful, aligning with the initial conceptualization, and adhering to key guiding principles. However, specific areas necessitate attention, including enhanced engagement with CHAG and private facilities, improved monitoring of NoP operations, and continued coordination between network facilities and NoP review meetings. According to study findings, Dormaa Central had lower NoP gains than Hohoe, particularly concerning provider ownership, and monitoring of NoP implementation. Dormaa’s NoPs reported fewer review meetings, a lack of sufficient supervision, and some CHAG facilities lacking awareness of them being part of NoPs. Staff attrition, incomplete handing over and lack of frequent training could explain this suboptimal performance. Digital health interventions are being leveraged globally to improve healthcare. However, in most of the sampled health facilities, even basic technology such as a laptop/computer was unavailable, posing significant hurdles to data management. Incomplete data entry is a big challenge to the smooth implementation of any health program – it prevents regular monitoring and necessary corrective actions, leading to suboptimal performance of the health systems. Having digital connectivity is key to transitioning from paper-based reporting to the more robust forms of digital reporting that can save time, and reduce error, thereby improving the quality of analysis. To ensure that NoPs function optimally and strengthen primary health care, constant efforts are needed to maintain awareness around NoP guidelines, roles and responsibilities, ensure the availability of resources to carry out required NoP activities and introduce more digital solutions. Specific actions ■ Appoint NoP coordinators at each health facility in the network to oversee and take responsibility for to implement smooth operations. recommendation ■ Enhance capacity building, knowledge sharing, and skill development through regular workshops, training, and routine supportive supervision by various health departments. These supervisory visits can include assessments on NoP implementation. ■ Strengthen monitoring and evaluation of NoP activities such as (a) using performance indicators on network-level functions like collaboration activities, quality improvement activities, (b) conducting regular review meetings of NoPs at the district level to keep abreast with the status of NoP implementation, performance monitoring and also as a means to make new staff aware of the NoP status of the health facility, and other aspects this status entails, (c) setting up a reward and recognition system to honor best performing NoPs (rewards could be financial with incremental funding going to all health facilities in the winning NoP, or through publishing names of winning NoPs), (d) assessing performance exclusively on DHMIS data, adopting a ‘no entry, no performance’ standard. This can improve data entry practices and strengthen precise quantification of drugs and commodities, enhance completeness of data, and facilitate timely interventions by higher authorities. ■ Adapt DHIMS to support network-level data capture and monitoring of performance indicators. In addition to basic data on the network members, there could be indicators at the ‘network level’ to track above-facility activities. ■ Increase community involvement through the use of scorecards to seek feedback on the quality of care provided at the health facilities- this can inform continuous improvement of the NoP operations. ■ Equip all health centers with laptops/computers and provide mobile devices with built-in apps to all CHPS-level health providers to ensure smooth and complete data capture. Other factors to be Sustainability of the NoP intervention depends on adequate funding to enable the network members to considered play their role within the NoP and for the network to function. The strong political and bureaucratic will to equip health centres for optimal primary health care is critical to ensure continued motivation in health workers and acceptance by the community. 15   RECOMMENDATION DESIGN REPORT Recommendation 6 – Build Capacity on Patient Referral to Make the System Work Better for Patients and Providers Rationale Streamlined referrals are a key feature of an optimally functioning health system and a crucial link between primary care and higher-level care available at hospitals. In the six study districts of Ghana, referral gatekeeping was found to be weak, with most health post staff continuing to refer to district hospitals rather than health centres. This was also found to be true in NoP districts. It was reported that referral decisions were based on service availability at the referral facilities and were officially recorded using referral notes. However, other factors, such as transportation services and family support at the referred facility, were often overlooked. Low levels of involvement of patients in the decision-making of referral facility lead to dissatisfaction with the referral process among most patients. Some patients also reported a lack of urgency and indifference from healthcare workers during emergencies, leading to delays in referrals and transportation arrangements. The financial burden often fell on patients for transportation, and referrals lacked discussions on the patient’s ability to pay for services at the referred facility. Follow-ups by referring health workers were rare and accompanying patients to referred facilities were uncommon except in highly critical conditions. Participants accepted down-referrals for simpler procedures closer to home to reduce medical transportation costs. They preferred commercial taxis over ambulances due to their high cost. Specific actions ■ Educate health staff on proper referral mechanisms and health services available at the various levels to implement of the health system. recommendation ■ Disseminate clear information on the services included under insurance at each level of care to ensure that patients are referred to facilities where the service for which referral is being made is available under insurance and will not lead to OOPE. ■ Strengthen collaboration with district hospitals on training, especially for physician assistants at health centres, as a strategy to reduce bypassing of health centres and ensure better referral gatekeeping. ■ Expand the National Ambulance Service (NAS) by increasing the number of ambulances at the district level and guarantee fuel provision/ regular maintenance to ensure 100% vehicle availability and optimal response time of vehicle. ■ Enhance collaboration between the NAS and the Community Emergency Transport System (CETS) at the community level to further strengthen the network for transferring patients from home/patient site to facility and inter-facility transfers. Other factors to be ■ Non-health public sector infrastructure enhancements such as improved communication networks, road considered repairs, should be advocated with concerned ministries in the government. ■ Explore multi-sector coordination through resources available with non-health departments and Ministries to make transportation available. ■ Referral gatekeeping can be successful only when the health centres have been sufficiently resourced to serve as a referral centre for the CHPS compounds. System strengthening to ensure adequate diagnostics, medication, and well-trained staff are a prerequisite to improving referral practices and avoiding over-referral to hospitals. 16 RECOMMENDATION DESIGN REPORT Recommendation 7 – Promote Health Equity through Increasing Awareness, Community Engagement, and Financial Protection Rationale Achieving health equity goes beyond mere access to healthcare services. It involves empowering communities through knowledge, involving them in decision-making processes, and ensuring financial mechanisms support equitable access to healthcare. Some of these elements were found to be suboptimal in the study areas. Although certain community members were informed about NoPs within their districts/communities through educational sessions at local health centers, indicating their engagement and potential to improve care quality, general NoP awareness was insufficient. Even among those aware of NoP implementation, utilization rates were suboptimal, suggesting gaps in understanding the comprehensive scope of services offered. Patients across the six study districts exhibited a strong inclination towards higher-tier health facilities. Approximately 72% of mothers surveyed preferred hospitals for deliveries, citing better service access, drug availability, provider professionalism, and avoidance of incorrect prescriptions. Addressing this preference requires raising awareness of local health services and building public trust to encourage the utilization of primary care facilities. Understanding community health needs and preferences and fostering patronage of NoPs is one of the key guiding principles of NoPs in Ghana. Aligned with the CHPS policy, Community Health Management Committees (CHMCs) act as intermediaries between the GHS and local communities to enhance healthcare delivery. CHMCs are pivotal in sensitizing and mobilizing communities to increase awareness and access to services within their sub-districts. A ‘Community Scorecard’ facilitates community leaders’ quarterly feedback on service quality, ensuring accountability and performance tracking of NoP initiatives at the grassroots level. However, the implementation of these strategies varied across the six study districts, with Hohoe reporting a comparatively higher number of community durbars, CHMC meetings, and use of Community Scorecards. A high OOPE incurred by patients was reported to be another concern in accessing healthcare across all districts, especially the two urban districts, Ayawaso Central and Atwima Nwabiagya. However, affordability was not found to be a primary driver of health facility selection, meaning patients were willing to spend more to be able to access good quality healthcare, with 72% of all MNH respondents using savings, 9% borrowing from friends/ relatives and 18% using either NHIS cover, private insurance or money lenders. Specific actions ■ Appoint a dedicated NoP focal person at the district in addition to the NoP hub lead, and to implement NoP coordinators at each spoke in addition to the Hub level NoP coordinator proposed in the recommendation Implementation Guidelines for NoP 2024 to improve NoP implementation awareness through advocacy among the public in each health facility’s catchment area. ■ In line with the guidance provided in the NoP Implementation Guidelines to mobilize and empower communities, leverage the CHMCs to facilitate awareness in the community about health services available at the various levels of the health system and access to services included under insurance. ■ Ensure that all trainings of health providers include a module on community engagement covering strategies to engage the community. ■ Use CHMC meetings for consultations about policy decisions affecting grassroot healthcare provision and access to services. ■ Enhance the DHIMS to include indicators that track critical community engagement activities. ■ Continue to optimize NHIS policies and payment structures to minimize OOPE for patients and prevent catastrophic health expenditures, especially for the urban poor. ■ Establish NHIA-credentialed health centers in areas with critical deficits in access to healthcare to extend a wide range of health services to these communities, thereby reducing OOPE. ■ Encourage private/NGO sector health facilities to be credentialed to provide services under the NHIS, especially in urban areas where there is higher patronage of private facilities, to extend financial protection to patients seeking care in the private sector. (continues on next page) 17   RECOMMENDATION DESIGN REPORT Other factors to be ■ System strengthening to ensure adequate diagnostics, medication, and well-trained health staff are a considered prerequisite to any advocacy and awareness-generating efforts. Advocacy driven health system users at health centres, if faced with poor quality care and lack of crucial medications, will lose trust in the primary care system, a situation that will further push people towards higher-tier facilities. ■ All community engagement interventions are subject to continued funding support from GHS. Recommendation 8 – Design Effective Urban Networks of Practice Rationale So far, the NoP initiative has piloted, implemented and refined networks in rural areas of Ghana with health centres as hubs and lower-level facilities as spokes. This study demonstrated that the facility mix is a different one in urban Ghana with strong private/NGO sector participation in primary care delivery. While maternal and HTN clients still strongly preferred hospitals, even for primary care services, health centre use was low, and use of private clinics/ mission hospitals and private providers was relatively more important. For MNH care in the urban areas, even the CHPS compounds were a more popular choice than health centres, indicating clearly that NoP design in urban areas cannot just replicate the rural blueprint of a model health centre as the hub. Physical access to health facilities was easier in urban areas than in rural districts. Most MNH respondents reached health facilities within 30 minutes, whereas many respondents from rural districts reported travelling for more than 60 minutes. The quick access in urban settings resulted from shorter distances to facilities and easier availability of taxis and private cars to reach them. A key concern of patients was, however, high out-of-pocket expenditure. For both MNH and HTN, the OOPE incurred in urban districts was, on average, higher than that reported in rural districts. For instance, on average 1336 Ghanian cedis OOPE were incurred for childbirth in Atwima Nwabiagya, and 973 Gh. Cedis in Ayawaso Central (vs. 387 Gh. Cedis in Dormaa Central to 816 Gh. Cedis in Ketu North). Specific actions ■ Map and assess health providers and health facilities from different sectors for inclusion in urban NoPs. to implement This would include private clinics and hospitals, mission and other NGO-supported health facilities, recommendation pharmacies and chemical sellers. ■ Promote public sector health facility upgrades to increase the range of services available at these facilities and to come under the National Health Insurance cover. Strengthening CHPS facilities and upgrading existing health centres to ‘Model Health Centres’ will further decentralize primary health care and reduce the out-of-pocket expenditure associated with travelling longer distances or with seeking care at private health facilities. A strategy of selectively upgrading specific CHPS compounds to health centers can also be considered. ■ Collaborate with the private sector through formal mechanisms such as Public-Private Partnerships (PPPs) for healthcare service delivery, MoUs for specific components of health system strengthening such as HR hiring, diagnostics support or use of private ambulance/ transportation services. ■ Implement comprehensive training and mentorship programs while following the GHS guidance for NoPs so that all private facilities staff are well equipped to handle NoP operations, intra-network resource sharing and referrals. ■ Provide HMIS access to all private health facilities being included in NoPs to enable review and monitoring of key performance indicators finalized at the time of entering into a partnership. ■ Implement a learning agenda on NoPs in the urban health system. Other factors to be ■ Successful collaborations with the private health sector to form NoPs depend on the existing trust and considered the willingness of the various stakeholders to collaborate. ■ PPPs necessitate additional processes such as contracting, monitoring and review, and the GHS may need to adopt the role of a payer of services. Hence, the availability of sufficient funds and contracting and fund management capabilities will determine the success of PPPs. ■ Strong advocacy by the GHS is essential to ensure public understanding of PPPs and their potential to enhance the accessibility and quality of healthcare. 18 RECOMMENDATION DESIGN REPORT Conclusion In summary, the recommendations are designed to address key barriers to accessing healthcare, particularly by enhancing affordability and ensuring the availability of essential resources at the facility and network level. By improving these aspects, the goal is to foster a positive shift in health-seeking behaviors among the population toward the primary care level. Central to the success of these efforts is establishing trust with communities, focusing on quality management of health services, and allocating sufficient financial resources. Through robust policy reforms and effective coordination across various sectors, there exists a clear pathway to elevate PHC standards. This comprehensive approach not only aims to strengthen PHC systems but also lay a solid foundation towards achieving UHC in Ghana’s rural and urban districts. Continuous learning and tailoring of the NoP approach to the local context will benefit healthcare users and providers alike. 19   World Bank 1818 H Street, NW Washington DC 20433 USA https://www.worldbank.org