CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA: Understanding the Bottlenecks and Closing the Implementation Gaps © International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Internet: www.worldbank.org; Telephone: 202 473 1000 This work is a product of the staff of The World Bank with external contributions. Note that The World Bank does not necessarily own each component of the content included in this work. The World Bank therefore does not warrant that the use of the content contained in the work will not infringe on the rights of third parties. The risk of claims resulting from such infringement rests solely with you. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or other partner institutions or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 4.0 International (CC BY 4.0) unported license https:// creativecommons.org/licenses/by/4.0/. Under the Creative Commons Attribution license, you are free to copy, distribute and adapt this work, including for commercial purposes, under the following conditions: Attribution – Please cite the work as follows: Care for Hypertension and Other Chronic Conditions In Samoa: Understanding the Bottlenecks and Closing the Implementation Gaps. 2020. Washington DC: World Bank. License: Creative Commons Attribution CC BY 4.0 Translations – If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in its translation. All queries on rights and licenses should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington DC, 20433, USA; fax: 202-522-2625; email: pubrights@worldbank.org. CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA: Understanding the Bottlenecks and Closing the Implementation Gaps World Bank authors: Nicole Fraser-Hurt, Shuo Zhang, Dayo Carol Obure Samoa Ministry of Health authors: Leausa Dr. Take Naseri, Dr. Robert Thomsen, Victoria Ieremia – Faasili, Athena Matalavea Table of Contents Acknowledgements .................................................................................................................................................................. i Preface ........................................................................................................................................................................................ ii Abbreviations............................................................................................................................................................................ iii Glossary of Terms..................................................................................................................................................................... iii Executive Summary................................................................................................................................................................. v I. Introduction.............................................................................................................................................................................. 1 Context of this analysis...................................................................................................................................................... 1 Aim of this study................................................................................................................................................................. 2 II. Methodology......................................................................................................................................................................... 5 The cascade approach..................................................................................................................................................... 5 Data sources........................................................................................................................................................................ 6 III. Study findings on hypertension...................................................................................................................................... 9 Who is hypertensive?........................................................................................................................................................ 9 Are hypertensive individuals getting care?............................................................................................................... 10 What are the gaps and bottlenecks in hypertension care?.................................................................................... 12 a. In primary prevention of hypertension................................................................................................................ 12 b. At the screening stage............................................................................................................................................ 15 c. At the referral and diagnosis stages.................................................................................................................... 18 d. At treatment initiation............................................................................................................................................. 20 e. At treatment monitoring and retention in care................................................................................................. 24 f. Achievement of treatment targets....................................................................................................................... 25 Has PEN Fa’a Samoa impacted hypertension care, and what are patients and providers experiences with PEN?.................................................................................................................................................. 29 IV. Study findings on chronic conditions and their care............................................................................................... 33 Burden of chronic conditions and care seeking behaviours................................................................................ 33 Why is the clinical management of chronic conditions often interrupted?....................................................... 35 Patients suggestions for supporting the management of chronic conditions................................................. 37 Provider suggestions for improving chronic care.................................................................................................... 38 V. Conclusions.......................................................................................................................................................................... 41 VI. Recommendations........................................................................................................................................................... 45 Citations and notes................................................................................................................................................................. 51 APPENDICES Appendix 1. Cascade Study 2018 Household Survey.................................................................................................. 54 Appendix 2. Patient File Review......................................................................................................................................... 68 Appendix 3. Focus Group Discussions ......................................................................................................................... 69 FIGURES Figure ES1. Hypertension cascade in Samoa ................................................................................................................. vi Figure 1. Cascade approach and its data sources ......................................................................................................... 6 Figure 2. Hypertension prevalence in Samoa men and women by characteristics ............................................ 10 Figure 3. Hypertension cascades in Samoa men and women .................................................................................. 12 Figure 4. Body mass index of men and non-pregnant women in Samoa ............................................................... 13 Figure 5. Average waist circumference by age group and gender .......................................................................... 14 Figure 6. Recalled learning themes from PEN Fa’a intervention .............................................................................. 15 Figure 7. History of blood pressure measurements in men and women ................................................................ 17 Figure 8. Coverage by PEN Fa’a screening and reasons for non-coverage ......................................................... 18 Figure 9. Reported outcomes of positive HTN screens and referrals ..................................................................... 19 Figure 10. Treatment recommendations received by people diagnosed with hypertension .......................... 22 Figure 11. Timing of treatment initiation among diagnosed hypertension cases ................................................. 23 Figure 12. Medicine prescription patterns in hypertension patients ....................................................................... 24 Figure 13. Hypertension cascades by villages’ PEN status ....................................................................................... 29 Figure 14. Chronic diseases reported by Samoa men and women ......................................................................... 33 Figure 15. Frequency of clinic visits by survey group .................................................................................................. 34 Figure 16. Care providers treating people’s chronic conditions ............................................................................... 35 Figure 17. Reasons for stopping the medication and treatment interruption ........................................................ 36 Figure 18. Preferred persons outside clinic for supporting medication adherence ............................................ 37 Figure 19. Favoured strategies to improving treatment adherence and outcomes ............................................ 38 Figure 20. Projected causes of pre-mature mortality in Samoa in 2030 ............................................................... 42 TABLES Table 1. Barriers and facilitators in the uptake of hypertension services by Samoans ........................................ vii Table 2. Topics covered in counselling sessions at HTN treatment initiation ....................................................... 21 In Appendix 1: Table 3. Survey response rates by population and village ........................................................................................ 54 Table 4. Demographic characteristics by population and village ........................................................................... 55 Table 5. Socio-economic characteristics of households by village ........................................................................ 60 Table 6. Resting blood pressure by population and village ....................................................................................... 61 Table 7. Prevalence of hypertension by population and village ............................................................................... 63 Table 8. History of blood pressure measurement by population and village ...................................................... 65 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA Acknowledgments This report was a joint effort by the Ministry of Health of Independent State of Samoa and the Health, Nutrition, and Population Global Practice of the World Bank. Ministry of Health Samoa: The study team was led by Leausa Dr. Take Naseri (Director General/Chief Executive Officer, Ministry of Health) and Dr. Robert Thomsen (Deputy Director General, Public Health Ministry of Health) and consisted of Victoria Ieremia - Faasili (Principal, Climate Change & Health) and Athena Matalavea (Senior Information Analyst). The MoH team made essential contributions in developing the household survey instruments, obtaining ethics clearance for the study, and implementing the patient file review in the field as well as analysing the patient-level data. World Bank: The study was led by Shuo Zhang (Senior Health Specialist, Task Team Leader), Dayo Carol Obure (Health Economist, co-TTL) and Nicole Fraser-Hurt (Consultant), and overseen by Marelize Gorgens. Maeva Natacha Betham Vaai (Liaison Officer to Samoa), Sabrina Terry (Team Assistant in World Bank headquarters in Washington DC) and Lynn Ioana Malolua (Team Assistant in World Bank office in Apia, Samoa) provided invaluable coordination and administrative supports to the implementation of the study. Graphic design for the study report was provided by Richard Ponsonby. The report was reviewed by Sutayut Osornprasop, Renzo Sotomayor, and Zara Shubber, who provided valuable peer review comments. A research team consisted of Lutia I Puava Ae Mapu I Fagalele Samoa (LPAMF) & Yale School of Public Health USA carried out house hold survey and focus group discussions in the field and provided excellent supports in analysing the survey data and developing the group discussion transcripts. The team members are Muagatutia Sefuiva Reupena (Director, LPAMF), Vaimoana Lupematisila (Senior Research Assistant, LPAMF); Dr. Nicola Hawley (Assistant Professor of Epidemiology/Chronic Diseases, Yale School of Public Health), Dr. Alysa Pomer (Associate Research Scientist, Yale School of Public Health) and Dr. Anna Rivara (Postdoctoral Research Fellow, Yale School of Public Health). The study team is very grateful to all the respondents in the Cascade Study 2018 Household Survey and the focus group discussants, for sharing their opinions, knowledge and insights with the researchers, and would like to thank all the health care managers and service providers in the participating health facilities for their inputs and support in this study. The study team is very grateful for the funding provided by the Decision and Delivery Science Global Solutions Group of Health, Nutrition, and Population Global Practice of the World Bank in supporting the study and for the valuable guidance provided by David Wilson and Marelize Gorgens throughout the study. All photography courtesy of Samoa Ministry of Health and World Bank. i PREFACE Preface I am pleased to present the ‘Care for Hypertension and Other Chronic Conditions in Samoa: Understanding the Bottlenecks and the Closing the Implementation Gaps Report’. This study was a collaborative effort by the Ministry of Health and the World Bank to identify breakpoints and gaps in the continuum of the hypertension care and to determine issues in chronic care in Samoa and potential solutions. Samoa is privileged to be the first Pacific Island Nation to conduct a cascade study on hypertension and other chronic conditions. The cascade study aimed to fill knowledge gaps on the burden of hypertension in Samoa, how many hypertensive individuals had been screened, referred and diagnosed, initiated with clinical management, retained in care and monitored according to the standard norms, as well as barriers and bottlenecks to NCD service uptake and delivery. The findings from this study will inform the Ministry of Health’s efforts to improve NCD care and the implementation of the Samoa Health System Strengthening Program for Results (PforR). The Ministry of Health expresses its gratitude to the World Bank for providing technical guidance and funding for this study. This has been a challenging yet rewarding experience for the Ministry of Health. The report will inform policy decision making and is crucial in moving forward with the efforts of the Government and its development partners in curbing the rise of NCDs in Samoa. The Government of Samoa appreciates all the efforts that have been made to complete this study and we look forward to implementing the recommendations of this report, in our endeavor to improve the health of Samoan’s through the prevention, management and control of NCDs. Ma le fa’aaloalo lava, Leausa Toleafoa Samau Dr. Take Naseri Director General of Health/Chief Executive Officer Ministry of Health ii CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA Abbreviations BMI Body mass index MOH Ministry of Health BP Blood pressure NCD Non-communicable disease CVD Cardiovascular diseases NHS National Health Service DBP Diastolic blood pressure OOP Out-of-pocket HTN Hypertension PEN Package of Essential NCD interventions IHD Ischemic heart disease PIC Pacific Island Country LMIC Low and middle income country SBP Systolic blood pressure METI Matuaileoo Environment Trust Inc. Glossary of Terms Blood pressure is the pressure of the blood in the circulatory system, measured in millimetres of mercury (mmHg). Blood pressure is recorded with 2 numbers: systolic blood pressure is the force at which your heart pumps blood around your body, diastolic blood pressure is the resistance to the blood flow in the blood vessels. Ideal blood pressure is 90/60mmHg - 120/80mmHg. A reading of 120/80mmHg - 140/90mmHg can indicate that a person is at risk of developing high blood pressure if no steps are taken to bring the blood pressure under control. Body Mass Index is an approximate measure of obesity and is calculated for a person as their weight in kilograms divided by the square of their height in meters. Hypertension means high blood pressure in a person. It rarely has noticeable symptoms but if untreated, increases the risk of serios health problems such as heart attacks and strokes. Hypertension is present if the systolic blood pressure is greater than or equal to 140 mmHg, or the diastolic blood pressure is greater than or equal to 90 mmHg. The degree of hypertension is often expressed in stages. Stage 1 = SBP 140-159 or DBP 90-99, Stage 2 = SBP 160-179 or DBP 100-109, and Stage 3 = SBP 180+ or DBP 110+. Normal weight is defined internationally as a BMI of up to 18.5 and below 25.0. iii GLOSSARY OF TERMS Glossary of Terms CONT/… Overnutrition is the overconsumption of food and nutrients to the point at which health is adversely affected. Overnutrition can develop into obesity, which increases the risk of serious health conditions, including cardiovascular disease, hypertension, cancer, and type-2 diabetes. Overweight is defined as abnormal fat accumulation that presents a risk to health. Overweight is a risk factors for a number of chronic diseases, including diabetes, cardiovascular diseases and cancer. A person with a BMI of 25 to 30 is considered overweight. Obesity is defined as excessive fat accumulation that presents a risk to health. Obesity is now recognized as a chronic or non-communicable disease posing major risk for multiple diseases including diabetes, cardiovascular diseases and cancer. A person with a BMI of 30 or more is generally considered obese. Severe obesity (also called extreme or morbid obesity) begins at a BMI of 40. That is a weight of 109 kg for a person who is 1.65 m tall and a weight of about 130 kg for a person 1.80 m tall. Primary prevention of hypertension includes measures which reduce the likelihood of people to become hypertensive. They involve both population-based approaches and targeted strategies focusing on high- risk individuals. The main components are moderate physical activity; maintenance of normal body weight; limiting alcohol consumption; reduction of sodium intake; maintaining adequate intake of potassium; and consuming a diet rich in fruits, vegetables, and low-fat dairy products and reduced saturated and total fat. Applying these approaches to the general population as a component of public health and clinical practice can help prevent blood pressure from increasing and can help decrease elevated blood pressure levels for those with high normal blood pressure or hypertension. Waist circumference is an approximate index of intra-abdominal fat mass and total body fat. It is measured at the mid-point between the inferior margin of the last rib and the crest of the ilium in the mid-axillary plane. There is an increased risk of metabolic complications for men with a waist circumference ≥102 cm, and women with a waist circumference ≥88 cm. iv CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA Executive Summary  he importance of non-communicable diseases (NCD) for the health status of Samoa’s population cannot be T overstated - NCD causes are responsible for 7 of 10 pre-mature deaths with 3 of 10 due to cardiovascular causes alone1. Many adults are hypertensive while rates of type-2 diabetes and obesity2 are among the highest in the world and on an increasing trend3,4. The 2013 Stepwise Approach to Surveillance (STEPS) survey found that 28.9 percent of the Samoa population are hypertensive, and 24.8 percent have diabetes. In response, the Government of Samoa has made NCD control and people-centered health services a priority in its Strategy for the Development of Samoa 2016/17-2019/20 and issued the National NCD Policy 2019-20235. In 2015, it contextualized and piloted WHO’s Package of Essential Non-communicable diseases (PEN) interventions and collected implementation experiences6. The PEN Fa’a Samoa is oriented toward community participation and outreach services. It places emphasis on early detection of NCDs, effective referral and increasing population awareness of NCD risk factors. However four years after initiating PEN Fa’ a Samoa program, it had only been rolled out to 17 out of total 431 villages in the country indicating the stagnant status of the screening. A World Bank study of NCD cost analysis in 2017 flagged low hypertension (HTN) treatment levels in Samoa as an issue of concern7. The widespread HTN and the significant cardiovascular disease burden and high premature deaths suggest that the continuum of HTN care is sub-optimal in Samoa. To support the Government in improving NCD care, the World Bank joined the Ministry of Health (MOH) in 2018 to conduct a study on HTN and chronic care. The primary aim was to identify breakpoints and gaps in the continuum of HTN care and determine issues in chronic care and propose potential solutions. The study findings inform Samoa’s policy formulation, intervention implementation and the World Bank’s Samoa NCD control project. As the first in the Pacific Region to do an implementation cascade study, other Pacific Island Countries (PICs) may also draw on the findings of the study. Considering that service delivery for chronic conditions often happens over multiple contacts between providers and users, the analysis used the “cascade framework” to investigate the continuum of hypertension care. The burden of HTN was estimated and answers sought to questions of how many hypertensive individuals had been screened, referred and diagnosed, initiated with clinical management, retained in care and monitored according to the norms. The endpoint of interest, documented blood pressure (BP) control while on treatment, was assessed for male and female patients, and by diabetes status and history of PEN exposure. The barriers and bottlenecks to service uptake and delivery were then determined among HTN cases and care providers as a step towards intervening to improve the chronic care. We sought to answer the following key questions based on the cascade analysis framework: 1. What is the total burden of hypertension, diagnosed and undiagnosed? 2. Of these, how many have been screened? 3. Of those with a positive screen as the high risk group, how many have been referred and diagnosed? 4. Following diagnosis, how many hypertension cases start treatment? 5. Do hypertension patients get monitored as per the norm while on treatment? 6. Are patients retained in care? 7. Do the patients achieve blood pressure control? 8. What are the barriers and facilitators affecting care cascade? v EXECUTIVE SUMMARY A household-based questionnaire survey with BP and anthropometry measurements was carried out on 1,207 participants aged ≥20 years from two PEN (where the PEN Fa’ a Samoa program has been Implemented) and two non-PEN villages (‘Cascade Study 2018 Household Survey’). The survey implementer, Lutia I Puava Ae Mapu I Fagalele/Yale School of Public Health, followed up with three focus group discussions (FGDs) to find out more about the causes of the gaps in the HTN care cascade and get perspectives from HTN patients and health providers. To learn about the long-term care provided to HTN patients, the MOH conducted a patient file review of 300 patients registered as cases since at least one year at one of the three health facilities serving the survey villages. Among the key findings were: Hypertension prevalence The Cascade Study 2018 Survey found that: • The prevalence of HTN in Samoan adults was 38.1% and most of these individuals had elevated BP while a minority had normal BP and were on anti-hypertensive medication • Men had higher HTN prevalence (39.9%) than women (36.2%), but the difference was not statistically significant (p=0.207). • The prevalence of elevated BP (35.6%) was similar to the level reported from the PEN Fa’a assessment on adults ≥18 years in 2015 (32.7%); it was higher than the prevalence reported from the 2013 STEPS survey but there were several methodological differences. Overall, it appears that HTN prevalence is still on the rise in Samoa, given the three data points from 2013 to 2018 of 28.9%, 32.7% and 35.6%. • The correlation between HTN and age was much steeper in women, with the oldest age group (≥70 years) having five times higher HTN prevalence than those aged 20-39 (in men, this factor was only two times) • In line with the pathology of diabetes, HTN was much more frequent in the 78 survey participants reporting to have a diagnosis of diabetes (63% HTN prevalence in diabetics) Risk factors for hypertension The PEN assessments in 2015 had identified high alcohol use and smoking in men, and low levels of physical activity in men and women as chief contributors to HTN in Samoa, akin to other PICs8. The 2013 STEPS survey reported high levels of obesity highlighting that overnutrition is a significant public health problem in Samoa. The Cascade Study 2018 Survey had the following findings: • Persons with high body mass index (BMI) were more likely to be hypertensive, whereby the BMI effect was stronger in men with HTN levels twice as high in obese versus normal-weight men • The survey confirmed the dramatic obesity epidemic with nine of 10 survey participants classified as overweight or obese (comparison with 2013 STEPS data suggest recent increases in the prevalence of obesity of about 9 percentage points in men and 8 percentage points in women over the five years) Women had especially high rates of severe obesity with one in 4 with a BMI ≥40 putting them at very high risk • of HTN, diabetes, and other preventable conditions • Women in all age groups and men in all groups above 40 years had average waist circumference measurements putting them well into the danger zone of metabolic complications In the two PEN villages, significantly fewer men were obese compared to non-PEN men, and the average waist • circumference was ~3cm smaller in PEN men compared to non-PEN men. In the PEN villages, 91% of survey respondents recalled learning about causes of HTN, especially foods high in sugar, fat and salt, but it was not possible to attribute these weight differences to an effect of PEN. vi CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA Regarding perceived hypertension risk, the survey found that Samoans are aware of the HTN problem, but that • primary prevention interventions need to become more effective to reduce risk behaviours fuelling the HTN and obesity epidemics. The review of patient data demonstrated that despite the population’s HTN awareness, care seeking is late: Among the 300 HTN patients reviewed, 36% had stage 2 and 35% stage 3 HTN at registration, suggesting late • diagnosis and acute cardiovascular risks.  elayed care seeking is a key contributor and needs to be addressed to prevent pre-mature deaths and •D comorbidities. Implementation cascades for hypertension Among hypertensive Samoan adults: Figure ES1 shows the hypertension cascade for Samoa adults. • 68% of men and 80% of women had ever had a BP screened for identifying HTN • 52% of men and 72% of women had a BP screen in the last 12 months • 13% of men and 38% of women had had a diagnosis of HTN by a health care provider • 10% of men and 29% of women had initiated treatment for HTN • 8% of men and 23% of women on HTN treatment had a monitoring BP check in the last 90 days • 2.5% of men and 7.2% of women had evidence of reaching the BP target while on HTN treatment  amoa’s hypertension implementation cascade clearly has multiple gaps, leading to very poor attainment of S the final outcome: BP control. A very low percentage of the hypertensive population is under regular NCD management and the majority of patients in care fail their treatment target. The significantly better screening coverage in women translated into higher linkage to care and clinical management levels, with hypertensive women three times more likely to have started HTN treatment, and three times more likely to have successfully controlled BP, compared to hypertensive men. FIGURE ES1. HYPERTENSION CASCADE IN SAMOA 100% 73.9% 61.4% 24.9% 19.2% 15.1% 4.7% Hypertension Ever screened Screened last Diagnosed On treatment Monitoring BP BP control 12 months last 90 days Sources: Cascade Study 2018 Household Survey, World Bank; Patient file data extraction Nov-Dec 2018 by MOH. Notes: 100% = All individuals classified as hypertensive in household survey based on BP result or reported HTN treatment; Ever screened= Reporting in survey to ever have measured BP; Screened in last 12 months = Reporting in survey to have undergone BP testing in last 12 months; Diagnosed = Told by a doctor or nurse about having high BP (reported in survey); On treatment = Reporting in survey to take HTN medicine; Monitored BP last 90 days = Using proportion of HTN patients with documented BP monitoring while in HTN care at clinic; BP control = Using proportion of HTN patients reaching BP target while in HTN care at clinic. vii EXECUTIVE SUMMARY Among hypertensive adults in PEN vs. non-PEN villages: • 79% of PEN vs. 69% of non-PEN hypertensives had a BP screen in the last 12 months • 21% of PEN vs. 17% of non-PEN hypertensives had initiated treatment for HTN • 6.4% of PEN vs. 6.7% of non-PEN hypertensives on treatment had evidence of reaching the BP target PEN provided positive results on screening, but did not succeed in improving the final outcome of hypertension treatment success. Significantly more PEN village residents had ever been screened for HTN compared to non-PEN residents (p=0.021), and although the peak of PEN Fa’a Samoa implementation was in 2015, more BP screening was still going on in PEN villages (p=0.057). This translated to slightly better diagnosis and treatment levels in PEN areas, but the final outcome of BP control was poor in all villages regardless of PEN history. Barrier and facilitators affecting the implementation cascade for hypertension The interviews in the Cascade Study 2018 Household Survey and the FGDs provided insight into people’s views and experiences on factors affecting their continuity of HTN care (Table 1). TABLE 1. BARRIERS AND FACILITATORS IN THE UPTAKE OF HYPERTENSION SERVICES BY SAMOANS Stage Barriers identified Facilitators identified Primary •U nhealthy food and nutrition habits • General education level prevention of • High levels of stress • Lower BMI HTN •C ultural pressures and obligations • Absence of diabetes • Lack of physical exercise •Non-sedentary occupation such as •H TN as a normalized condition farming •Belief that HTN risk is controllable by own actions Screening & • Fear of diagnosis • Use of antenatal care services diagnosis • Denial of a positive screen •Living in a PEN village (more • Refusal of accept illness screening, better referral) • Not feeling at risk of HTN •Understanding the importance •N ot being aware of screening of following a referral after a positive service screen •L ow ability to pay for health care •P reference for traditional healing practices Treatment start  ow ability to pay for medication or •L • Belief that medication works transport to the health facility  ack of understanding of the •L asymptomatic condition needing treatment viii CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA TABLE 1. BARRIERS AND FACILITATORS IN THE UPTAKE OF HYPERTENSION SERVICES BY SAMOANS cont/… Stage Barriers identified Facilitators identified Treatment  ow ability to pay for recurrent care •L • Convenient access to medication maintenance & costs • Belief that the medication works monitoring nsufficient patient education about •I •Good understanding of HTN condition importance of clinical management •Family support in treatment adherence  ailing to obtain drugs due to •F •Successful use of reminder strategies stock-outs (alarm clocks, linking pill-taking to  ong waiting times at health •L activities like meals, also SMS, social facilities media, wall calendars, health care  egative staff attitudes towards •N visits, pill boxes patients  ack of understanding that •L condition and treatment are life- long  orgetting to take pills or running •F out  oor grasp of medication regimens •P  onflicting treatment advice as re- •C cord system fails to provide patient history  oor communication of treatment •P monitoring results Source: Cascade Study 2018 Household Survey and focus group discussions, World Bank People’s views on barriers to successful hypertension treatment were mirrored by care providers’ opinions on service delivery bottlenecks. Problems with human resources (staffing at the nursing and clinician level) were a key concern of providers - staff shortages impacted their ability to identify HTN cases, counsel and treat them, do community outreach, or provide health education services necessary for prevention. Also, lack of physical resources was a shared experience of providers as maintenance of existing equipment and ordering of new equipment are challenging. Budget cuts were given as a reason for receiving low quality equipment. Many providers talked about prior experiences where they were able to offer more programming in prevention, but all agreed that the pressure on health services was much greater now than in prior years. Some providers felt that patients needed education to ensure they utilized the correct services. Patients and providers spoke about the government-supported Matuaileoo Environment Trust Inc (METI) program promoting a whole food, plant-based diet. METI does village outreach visits, organizes seminars, and published testimonies of patients who have reversed their HTN or diabetes through dietary changes, sometimes leading to discontinuation of the medication. Quality of hypertension care  The standardized NCD case management should entail both pharmacologic treatment and behaviour interventions, but the study found that patient advice focussed on medication, and advice on nutrition or weight-loss was less common. About a quarter of diagnosed HTN cases did not receive counselling when commencing treatment, and there was evidence that for many counselling was a one-off service despite the chronicity of HTN and poor BP control. As patients recalled the content of their HTN education, several topics ix EXECUTIVE SUMMARY seemed poorly covered (e.g. medication schedule, long-term adherence, managing missing doses, side effects, switching regimens). There was little focus on real-life challenges and practical tips to master these and prevent treatment interruptions. W  hile most HTN patients started treatment soon after diagnosis, a minority had never commenced treatment at all. Overall, the timing of treatment start seemed adequate. Prescribed medicine came from four classes of HTN drugs with ACE inhibitors being most frequent, especially in diabetics. There were also three types of supportive drugs prescribed (aspirin, statins, beta-blockers). Only 3% had five or more of these seven drug types prescribed. The level of “polypharmacy” was therefore very low, perhaps reflecting the under-treatment already highlighted in the 2017 National NCD Cost Analysis. NCD case management protocols were not adhered to. HTN patients should be followed up at least once every quarter, but their files suggested that patient monitoring was insufficient with only 61% of HTN patients having evidence of a BP check within 90 days, and only 60% having glucose and cholesterol checked within 12 months. Co-morbid HTN/diabetes had even lower 90-day BP monitoring coverage (51%) but slightly better lab test coverage (68%). Overall, only 13% of the HTN patients had their body weight recorded when put onto the Hypertension Register, and body weight was poorly monitored during follow-up with 93% of HTN patients lacking any record of bodyweight in the past 12 months. BMI calculation was not possible in the majority of cases, as only 38% of patients had height and weight recorded. It therefore appeared that there was no systematic tracking of weight or BMI for clinical management of HTN. The study team could not rule out that measurements had been taken but not recorded in the patient file.  etention in care over the last six months was also insufficient. Only 54% of the HTN patients had evidence of a R facility visit with BP check during the past six months. There was no gender difference, but there were large differences between the three health facilities (Faleolo HC 40%, Lufilufi HC 45%, and Sataua DH 78%). Therefore, many patients show signs of having dropped out of care, or at least not being actively cared for by their health facility.  he sub-optimal clinical management of HTN culminated in poor BP control, as already stated earlier. Among those T without a diabetes diagnosis, 34% achieved the BP target of <140/90 mmHg, but among co-morbid HTN/diabetes patients, only 9% achieved the BP target for diabetics of <130/80 mmHg. While there are many reasons for patients to fail their BP targets (Table 1), the non-availability of anti-hypertensive drugs at clinics was highlighted as an important supply side factor, as was patients’ poor understanding of HTN management. This was responsible for patients stopping the medication once BP readings returned to normal, or when symptoms disappeared. Observations about PEN Fa’ a Samoa experiences  Most of the female respondents were familiar with the PEN program and expressed appreciation about the training of facilitators and getting care closer to home. Their better knowledge about PEN was likely due to the PEN’s implementation through Village Women’s Committees. Many respondents felt PEN had been under- resourced and hinged on unpaid community-based work. They also commented negatively about the MOH not giving feedback about PEN results. Men were less familiar with PEN but expressed a desire to see the program return to their village. In case of PEN continuation, the women suggested PEN is combined with METI and that the PEN intervention also addresses HTN treatment, monitoring and retention in care, and not only screening. Care providers expressed appreciation for the underlying goals of the program, believing that if implemented correctly it can help the healthcare system to support community outreach. They questioned PEN’s sustainability and attributed the failure to fully launch PEN to insufficient investment in training, human resources, equipment x CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA and data recording systems, resulting in relatively small reach of PEN. Providers expressed enthusiasm for re- starting or expanding the program, but highlighted the need for PEN to better align and coordinate with health facilities. Burden of chronic conditions and care seeking behaviours In the Cascade Study 2018 Household Survey, 29% of women and 16% of men said they have at least one chronic disease or condition, which was most often high BP or diabetes, especially in women (who have higher diagnosis rates). Some of the other conditions respondents had, such as heart disease, eye problems, and kidney disease, may be complications of undiagnosed, untreated or uncontrolled HTN and diabetes. Another nutrition-related disease reported by 2.3% of men was gout. Many conditions were reported by just a few respondents, e.g. injuries, joint pain, headaches, mental health conditions (nobody reported gastric ulcers, rheumatic fever or cancer), and some of the low frequencies may have been due to under-reporting during the survey interview.  lmost three-quarters of chronic patients were mainly looked after by doctors, and about 1 in 10 mainly received A care from nurses. Only 6% had a dietician involved in their care, and few got care from traditional healers or counsellors. Their chronic care cost patients on average WST 50 (US$ 20) per month. Similar to HTN treatment, chronic patients stopped medical treatment due to problems accessing medications and getting timely medication refills, feeling better or having normal test results, not prioritizing health and wellbeing, and opting for alternative treatment approaches.  he study team discussed possible solutions to the poor continuity of HTN care with patients and providers. T Suggestions concerning health facility capacity to provide efficient, flexible and friendly NCD services received a lot of support. Better patient information was also welcomed by many individuals, including more counselling at treatment start. Several strategies for supporting patients to maintain treatment and navigate challenges were also favoured, among them phone reminders and community support groups, as well as convenient drug refill options. Interestingly, participants from PEN villages were consistently more likely than those from non-PEN villages to welcome ideas on adherence support, and were more familiar with use of community health workers. This may be due to the better NCD awareness in PEN villages as the intervention had sensitized the community to the issues around NCD care. Recommendations (see report section VI for full recommendations):  cale-up public awareness communications on nutrition, HTN and diabetes, by using a mix of communication 1. S channels, harmonized and evidence-informed messages, and branding strategies, and linking demand creation for screening to screening provision. Address pressures and obligations specific to Samoa’s culture, which cause stress and contribute to the NCD epidemic. Support and enhance these interventions with robust taxation policies which reduce NCD risks and curb unhealthy diets in the population. mprove PHC-level chronic care provision by redesigning how services are delivered, making better use 2. I of outreach, triage, decentralisation and task shifting, drawing from the PEN Fa’a approach - so that chronic care is coordinated and delivered by PHC workers and through training and facility-level optimisations gains efficiency, attractiveness and impact.Concrete task shifting interventions concern the expanded role of nurses in outreach, the village women committees providing health promotion and patient follow-up, as well as the creation of multidisciplinary care teams in the rural hospitals including physicians stationed in the rural district hospitals. Needs-based investment in equipment such as automated (digital) monitors is essential to improve the quality of HTN care. xi EXECUTIVE SUMMARY  nsure the patient-level medical record system is more effectively used to support frontline staff in 3. E providing continued and coordinated chronic care to patients, by maximising the patient interaction through joint treatment planning and phone-based communication, and by using data dashboards, lists, exception reports, and other functionalities for pro-active data use in patient management. The patient experience should also be improved by selectively decentralizing point-of-care tests to rural district hospitals, with results recorded in the patient-level medical record system for quality patient management. Scale up PEN Fa’a Samoa to more villages to expand the screening and early detection of cases. Leverage 4.  the comparative advantages of village-based groups and individuals to improve the effectiveness of NCD screening, referral, diagnosis and treatment, by expanding the community aspects of PEN Fa’a Samoa that worked well, broadening the response with non-health actors, and directing effort to closing the gaps in the continuum of HTN care including screening for NCDs. Adaptation work of the CVD risk chart, based on PEN Fa’a Samoa pilot results, should be completed so care efforts can be targeted to individuals at greatest risk of CVD events.  ystematically address the bottlenecks and barriers to NCD treatment adherence: Improve access and 5. S availability of medications; Strengthen provider understanding of barriers to adherence; and Ensure the provision of patient-centered adherence support – this requires building better knowledge among health personnel on continuity of and barriers to NCD care, and how to advise patients on practical issues helping with treatment adherence, while also trialing new approaches. Several measures should be taken to improve access to essential medicines (facility-specific essential NCD drug list, drug supply planning with proper forecasting, accountability mechanism for drug management and supply planning, enhancing the compliance with standard operating procedures). Clinical NCD guidelines must be aligned with WHO recommendations.  nsure body weight monitoring and the management of overweight and obesity is an integral part of 6. E hypertension and diabetes care, by systematizing weight measurements, patient feedback, and clinical action, and increasing the number and capacity of dieticians and nutrition workers. NCD care guidelines and standard operating procedures must integrate weight monitoring and active management. xii INTRODUCTION I. Introduction CONTEXT OF THIS ANALYSIS For non-communicable disease (NCD) control to succeed, people must be effectively targeted by primary and secondary prevention so that their risk of disease and complications can be minimized. High blood pressure (BP), high plasma glucose, obesity, tobacco and alcohol abuse are the leading risk factors for numerous NCDs and thus the global focus of primary prevention. Once people have conditions like hypertension (HTN) and diabetes, they need to be successfully diagnosed, treated and monitored (secondary prevention), which will reduce the occurrence of clinical diseases, co-morbidities, and pre-mature deaths. In general, NCD patients must remain in care for long periods of time. Service delivery for chronic conditions generally happens over multiple contacts between providers and users, requiring life time continuity of care. NCD patients need to benefit from early clinical management, adhere to it and have regular check-ups to ascertain the achievement of treatment targets. In the case of HTN, treatment is often lifelong and requires consistent medication adherence to maintain normal BP levels. Other measures like weight loss, stress management and physical activity can enhance clinical outcomes and further reduce cardio-vascular disease (CVD) as well as diabetes incidence. Samoa, classified as an upper middle-income country with a gross national income (GNI) per capita of US$4,120 in 2018, is a small Polynesian island state located in the South Pacific. The population of Samoa is approximately 196,000 people distributed between the two main islands (Upolu and Savaii) and two smaller islands (Apolima and Manono). Several studies point to the large and growing burden of NCDs in Samoa where 68% of pre-mature mortality is attributed to NCDs9. The 2013 STEPS survey reported that 29% of the adult population is hypertensive and 25% have diabeteand alarmingly, the rates have been rising The National NCD Cost Analysis Study found comparatively low use of essential HTN drugs implying serious under-treatment or sub-optimal treatment of cases11. Studies on HTN treatment outcomes in other countries have indicated that adherence to HTN treatment is sub- optimal and poses a serious threat to the long-term success of the health sector response to high BP levels in populations12,13,14. This observation of poor adherence to long-term treatments extends to other chronic conditions, including diabetes, and calls for the strengthening of chronic care models in general. While Samoa is facing significant challenges of rising NCDs, the current health service delivery system is not well poised to tackle the challenges. The healthcare service delivery system is largely publicly-owned and heavily hospital-centric, with patients bypassing primary health care (PHC) and overcrowding the higher level health facilities. There are two referral hospitals: the main National Referral Hospital in Apia, Upolu, and another referral hospital on Savaii Island. There are 11 rural health facilities comprised of 6 rural district hospitals (3 on Upolu and 3 on Savaii), and 5 community health centers (3 on Upolu and 2 on Savaii). The allocation of resources is skewed towards the national hospital, with the PHC facilities largely under-resourced and under-staffed. Staff numbers in rural hospitals are insufficient to cater for the increasing population. Doctors are concentrated in the main referral hospital in Apia. The other 11 health facilities are staffed mainly by nurses, with doctors from the main referral hospital visiting the district hospitals one day per week or less. The doctors bring in the medicines when they visit the rural district hospital and carry back the laboratory samples to the main hospital for testing and final diagnosis. 1 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA In response to the NCD crisis, the Government of Samoa, in its Strategy for the Development of Samoa 2016/17- 2019/20, has prioritized inclusive, people centred health services with emphasis on NCD control and management15. The National Health Sector Plan 2008-2018 emphasized prevention and primary care within the health systems strengthening efforts16. The National NCD Policy 2019-2023 aims at decreasing the incidence of NCDs and reducing complications and premature mortality due to NCDs. The strategy draws on experiences with PEN Fa’ a Samoa implementation (Box 1). BOX 1 THE PACKAGE OF ESSENTIAL NON-COMMUNICABLE DISEASE (PEN) INTERVENTIONS IN SAMOA Samoa adopted the WHO NCD intervention package in 2013 . The aims of the Fa’a Samoa (‘Samoan way of life’) version of the package are to address key issues in health system delivery, in particular integrating community participation and village outreach services to: (i) ensure early detection of NCDs in those at risk; (ii) establish mechanisms for referring the high risk population to rural district health facilities for treatment and follow-up; and (iii) increase awareness of risk factors18. Several villages from the islands of Upolu and Savai’i were selected for participation in a pilot phase. The community based NCD screening occurred from February to March 2015. One outcome was that by being made aware that some NCDs have no symptoms in their early stages, the cultural belief that illness is only present when a person feels ill was overcome in the intervention villages. The main lessons learnt (cited from: Bollars et al 2018) were:  he Fa’a Samoa package strengthened links between health services and their communities through •T application of a community-focused, participatory approach.  roviding the screening results to the community increased the understanding of the risks of such •P diseases and their asymptomatic nature on the early stages.  ata management and follow-up of people referred to the health-care facility could be improved to •D monitor progress of the intervention over time. To support the Ministry of Health identify opportunities to improve the continuum of care for NCDs in Samoa, the World Bank, in collaboration with the Samoa Ministry of Health and a research partner, conducted a hypertension implementation cascade study. HTN was used as a tracer condition for NCDs and for the continuum of long-term care. Successful management of HTN is a priority for the health system, given cardiovascular diseases account for 29% of premature death in Samoa according to the Global Burden of Disease estimations. AIM OF THIS STUDY The primary aim of this study was to identify breakpoints and gaps in the continuum of care for HTN in Samoa and to reflect on the provision of services for chronic NCDs. The study determined the status of men, non- pregnant women, and women with current/recent pregnancy in terms of HTN screening, diagnosis, treatment, BP monitoring, and treatment success. This was carried out in order to inform the design of effective NCD prevention and control interventions in Samoa and contribute to better decision making and more effective resource allocation. The study also informed the design of the World Bank financed health project which aims to support systematic 2 INTRODUCTION NCD prevention and control in the country over five years, from 2020-2025. To learn from previous implementation, the analysis reflected on the experiences of PEN Fa’a Samoa pilot and compared HTN care in PEN and non-PEN villages. Being the first in the Pacific to conduct an implementation cascade study, the results from this study in Samoa can be used to inform the design of NCD control interventions in other PICs. The report introduces the “cascade framework” and study design in Section II, and summarises the data sources and sampling. Section III provides the results of the implementation cascade study by answering key questions about HTN care in Samoa. Section IV presents findings on chronic conditions and their care. Section V offers conclusions and Section VI presents the study team’s recommendations. Additional data tables and details on methodology are provided in appendices. 3 METHODOLOGY II. Methodology THE CASCADE APPROACH Service delivery for chronic conditions often happens over multiple contacts between providers and users. When continuity is maintained across these contacts, health outcomes improve, and user satisfaction is higher. Identifying bottlenecks and gaps along the service continuum which prevent people from continuity of care, is an essential step towards intervening to improve the management of chronic conditions. The “cascade framework” provides a systematic way of identifying these bottlenecks. For HTN, the cascade of service implementation begins with identifying undiagnosed individuals through screening and diagnosis. The framework provides a lens to examine if HTN is diagnosed, appropriate treatment is initiated, the patient is followed up and compliant with care, and BP control is achieved. While the cascade framework is useful for understanding retention in care at the population level, the cascade may not necessarily be linear for the individual patient. For example, a patient that had attained BP control, may cease to adhere to clinical management, and regress on the HTN care cascade. In line with the cascade framework, we sought to answer the following questions: 1. What is the total burden of hypertension, diagnosed and undiagnosed? 2. Of these, how many have been screened? 3. Of those with a positive screen as the high risk group, how many have been referred and diagnosed? 4. Following diagnosis, how many hypertension cases start treatment? 5. Do hypertension patients get monitored as per the norm while on treatment? 6. Are patients retained in care? 7. Do the patients achieve blood pressure control? 8. What are the barriers and facilitators affecting care cascade? We used a mix of methods, consisting of a household survey with measurement of BP and anthropometry, a secondary data analysis of HTN patient data from health facilities on the case management, and focus group discussions (FGDs) with HTN cases and health care providers, to study the continuum of HTN care in Samoa (Figure 1). The relevant literature was also used for comparisons and context. The Cascade Study 2018 Household Survey and the FGDs were conducted by a service provider (Lutia I Puava Ae Mapu I Fagalele [LPAMF]/ Yale School of Public Health) while the patient file data were extracted by personnel from the MOH. 5 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA FIGURE 1 CASCADE APPROACH AND ITS DATA SOURCES HYPERTENSION IMPLEMENTATION CASCADE Estimated HTN BP screened Diagnosed HTN treatment BP monitored BP controlled burden initiated according to the norm Patient file data extraction Household survey with BP measurement 2. FGDs with HTN patients and care providers Source: Authors’ drawing DATA SOURCES Cascade Study 2018 Household Survey in two PEN and two non-PEN villages – We conducted a household survey in four rural villages19 i.e. Mulifanua (PEN) and Fagaloa (non-PEN) on Upolu Island, and in Neiafu (PEN) and Vaisala (non-PEN) on Savai’i Island in November-December 2018. These villages on each of the two islands were selected based on: a) their population size to ensure sufficient adult survey participants in a maximum of households to prevent clustering of observations, b) whether they had been part of the PEN pilot or not, and c) geographical location (Upolu/Savai’i). Survey participants aged 20 and above were randomly selected from a household list. Consenting participants were interviewed and BP, height, body weight and waist circumference were measured. The household survey provided data on BP levels from 554 men, 555 non-pregnant women and 98 women with current (41) or recent (57) pregnancy in the past 12 months - for additional data tables from household survey, see Appendix 1 (Table 3 on response rates, Table 4 on demographic characteristics, and Table 5 on socio-economic characteristics). The prevalence of HTN in the study population was determined through a combination of questions asked during the survey interview, and the physical measurements of systolic and diastolic BP. Persons who met any of the following three criteria were considered “hypertensive”: a) Systolic BP greater than or equal to 140 mmHg; b) Diastolic BP is greater than or equal to 90 mmHg; or c) Reporting the use of oral medication for high BP either taken daily or not every day. 6 METHODOLOGY Medical files of registered HTN patients in three health facilities – Existing routine data was extracted by Ministry of Health officials at the three rural health facilities serving the survey villages in November 2018. 100 individuals were randomly selected from each of the three Hypertension Registers in Faleolo Health Centre, Lufilufi Health Centre and Sataua Rural District Hospital. See Appendix 2 for details. Focus group discussions with community members diagnosed with hypertension, and health care providers – Three focus groups were conducted in February 2019: (1) women with hypertension (Mulifanua; patients), (2) men with hypertension (Mulifanua; patients), (3) health facility staff (providers). See Appendix 3 for implementation notes. The data collection was described in a study protocol. The protocol, data collection instruments and informed consent forms for study participants were reviewed and cleared by the Human Research Committee of the Samoa Ministry of Health. 7 STUDY FINDINGS ON HYPERTENSION III. Study Findings on Hypertension WHO IS HYPERTENSIVE? Overall, 38.1% of the participants in the Cascade Study 2018 Household Survey were hypertensive, of which 2.5% had normal BP but reported taking HTN medicine (elevated BP was therefore found in 35.6% of participants). The HTN prevalence for men was 39.9%, slightly higher than the prevalence in women at 36.2% (p=0.21) – see Figure 2 (with detailed statistics provided in Tables 6 and 7 of Appendix 1). Among the women with current or recent pregnancy, HTN prevalence was only at 5.1% but they were much • younger (median age 29 years as opposed to 47 years for men and non-pregnant women) As expected, HTN levels increased with age, whereby age had a stronger effect on BP in women than men • Also as expected, the risk for HTN was higher in persons with high BMI, and the BMI effect was stronger • in men where HTN levels were double in obese versus normal-weight men In line with the pathology of diabetes, HTN was much more frequent in the 78 survey participants • reporting to have a diagnosis of diabetes (63% HTN prevalence) Higher education appeared to correlate with lower HTN levels. The difference in HTN prevalence in female • secondary school versus university graduates was smaller (4 percentage points) compared to men (9 percentage points). There was no clear pattern between household income and HTN risk in women, but in men higher income • was associated with lower HTN rates (however, this was an effect of age with younger men tending to earn more) There were no significant differences in HTN prevalence based on a village’s PEN status •  he 35.6% prevalence of elevated BP was close to the 32.7% reported from the PEN Fa’a assessment •T in 2015 on 1,536 adults aged 18+ years20. For a comparison with the 2013 STEPS survey and discussion of methodological difference: see endnote21.  verall, it appears that HTN prevalence is still on the rise in Samoa, given the evidence from surveys •O over the period 2013-2018 of 28.9%, 32.7% and 35.6%. 9 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA FIGURE 2. HYPERTENSION PREVALENCE IN SAMOA MEN AND WOMEN BY CHARACTERISTICS Prevalence of hypertension: Men 56.9% 54.5% 50.6% 48.1% 42.5% 45.9% 44.3% 42.5% 39.9% 39.9% 40.9% 39.3% 36.9% 33.7% 33.3% 26.7% 24.7% Total (554) PEN Mulifanua (153) PEN Neiafu (141) Fagaloa (154) Vaisala (106) 20-39 years 40-59 years 60-69 years 70+ years Primary Secondary compl Coll/Univ comp 25 to <30 (overweight) Known diabetic No diagnosis of diabetes <25.0 (normal range) ≥30.0 (obese) All By PEN / Village By age group By education level By BMI By diabetes Prevalence of hypertension: Women 73.3% 66.1% 56.0% 39.7% 44.6% 36.2% 35.3% 38.2% 38.4% 35.7% 38.6% 34.0% 31.7% 30.3% 32.7% 28.2% 14.9% Total (555) PEN Mulifanua (150) PEN Neiafu (131) Fagaloa (162) Vaisala (112) 20-39 years 40-59 years 60-69 years 70+ years Primary Secondary compl Coll/Univ comp 25 to <30 (overweight) Known diabetic No diagnosis of diabetes <25.0 (normal range) ≥30.0 (obese) All By PEN / Village By age group By education level By BMI By diabetes Source: Cascade Study 2018 Household Survey, World Bank Hypertension was seen as a frequent condition in older people. FGD participants believed that having normal BP would be almost unheard of among the elders in their community: “At his age, if his blood pressure is normal, he could be someone from another planet.” (Male FG participant) ARE HYPERTENSIVE INDIVIDUALS GETTING CARE? We brought together the data from the Cascade Study 2018 Household Survey and clinic data on HTN patients to determine the continuum of HTN care in Samoa. 10 STUDY FINDINGS ON HYPERTENSION Based on the data collected and used in this study, Samoa’s hypertension cascade is as follows: • 73.9% of hypertensives have ever been screened • 61.4% of them have been screened in the last 12 months • 24.9% of hypertensives have been diagnosed • 19.2% of them have been initiated on anti-hypertensive treatment • 15.1% of them had their BP monitoring in the last 90 days, and • 4.7% of hypertensives have evidence of BP control Clearly, there are multiple, significant gaps in this cascade, starting from screening, to diagnosis to patient management, leading to very poor attainment of the final outcome: BP control. A very low percentage of the hypertensive population is under regular NCD management and the majority of patients in care fail their treatment target. The cascades for men and women are shown in Figure 3. The 100% columns represent the 221 men and the 201 women who either had elevated BP during the survey measurement or reported to be on HTN medication. The first five columns of the cascades in figure 3 reflect survey participants who were classified as hypertensive (1st), reporting to ever have had a BP screen (2nd), reporting a BP screen in the last 12 months (3rd), diagnosed with hypertension (4th) and on hypertension treatment (5th). The remaining two columns reflect proportions of patients on hypertension treatment who had BP measured in last 90 days (6th) and reaching the target BP level (7th), using patient file data to calculate the proportions. The HTN cascades for men and women show that: Among people classified as hypertensive (first column, 100%), the coverage of BP screening was much • higher in women compared to men (e.g. 72% vs. 52% measured BP during last 12 months, screening column) Due to better screening, more of the hypertensive women (38%) have been told that they have high BP • compared to men (13%, diagnosis column). But for both genders, the diagnosis rate is very low indicating significant gap on referral and diagnosis.  he lower screening coverage in men combined with the lower level of diagnosis lead to very poor •T linkage to HTN care of hypertensive men: Only an estimated 10% of hypertensive men were on treatment compared to 29% of hypertensive women. For both men and women, about 77% of those diagnosed are on treatment. Using data from clinic-registered HTN patients to complete the cascade, we found that an estimated 2.5% • of all hypertensive men (diagnosed and undiagnosed) and 7.2% of women have treatment-controlled BP. We also estimate that about 8% of hypertensive men and 23% of hypertensive women are regularly monitored by clinic staff for their BP as part of their HTN care indicating poor quality of case management. The HTN care cascades by villages’ PEN status is discussed in later sections and shown in Figure 13. In brief, • PEN villages had significantly better screening coverage than non PEN villages (70.0% vs. 62.4%), and the main driver was higher coverage of men in PEN villages compared to men in non-PEN villages. 11 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA FIGURE 3. HYPERTENSION CASCADES IN SAMOA MEN AND WOMEN 100% 100% 81.0% 71.6% 68.3% 52.0% 37.8% 29.4% 22.5% 13.1% 10.0% 8.3% 7.2% 2.5% Men Women HTN Ever screened Screened last 12 months Diagnosed On treatment Monitoring BP last 90 days BP control Sources: Cascade Study 2018 Household Survey, World Bank; Patient file data extraction Nov-Dec 2018 by MOH. Notes:Notes: 100% = All individuals classified as hypertensive in household survey based on BP result or reported HTN treatment; Ever = All individuals 100%Reporting screened= in surveyclassified to ever have measured BP; in as hypertensive household Screened survey in last based 12 months on BP result = Reporting or reported in survey to have HTN treatment undergone BP testing in last Ever 12 months; screened= Diagnosed Reporting Told byto in=survey a doctor or nurse ever have about having measured BP high BP (reported in survey); On treatment = Reporting in survey to take HTN Screened medicine; in last Monitored 12 months BP last 90 = Reporting days = Using in survey to haveproportion undergone of HTN BPpatients testing with documented 12 monthsBP monitoring while in in last HTN care at clinic; = Diagnosed BP by a=doctor control Told Using proportion of HTN or nurse about patients having reaching high BP target inwhile BP (reported in HTN care at clinic. survey) On treatment = Reporting in survey to take HTN medicine Monitored BP last 90 days = HTN patient clinic data on BP monitoring of registered HTN patients BP control = HTN patient clinic data on reaching BP target among registered HTN patients WHAT ARE THE GAPS AND BOTTLENECKS IN HYPERTENSION CARE? a. In primary prevention of hypertension To understand gaps in primary prevention, we collected views from FGD participants on community awareness of HTN and its prevention (see Glossary for ‘primary prevention’). Unhealthy food and stress were by far the most commonly mentioned causes of high BP, with cultural pressures of life in Samoa mentioned by both men and women, patients and care providers. A minority of comments focused on the importance of exercise, genetics, and changes to the physical environment: “Food is the reason of these things it’s all because of food.” (Female FG participant) “…also faalavelave in our family, easy that stuff it’s gonna make us pressure and make us worried, that t he other things that get a high blood pressure.” (Female FG participant) “I also think one of the reasons why high blood pressure is very serious in Samoa is because of the enormous and massive demand exerted by our cultural and churches obligations.” (Male FG participant) “there’s a lots of things or the reason why the people get these non-communicable diseases. The first thing is the food, the other one is the social life of a Samoan people, what I mean is the life of Samoa, there’s a lots of stuff, another thing is the genes of the oldies, between the couples, children’s, the social life of a Samoan people” (Female FG participant) 12 STUDY FINDINGS ON HYPERTENSION Samoa health care providers said the causes of hypertension are due to nutrition habits and lack of exercise, as well as cultural and family pressures: “Most of the time it the family issues, it’s not the food, they hardly consume fat but the main cause is problems in the family. Samoa is not like overseas countries, they are free like a dove, However, Samoa has a death sentence of pressure and stress hence the high blood pressure, it’s not the food.” (Nurse, Providers FG) “The definition of the NCD is ”Non Communicable Diseases” cause by lack of exercises, never taking NCDs medication and eating unhealthy food.” (Nurse, Providers FG) We asked survey participants whether they have access to fruit and vegetables: 80% are growing fruit and/or vegetables; and 61% consume fruit/vegetables several times a day, 13% consume them once a day and 26% consume them less often. Since nutrition is a main factor in NCD prevention, we assessed the level of obesity in the Cascade Study 2018 Household Survey, using body mass index (BMI), defining BMI of 40 and above as ‘severe obesity’. The BMIs are generally high in Samoans with 9 of 10 adults classified as overweight or obese (Figure 4): • BMIs tend to be higher in women Women have especially high rates of severe obesity (25%, BMI ≥40), putting them at very high risk of HTN, • diabetes, and other preventable conditions Obesity rates are higher (M: 53.4%, F: 76.6% - ages 20+ years) compared to 2013 STEPS data (M: 44.8%, F: • 68.8% - ages 18-64 years) • In PEN villages, significantly fewer men were obese compared to non-PEN men FIGURE 4. BODY MASS INDEX OF MEN AND NON-PREGNANT WOMEN IN SAMOA Men 51.6% Women 43.8% (non-pregnant) 32.0% 25.0% 14.6% 16.2% 9.6% 6.9% 0% 0.4% < 18.5 18.5 to <25.0 25.0 to < 30.0 30.0 to < 40.0 ≥40.0 (underweight) (normal) (overweight) (obesity) (severe obesity) Source: Cascade Study 2018 Household Survey, World Bank 13 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA  any adults have a waist circumference which puts them at increased risk of metabolic complications M (Figure 5): • The average waist circumference of men aged up to 40 years was below the cut-off  ll other age/sex groups had an average waist circumference within the danger zone of metabolic •A complications with all women age groups well above the cut-off of 88 cm • In PEN villages, men’s average waist circumference was ~3cm smaller compared to non-PEN men FIGURE 5. AVERAGE WAIST CIRCUMFERENCE BY AGE GROUP AND GENDER Average waist circumference (cm) 109.4 110.3 109.4 109.0 107.8 107.1 106.3 102 97.0 88 20-39 40-59 60-69 70+ Risk 20-39 40-59 60-69 70+ Risk years years years years cut-o years years years years cut-o Men Women Source: Cascade Study 2018 Household Survey, World Bank. Note: Cut-off to increased risk of metabolic complications as in 2013 STEPS report p37 In the two PEN villages, we asked whether people had learnt about risk factors for HTN and diabetes during the PEN Fa’a intervention. Figure 6 shows that most villagers recalled learning about causes, especially the importance of unhealthy food which is high in sugar, fat or salt. Only 5% (9% of men, 3% of non-pregnant women) said they did not learn about these risk factors. Gender differences were largest for recalling learning about smoking (28% of men, 16% of non-pregnant women), stress (30% vs. 38%), unhealthy food (95% vs. 88%) and obesity (53% vs. 48%). 14 STUDY FINDINGS ON HYPERTENSION FIGURE 6. RECALLED LEARNING THEMES FROM PEN FA’A INTERVENTION PEN Fa'a villages: Learnt about causes of high blood pressure or diabetes 91% 91% 50% 49% 34% 21% 19% 11% 2% 5% 3% Unhealthy food (high sugar/fat/salt) Overweight/obesity Lack of physical activity Stress Smoking Alcohol Genetic/‘running in family Other Did not learn about causes Don’t Remember Learnt about causes Yes Specific causes learnt about No Sources: Cascade Study 2018 Household Survey, World Bank The risk factor assessment done in 2015 during PEN implementation had found high alcohol use and smoking in men, and high levels of physical inactivity in men and women (Bollars et al. 2018). These risk factors, together with overnutrition, are typically addressed by “primary prevention” measures (see ‘Glossary of terms’ for a list of core measures). While there is awareness of HTN as shown in the FGDs, our results suggest that behaviour change is yet to realize. There is therefore a need to intensify activities on primary prevention of NCDs to reduce the various risk behaviours fuelling the epidemic in Samoa. These activities need to address the socio-cultural pressures associated with NCD risk among Samoans. b. At the screening stage Screening is a major bottleneck in the continuum of HTN care, especially in men: 48% of all - undiagnosed and diagnosed - hypertensive men had no BP measurement during the last 12 months (versus 28% of women), preventing full diagnosis and linkage to HTN care (Figure 3). We discussed BP testing during the FGDs to explore reasons for gaps in HTN screening. There was broad agreement among these FGD participants that a diagnosis with high BP was a ‘scary’ thing. Many believed that a diagnosis with high BP meant that someone would die [opinions were likely influenced by two recent deaths of HTN cases in the community]. Some women speculated that this could be a reason why some 15 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA people did not seek a HTN diagnosis. Fear was also linked to denial. “I suppose if for the first time you are being told that you have high blood pressure, you would be devastated, and very worried, thinking that you will die soon.” (Male FG participant) “….because I know when you get the high blood pressure that mean your ready to have a long rest - mean you die.” (Female FG participant) “…. I’m very worried and disappointed because if I went for a check at the hospital and the doctor say a sad news about my health and that time I’m very feel sad and also my self esteem is very low, and I’m like ohhh no I feel like I’m dying. We should protected and respond things that we know it better for our life.” (Female FG participant) “I grew up as a healthy kid, never suffer any major health problem. Last year, I suffered a severe headache. My wife accompanied me to see the doctor. After my check, the doctor in a very scary language warned me that I am close to death. I ask him what he meant. He said my blood pressure is too high - 170.” (Male FG participant) “For me the most shocking thing with our fight with NCDs is the boldness of our people to hide or tell the truth about their health. Some people died unexpectedly because of this silly attitude. Some people categorically lied that they don’t have high blood pressure, referring to a makeup family history where grandparents and parents never contracted high blood pressure or other form of NCDs.” (Male FG participant) In addition to fear of being found with high BP, conversations about barriers to BP testing also revealed the ability to pay for treatment and access to the health facility as factors (these were also reasons for not using medication or attending follow up appointments). Access to health facilities can be a challenge for these village residents in rural areas. “They should have money to buy the pills. But if they don’t have money to buy the pills they did not go to look for thing to cure it.” (Female FG participant) “The first thing is no money. Low income. No one of the family work to get money, like that. Two, some people they had high minded they say they didn’t had blood pressure. They have their own points they say they didn’t get high blood pressure but they already had high blood pressure. The other thing people of our country they didn’t care they don’t even care” (Female FG participant) The household survey data showed that in the three population groups, the women with current or recent pregnancy had the highest screening coverage (Figure 7). Still, it was not universal as about one quarter of these women said they had never had their BP measured. Regarding the gender gap in screening levels, exposure to BP screening during antenatal care can partly explain why women overall have higher screening levels. In many settings, women tend to use health services better than men, especially for prevention, and are in closer contact with services through visits with young children. Occupation and employment can be reasons for men to stay away from health services, unless they have an acute illness (“work” was one barrier mentioned by men for non-participation in the PEN screening). Lastly, there can be cultural constructs of masculinity and toughness reducing men’s interest to seek health care services. 16 STUDY FINDINGS ON HYPERTENSION FIGURE 7. HISTORY OF BLOOD PRESSURE MEASUREMENTS IN MEN AND WOMEN 81% 77% 76% 71% Prevalence of self-reported 70% 69% 67% 65% 65% 63% 62% 60% 58% 56% 57% 51% 51% BP measurment 46% 47% 48% 48% 37% Ever In last 12 months Ever In last 12 months Ever In last 12 months Men 20+ years Women 20+ years Women w current/recent pregnancy Total PEN Mulifanua PEN Neiafu Fagaloa Vaisala Sources: Cascade Study 2018 Household Survey, World Bank We also looked at the coverage of screening in PEN versus non-PEN villages, and found that in the small sample of four villages in our comparison, PEN had contributed to significantly higher screening coverage (70.0% vs. 62.4%). This result was driven by the low screening coverage in Fagaloa, one of the non-PEN villages. Furthermore, this difference was driven by coverage in men i.e. the higher BP testing in PEN village men (66.3%) compared to non-PEN village men (56.2%). There was also a significant difference between PEN and non-PEN villages in frequency of BP checks during the past 12 months: a greater proportion of participants from PEN villages reported ‘never’ having their BP checked in the last 12 months than those from non-PEN villages (22.3% vs. 16.6%). At the same time, 37.9% of PEN villagers had their BP checked three times or more compared to 29.9% of non-PEN villagers, perhaps indicating better linkage to care in the PEN villages if they were screened and referred. Participants from PEN villages were more likely to have received a BP check at their nearest government facility, while those in non-PEN villages were more frequently receiving BP checks outside of the health system (e.g. by their local women’s committee, at screening days). Table 8 in Appendix 1 provides detailed statistics. Among PEN Fa’a villagers, we determined the reasons why some villagers had not participated in the PEN Fa’a screening (Figure 8). Overall, 72% said they had not participated, and the main reason given for not participating was not being aware of the screening being implemented. Only 2% said they failed to participate because of not seeing the need to screen for diseases. The lower PEN Fa’a coverage among men (23.5%) compared to women (35.9%) was mainly due to men being at work, according to the reasons given by men. In the Cascade Study 2018 Household Survey, we also determined the coverage of glucose testing for diabetes screening, which was at 55.6% lower than the HTN screening. Men had lower glucose screening levels (48%, 263/553) than women (non-pregnant: 63%, 347/555; current/recent pregnancy: 62%, 61/98). There was no differences between PEN and non-PEN villages in terms of diabetes screening. 17 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA FIGURE 8. COVERAGE BY PEN FA’A SCREENING AND REASONS FOR NON-COVERAGE 76.5% Men Women 64.1% 47.6% 45.0% 35.9% 23.5% 20.9% 20.6% 12.4% 16.1% 7.1% 7.6% 6.1% 6.1% 1.8% 0.4% 2.2% 2.2% 1.1% 2.8% Covered Not covered Did not know about it Due to traveling Not o ered in sub-village Because of work Lack of time No perceived need Did not like screening Other Covered Not covered Did not know about it Due to traveling Not o ered in sub-village Because of work Lack of time No perceived need Did not like screening Other Yes No Reason for not having participated in PEN Fa'a screening Yes No Reason for not having participated in PEN Fa'a screening Sources: Cascade Study 2018 Household Survey, World Bank In summary, screening is a major bottleneck in the continuum of HTN care, especially in men  lmost half (48%) of all hypertensive men had no BP check for at least one year, preventing their entry •A into HTN care (the problem was smaller in hypertensive women with just over one quarter not having had a BP check during the year). BP screening is more frequent for younger women during pregnancy, but coverage was incomplete • according to women’s recall of BP testing in their antenatal care  aps in HTN screening arise through fear of being found with high BP, and the concern over costs for •G HTN medication and care, but also practical reasons like being at work. The PEN pilot has contributed to higher screening coverage, especially men seemed to take advantage • of the screening service, however, almost three-quarters of PEN villagers had not participated in the PEN Fa’a screening, mainly due to not knowing about it. BP measurements during the 12 months prior to the Cascade Study 2018 Household Survey happened • among some of the PEN villagers who check their BP more often compared to non-PEN. At the same time, more PEN villagers never check their BP compared to non-PEN, so the effect of PEN on the uptake of BP checks is mixed. The observation that PEN villagers were more likely to receive BP checks in government health facilities • compared to non-PEN villagers is a positive sign of the PEN pilot. c. At the referral and diagnosis stages We explored the breakpoints around referral as our cascades suggested that some hypertensive individuals, although screened, may not get diagnosed and linked to care. 18 STUDY FINDINGS ON HYPERTENSION In the household survey, 10% of all respondents recalled having an elevated BP in a screening event. We  assessed whether these individuals had a successful referral (Figure 9): About 7 out of 10 screen-positives received a referral (red and pink sectors in right-hand pie), but 3 out of 10 • missed out on a referral (grey in pie) Most people with a referral went to a health facility for diagnosis but there were important differences by PEN  status:  EN villages were better at care seeking after a positive HTN screen: PEN villagers had significantly better •P follow-up with a health care provider after being referred (69% followed the referral successfully, p=0.033) compared to non-PEN (50% followed referral). PEN villagers also had significantly lower failed referrals (5% not following referral) compared to non-PEN (18% - p=0.016).  EN villagers may have a better understanding of the importance of accessing HTN care once found •P with elevated BP. In the FGDs, there were many examples given of villagers having positive BP screens but not followed the referral advice, or even refusing to acknowledge the screening result. This lack of taking the test result seriously stood in contrast with the anxiety the participants themselves expressed about HTN. Disbelief about the positive BP test may have prevented some to follow the referral, or fears of getting the HTN confirmed. FIGURE 9. REPORTED OUTCOMES OF POSITIVE HTN SCREENS AND REFERRALS 1% BP screen negative 3% BP screen+, referred successfully 90% 10% BP screen+, referred unsuccessfully 6% BP screen+, not referred Sources: Cascade Study 2018 Household Survey, World Bank “We found here in your team that we had blood pressure. And then walk at the back of the house and laugh they said that I had high blood pressure but I’m not, it’s seems like a joke but they know already and decided not to go to the hospital but the husband find out later having blood pressure and die later.” (Female FG participant) “The other thing they did not believe it. They do not believe if we mention the hospital. They do not believe.” (Female FG participant) 19 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA For referral, Upolu residents (Fagaloa and Mulifanua) were about equally likely to be sent to their local district health facility or to Moto’otua Hospital in Apia. In Savai’I, over 90% said referral was to their local district health facility (and not to their tertiary care facility, Tuasivi). The local health facilities reported as the location for referral were (with only one exception) always the closest facility to the survey village (Lufilufi for Fagaloa; Faleolo for Mulifanua; Sataua for Vaisala and Neiafu) – participants did not report being referred to district hospitals other than their own. Care providers agreed that successful referral is very important as many new cases were identified through community outreach. “Most of the work we conduct is our programs in the villages and we find people new cases and patients with long term high blood pressure and they stay home because they don’t want to come to the hospital.” (Nurse, Provider FG) While providers did acknowledge challenges patients might experience in accessing health services because of cost or distance, their conversation centered on ‘denial’ or alternative treatment seeking behavior. “There are very many people that have the mentality of denial or refuse to accept their illness” (Nurse, Provider FG) “…and another thing there are some Samoan people that just sit at home and experience headaches and they are seen by the traditional healers and it ends there.” (Nurse, Provider FG)  ommunity outreach for hypertension screening is only effective if referrals to a health facility for full C diagnosis work, and care providers were fully aware of this. However, the patient pathway sometimes faltered at the referral stage  bout a third of hypertensive individuals, although screened, missed out on getting referred to the health •A facility after screening positive for HTN Referrals were almost always made to the closest health facility, the distance to the referral facility was • therefore as short as possible  EN villagers had significantly better referral rates compared to non-PEN villagers, and significantly lower •P levels of failure to follow a referral. These improvements may be due to greater awareness about the risks of untreated HTN through the implementation of the PEN pilot.  eople who were in denial about their HTN, did not belief a screening result, or had fears of a full •P diagnosis, are all at risk of ignoring the referral after a positive HTN screen. Care providers also knew about cases choosing alternative treatments, and about financial and geographical access acting as barriers to utilization of HTN services. d. At treatment initiation This section draws on the medical data from 300 registered HTN patients which we reviewed for the purpose of this study (see Figure 1 for cascade methodology). Some patients (7%) had normal BP levels at registration, possibly due to successful treatment started prior to getting registered as HTN case. Among those hypertensive at registration, 22% had stage 1 HTN, 36% stage 2 and 35% stage 3 HTN, suggesting late diagnosis and acute cardiovascular risks. Many of them might have been newly diagnosed on the date of registration as HTN case. As we wanted to learn about long-term care, we included patients who had been 20 STUDY FINDINGS ON HYPERTENSION registered as HTN cases for at least one year. 42% had been registered for 5 years or longer and 62% of the HTN patients were female, in keeping with the better screening coverage in women. From the cascades in Figure 3, we have an estimate of the gap between HTN diagnosis and being on HTN treatment. Here, we assess the step of treatment initiation in more detail – the counselling provided, the treatment approaches, time to treatment, and what medication is being prescribed. Treatment of HTN should consist of a combination of medication and healthy living counselling. The recalled counselling by survey respondents at their HTN treatment initiation was as follows: • 75% had received counselling (44% had one session, 31% had two or more sessions) • 24% had no counselling session • One patient recalled receiving just one group counselling session The counselling sessions covered knowledge topics well but seemed to miss out on focusing on treatment adherence and practical tips to overcome barriers to sustained treatment adherence (Table 2): The 68 household survey respondents who had ever been given treatment recommendations upon HTN TABLE 2. TOPICS COVERED IN COUNSELLING SESSIONS AT HTN TREATMENT INITIATION Topics covered in counselling sessions Frequency • General education on high blood pressure Most counselling sessions • BP monitoring Some counselling sessions • Education on diabetes • Patient support • Education on heart diseases • Treatment goals (BP target, etc.) • Medication schedule Few to very few counselling sessions • Importance of adherence • Reminder strategies • Storing medicine • Challenges to adherence • Switching drug regimens • Managing missing doses • Side effects/drug interactions • Planning treatment ahead for trips Sources: Cascade Study 2018 Household Survey, World Bank diagnosis mainly received advice to take medication/pills (Figure 10). Given the strong links between HTN, obesity and diabetes, the nutrition-related advice seemed too infrequent or not well recalled by patients. Some HTN patients thought that life style changes could be effective as a cure for high BP, while others were adamant that regular medication was the only option. Examples of dietary changes included consuming more vegetables, soups and green tea. Men in particular believed that exercise could play a role in managing the condition or even replacing medication. 21 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA FIGURE 10. TREATMENT RECOMMENDATIONS RECEIVED BY PEOPLE DIAGNOSED WITH HYPERTENSION Recommendations received by HTN patients (%) Take tablets/pills 94% Eat healthier food/balanced diet 29% Reduce stress levels 24% Increase physical activity 18% Reduce body weight 16% Reduce salt intake 10% Grow own food/vegetable garden 9% Reduce sugar intake 6% Stop smoking 4% Stop alcohol 4% Sources: Cascade Study 2018 Household Survey, World Bank “We thought that we can cure this diseases in terms of eat a healthy food, doing a exercise, and keep focus on your medicine” (Female FG participant) “I know it’s a non-communicable disease so it’s my own self can protect from disease.. Don’t depend on pills and treatment - you can help yourself like do more exercise and eat a healthy food, not too much….” (Male FG participant) “Well me, I never try other way to cure my blood pressure, only pills. I’m happy for that way. My blood pressure always hundred forty that’s tell me it is not high” (Female FG participant) “I can still remember clearly the advice by the doctor. Cut down on the consumption of fatty, oily, and salty food and many other foods and drinks that are not suitable for hypertensive cases. As a farmer I’m not taking my high blood pressure medication anymore. I rely on the exercises from the work I do on my plantation. After the first month, I felt good and realized it’s working for me, so I continue with it.” (Male FG participant) The nurses who participated in the focus group reported being the front line providers for those with hypertension, describing how they would take patient history, record blood pressure, and then choose the following actions depending on symptom severity – either immediate referral or providing education about diet and lifestyle before asking patients to return on days where the doctors were visiting their health centers. “In the time being an identified case with high blood pressure for the first time they have come to us, it doesn’t mean they get treatment, we advise according to diet and other education and give them a follow up appointment to check blood pressure or when the doctor refers to us.” (Nurse, Provider FG) 22 STUDY FINDINGS ON HYPERTENSION For patients to blame the HTN just on factors they cannot modify (like genes, climate change) is not empowering them for change. “when I’m pregnant I went to Faleolo for my check and the nurse confirm that I got the high blood pressure and asked ‘is someone of my family or my parents high blood pressure?’ And I say yes sis, it’s my mother. And now I thinks that’s the reason why I get this disease.” (Female FG participant) “If we look back, in the olden days, we did not have any NCDs, and now with the detrimental impact of climate change, we are experiencing devastating problems affecting the lives of our people as well as our physical environment.” (Male FG participant)  reatment initiation is generally soon after diagnosis with 73% of survey respondents reporting treatment T start on the day of HTN diagnosis or shortly after - Figure 11. Still, about 1 in 10 HTN cases never started treatment according to people’s recall •  he swift treatment initiation was confirmed with the clinic data of HTN patients - 88% of patients were •T prescribed HTN treatment on the day of registration/diagnosis. Another 5% were prescribed medication within the first month and a further 4% within 12 months of registration/diagnosis.  onsidering the variable patient profiles and HTN treatment needs, one can conclude that the timing of •C treatment start is adequate. FIGURE 11. TIMING OF TREATMENT INITIATION AMONG DIAGNOSED HYPERTENSION CASES 3% 6% Treatment started on day diagnosed 5% Treatment started within a week 8% Treatment started within a month 5% Treatment started after one month 73% Treatment never started but o ered Treatment never started and never o ered Sources: Cascade Study 2018 Household Survey, World Bank T  he HTN patients we reviewed were prescribed four classes of HTN drugs and three types of supportive medication i.e. aspirin, statins, beta-blockers, the most frequent being ACE inhibitors, especially in diabetics (Figure 12). Only 3% had five or more of these seven drug types prescribed (“HTN polypharmacy”). 14% had four or • more of the seven types prescribed. HTN polypharmacy (5 or more of the 7 classes) for ages 65 and above was still very low at 2%, and the • same held for HTN/diabetes patients (only 1%). 23 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA FIGURE 12. MEDICINE PRESCRIPTION PATTERNS IN HYPERTENSION PATIENTS 89% 57% 37% 36% 27% 30% 26% 24% 23% 21% 10% 9% 6% 5% Patients with HTN only (168) Patients with HTN and diabetes (132) Thiazide-diuretic Calcium channel blocker ACE inhibitor Angiotensin receptor blocker Aspirin Statins and other lipid-lowering drugs Beta-blocker Source: Patient file data extraction Nov-Dec 2018 by MOH At the time of the patient record review, 44% of HTN cases were also known to be diabetic (all type-2). Among the 132 HTN/diabetes patients with complete data, the majority were on pharmacological treatment: 82% were on oral sugar-lowering drugs and 11% on insulin, alone or in combination with oral drugs and 7% did not have any treatment recorded within their patient file. e. At treatment monitoring and retention in care Monitoring of BP, glucose and cholesterol: According to Samoa guidelines, HTN patients should have their BP checked at least every 90 days, and get tested for glucose and cholesterol (LDL) at least every 12 months. We looked for evidence in the patient files whether these monitoring checks were being carried out and where the gaps were (we could not include checks which had not been recorded in the patient file).  verall, 61% of HTN patients had evidence of a BP check within 90 days, and 61% had glucose and cholesterol •O checked within 12 months; differences between male and female patients were very small  mong co-morbid HTN/diabetes patients, only 51% had a monitoring BP within 90 days. Coverage of annual •A laboratory tests was slightly higher at 68% (glucose and/or HbA1C, cholesterol). It is possible that some diabetes cases had check-ups elsewhere which were not noted in their medical file  mong those without diabetes, 63% had been covered with a BP check within 90 days, and 56%22 had had •A the laboratory tests done within the 12 months. Monitoring of bodyweight or BMI: Counselling and education about a healthy diet, physical activity and weight control is part of both HTN and diabetes care. To assess whether patients successfully implement the advice, a patients’ body weight or BMI needs to be monitored and evaluated as part of the treatment plan. 24 STUDY FINDINGS ON HYPERTENSION  nly 13% of the HTN patients had their weight recorded at registration (Faleolo 4, Lufilufii 21, Sataua 13) •O  ody weight was poorly monitored - 93% of HTN patients had no record of their bodyweight in the past 12 •B months The data for calculating patients’ BMI was often not recorded in the patient files: Only 38% of the patients • had their height recorded in addition to their weight. There was very little evidence found in the patients’ files that suggest BMI was regularly or systematically tracked for patient management Among the patients in HTN care with data for BMI calculation: 18% were overweight and 77% obese (severe • obesity was present in the 8% of patients). Retention in care and patient attrition over six months: We also looked at retention in HTN care, defining it as a clinic visit in the last 6 months with a BP result recorded in the file. Those not retained may have come back to the clinic after the 6 month period, but these patients would not benefit from quality HTN care as clinic contacts need to be more often, even for long-term HTN patients. The views below collected during FGD provide reasons on why patients fail to adhere to the treatment.  verall, 54% of the HTN patients were retained in care, judging from a BP recorded over the past six •O months, and there was practically no difference between male and female patients (M 53.9%, F 54.6%) However, retention in care showed large differences between the three health facilities: Faleolo HC 40%, • Lufilufi HC 45%, and Sataua DH 78%  onversely, non-retention (attrition) was at 46% (M 46.1%, F 45.4%), and the scale of the problem differed •C between the facilities (Faleolo HC 60%, Lufilufi HC 55%, and Sataua DH 22%) f. Achievement of treatment targets The marker for success in HTN treatment is for patients to achieve a BP <140/90 mmHg (or <130/80 mmHg if diabetes is present). We used the last recorded BP result in the patient files to determine BP control treatment success.  verall, only 19% of the patients in HTN treatment achieved their BP target; the level was identical for O male and female patients Among co-morbid HTN/diabetes patients, only 9% achieved the BP target of <130/80 mmHg • Among those without diabetes, 34% achieved the target of <140/90 mmHg • While there are many reasons for patients to fail their BP targets, the non-availability of anti-hypertensive • drugs at clinics was highlighted Another major reason for poor outcomes is patients stopping medication due to lack of understanding the • clinical management of HTN - this includes stopping the medication once BP readings return to normal, or when symptoms disappear. 25 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA “We had only one day available for our whole district to the hospital at Faleolo. We only had Tuesday for the whole week? Oh no we had two days Tuesday and Thursday for their service. So by the time we come to visit them the medication was finished that’s what happen” (Female FG participant) “…. if I told her for a pills of the blood pressure she also say the pills is finished, it’s nothing, and I goes, hey is that how you serve a people?” (Female FG participant) “In the morning, in the hospital it’s many of them - they are coming for treatment, but not all the time we’ve got anti-hypertensive drugs, because only the central ones in the hospital. And the doctor comes in only once a week, so we advise them to come when the doctors visit…” (Nurse, Provider FG) “Sometimes my checks came out high and some are low… I ask the doctor if my high blood can be cure completely by the medication I’m taking. He happily informed me that medication can certainly cure high blood pressure. Since there is no change for the better in my case, I’m giving up on my medication” (Male FG participant) Regarding the availability of HTN medication, it can be challenging for providers to ensure continuity of supplies especially outside Apia. The medications are brought in by the physician on the day she/he visits the district hospital, which may not be enough to meet the needs before physician’s next visit. Doctors prescribe what was accessible to them, and if they need to substitute, patients may refuse the drug. “… there are a lot of medications here for anti--hypertension medicine that is not available in our clinic… so they have to wait for the outreach clinic to come out and deliver. They go, and at the time they come the next week there is no Doctor.” (Nurse, Provider FG) “… there is a Doctor however they are not able to bring a lot of medication, therefore when it leaves the people no choice but to go to Apia to received their medications.” (Nurse, Provider FG) “We do have… but not all them - what are they called - the essential medicines. Remember each hospital has a pharmacy, there is something I would like the Doctors to work with us. If they can write or prescribe to bring these medication and injections, would be great if they can stock the need when the time arises.” (Nurse, Provider FG) “… if this person come to pick up the medication but the medications is out of order but the doctor change their medication. Then suddenly, they don’t want to take their medication because they think the pills are wrong. Then… they don’t even take it anymore” (Nurse, Provider FG) M  any patients are dissatisfied with the hypertension care they receive, and this is a contributor to few patients having treatment success Waiting times for follow-up visits were a major problem in patients’ views. • Female participants in particular explained that the negative attitudes of doctors may prevent patients from • feeling comfortable sharing their symptoms or asking questions The lack of clarity from medical professionals about diagnosis and treatment was a frequent criticism – • information can be conflicting, test results not communicated, or treatment regimens not explained clearly f provider-patient communication is poor, patients make wrong conclusions on the clinical management of •I their HTN (however, providers generally believed that they communicated clearly and it was the patients not understanding the treatments) For some patients, this means losing trust in the health system and not visiting, or losing trust in a medication • and turning to traditional medicines or alternative cures 26 STUDY FINDINGS ON HYPERTENSION “I got there at 9.30am but I went out at 3.30pm. I was in there for that long, because it seems like they just concern about the children that were taken on that day for treatments. But I can’t wait in the hospital for 7 hours.” (Male FG participant) “We keep waiting a long time for the doctors or the people working at the medical records those who passing the papers to us to see the doctors.” (Female FG participant) “…and they way they look to people they argue and show their long faces to people, those faces they show to the people they are shy of share others they already had those diseases.” (Female FG participant) “For me I don’t want to see the doctors at the health center because of the long queue, the attitude of the doctor. The inconsistency in the doctor diagnosis and the prescribe medications is very scary to me.” (Male FG participant) “And the nurse at the hospital she didn’t told me the result of my hypertension, just only check and say nothing, but a local private cardiologist said and inform me about the reason.” (Female FG participant) “,,, he measure my blood well and he said ohhh it too high, and I told her can you try another machine, and she do it. The result say it’s normal, so on that time I’m so confusing …” (Female FG participant) “I went to see him [my Family Doctor] as part of my monthly check up. He said your blood pressure is kind of high but not really high. So I did not really know what he meant… He went on and prescribed some medication with the advice to take everyday. By doing this, I’m more confused.” (Male FG participant) “I am a hypertension case, I consulted the doctor, she prescribed blood pressure medication for me. I took my medication for a while when I started coughing regularly, I consulted a private doctor and she advice me to stop taking my hypertension medication. Still worried and confused about my blood pressure, I started taking herbal medicine.” (Male FG participant) Problems with human resources (staffing at physician, nursing and clinician level) is a main concern of providers. Lack of staff was impacting their ability to identify HTN cases, treat them, or provide health education services necessary for prevention. “I think rural area they need it [more staff] because not only doctors wants to go there. Doctor need a nurse and also the people in the pharmacy with a drugs and all those staff.” (Nurse, Provider FG) “In Savai’i we have four interns working here to cover all of Savai’i, they cover Savai’i and are based in Tuasivi. So it’s impossible for four interns, three and the doctor who is a registrar to come once a week.” (Nurse, Provider FG) “We had a program, on Mondays, Wednesdays and Fridays. We have a PA system that the Nurse use to talk from the mic, in explaining NCDs in every way in-front of the waiting area. In protection, like how you feel when you have hypertension? What food you need to eat? What things you need to do? How to protect you from NCDs? You need to do more exercises, eat healthy foods…. Not only hypertension, but all the NCDs like Diabetes, Heart Disease…etc. The days that the Pharmacy open, their appointments and everything for their information. Not only for those who got hypertension, but for everyone who are in the waiting area, ready to go see the doctors. That was Tuasivi’s process, but at the moment, the short of our midwives, stop that program from continuing.” “Well usually for the first time there’s nothing they can do because back there you know ‘cause there’s no doctors. We usually hold them - we observe them three times or four times and call the doctor and await for their orders such as blood tests or confirm other orders.” (Nurse, Providers FG) 27 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA “So I know one of the big challenges of the role of Nurse manager in terms of diagnostics and management is the small amount of resources and technology and other things, to help us to do our jobs” (Nurse, Provider FG) Lack of physical resources was another shared experience of the providers as they find maintenance of existing equipment and ordering of new equipment a challenge. Budget cuts were given as a reason for receiving low quality equipment. “We need the quality. The best quality, not the quantity. We order and the pharmacy provide us the sphyg’s (coll.) but the problem with the sphyg, when you pump it high - like about 200 - it’s broken.” (Nurse, Provider FG) “… we leave it … because there are no resources for cholesterol. Also the blood pressure monitor, we told them it was broken, we need another one but there is no respond, and suddenly stopped.” (Nurse, Provider FG) Many providers talked about prior experiences where they were able to offer more programming in prevention, but all agreed that the pressure on health services was much greater now than in prior years. Some providers believed that patients’ access to cars lead to more health visits, while others felt that education was needed to make sure that patients were utilizing the correct services. Specifically, that primary care options needed to be improved so that patients with a common cold or flu were going there, rather than the tertiary care facilities where their presence was increasing waiting times for other, more urgent issues: “At this time, the hospital is very full, the people get a chance to come to the hospital anytime, because they all have cars.” (Nurse, Provider FG) “… there were times, we tried to explain to the people to make them understand, that all you need to do to the ‘Flu’ is to stay home, drink the water and take some panadols, than to come to the hospital, waiting for a long time at the long queue and will spread the flu to other people, but that’s the only reason of coming to the hospital. But due to the shorts staff, some program cancelled.” (Nurse, Provider FG) One program that was widely discussed during the focus group was the METI program – a program sponsored by the National Health Service that promotes a whole food, plant-based diet. The program is active in village outreach and it appeared as though they had held several seminars in Mulifanua. METI are also well known for weekly columns in the Samoa Observer with testimonies from participants who report reversing their hypertension and diabetes using the plant based diet and the fact that they are able to stop taking their prescribed medications after following the program. The reliance on this program could be problematic for promoting adherence to treatment regimens: “I so happy that I know I got the hypertension and I’m ready to have a diet and also take care myself. I told them I don’t want a pills for my blood pressure because I want to join the program of METI, because I know I have a right to join them there’s big change for me and my life.” (Female FG participant) “I understand that every Friday, the METI program send a team to run a diet program for the people of Mulifanua. I know that there are success stories of the METI program. I know people of our village with very bad hypertensive problem who participated in the METI and are now fully cure.” (Male FG participant) “That’s why I didn’t go the hospital at Faleolo because I hope and I’m sure METI is good for me. I’m so happy that I try their diet food. I know the METI they not did the full check but I’m very happy for their program. About their soup they don’t want a meat just only a vegetables that it and I know it is very good for me and my life for the future.” (Female FG participant) 28 STUDY FINDINGS ON HYPERTENSION HAS PEN FA’A SAMOA IMPACTED HYPERTENSION CARE, AND WHAT ARE PATIENTS AND PROVIDERS EXPERIENCES WITH PEN? Given the World Bank funded new NCD project will support rolling out of PEN Fa‘a Samoa screening, it is important to review the results of the earlier pilot of the program.  e assessed whether PEN Fa’a Samoa implementation in Mulifanua and Neiafu had led to any differences W in the HTN care cascade compared to the two non-PEN villages Fagaloa and Vaisala. We found the following among people classified as hypertensive in the survey (Figure 13):  mong people with the condition, screening coverage was higher in the PEN areas (+ 10% ever BP •A screened in PEN) and there was evidence of recent BP screening activity also being higher.  etter screening translated to slightly higher diagnosis and treatment levels in PEN areas (+ 3-4% in PEN) •B • The final outcome, BP control, was however similarly low in PEN and non-PEN areas at 6-7% of all hypertensives (note that the below cascades could only use household survey data whereas the cascades for men and women in Figure 3 used clinic data for better sample sizes and precision) FIGURE 13. HYPERTENSION CASCADES BY VILLAGES’ PEN STATUS 100% 100% PEN villages 79.0% 69.2% Non-PEN villages 66.2% 57.2% 26.0% 23.1% 21.0% 16.8% 6.4% 6.7% Hypertensive Ever screened Screened last Diagnosed On treatment BP control 12 months Sources: Cascade Study 2018 Household Survey, World Bank Most women in the focus group were familiar with the PEN program and appreciated the attempt to bring healthcare to the community and the training that the PEN facilitators had received, however, There was disappointment about the lack of resources and village women not being paid for their work for • PEN Fa’ a Samoa program, and frustration at the lack of communication of results from the Ministry of Health Women’s comments about the program were more negative than positive and not all thought it should • continue  en were less familiar with PEN but expressed a desire to see the program return to their village •M In case of PEN continuation, the women suggested PEN is combined with METI and that PEN provides • medication and monitoring, and not only screening. 29 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA “The PEN. Is one of the important program. They started with no continuation.” (Female FG participant) “They have no money to pay the girls” (Female FG participant) “Suggestions we can still continue on the PEN because they work partnership with the METI here and add something up to the PEN. It should be medicine to cure. They should not just measure and tell them about the readings of the blood pressure without doing anything” (Female FG participant) “ …. most of the time if the not enough strips to measure the sugar then we keep waiting for them. Or else they will just do their job but no treatments…. But they do not just measure the sugar, measure the blood and then just leave at there and gone, without nothing given to them for treatments.” (Female FG participant) “It was good because the girls keep follow up every week, cause every week they do their work. Because they went every day to do their work, so the people waiting for them and expecting them to come over. They didn’t mind them…. and then they just stopped and they said they had not enough equipment or other they takes some equipment to used because we had nothing it’s all broken.” (Female FG participant) “They just come and pick up the reports from the girls who did the work but they didn’t say anything thing about what’s going on whether the village is okay or what whose okay or what” (Female FG participant) “They [The Ministry of Health] just push back their plan and they never come back.” (Female FG participant) “Well me and my thoughts, I don’t want PEN anymore, even they here to do their job I don’t want it anymore, they here before they didn’t do their job properly even their equipment, It is not enough, the other thing is they don’t work together, and they stand up and leave us but we don’t know what happened.” (Female FG participant) “I remember vividly it was my participation in the PEN that first revealed I have high blood pressure. I was given a referral, and I saw a doctor and it was the start of my treatment.” (Male FG participant)  Care providers expressed appreciation for the underlying goals of the program, believing that if implemented correctly it can help the healthcare system to support community outreach, however, Providers questioned PEN’s sustainability and echoed patients’ views about why PEN had not been entirely • successful in its initial launch, citing lack of adequate training of the PEN work force, problems with recording of patient data, lack of equipment, and the relatively small reach of the program Providers also expressed a sentiment that PEN was just ‘one more program’ that was being added to their • workload without the provision of adequate supplies or support When re-starting or expanding the program was discussed, providers generally expressed enthusiasm, but • talked about the need to better align and coordinate with the health facilities while being self-sufficient 30 STUDY FINDINGS ON HYPERTENSION “At Sataua, one improvement I’ve seen is recruitment of people from the PEN program. We see them as appointments and I’m beginning to see better documentation this time.” (Nurse, Provider FG) “For example, Lalomalava village, it has been a great step forward in the process for the first time, and then the participation even the peer groups that trained they become success. They just went on their notebooks and said that this man his BP is…, and diabetes is… then disappeared slowly. So it’s the sustainability…” (Nurse, Provider FG) “To me now, I look at the concept behind the PEN, it’s nice because it can be connected. But that’s what I mean, if it can connect so we need to provide infrastructure so that it can be completed to others so it can be connected.” (Nurse, Provider FG) “… I know some men of this village, they are fat but their weight written on the referral are like children’s weight. So, it’s not match if we do the BMI.” (Nurse, Provider FG) “This means the PEN program needs to reboot. Because is only the screening can work on with but can’t work on treatments and diagnose.” (Nurse, Provider FG) “So there were programs introduced to us, they said, “Go and do this”. We do it first and finished, and then another program, and they also said, “And do this one” Like they just start it, and endless. So what I mean, we the staff, are happy to work if we have enough resources to do the job, but if not, we don’t know what to do.” (Nurse, Provider FG) “It’s important that it does not contradict hospital protocols. .” (Nurse, Provider FG) “ Instead of our workforce attending PEN, it needs to stand on it’s own… When screening, referral, referral to hospital then the hospital can work together with them because we give them tests and bloods tests. There is a protocol in place.” (Nurse, Provider FG) “.. we have outreach programs through the Ministry of Health like taking blood pressure... So we are Nurses and we need to be informed because we have patients who come in who state they had their blood pressure taken at their village...” (Nurse, Provider FG) 31 STUDY FINDINGS ON CHRONIC CONDITIONS AND THEIR CARE IV. Study Findings on Chronic Conditions and Their Care BURDEN OF CHRONIC CONDITIONS AND CARE SEEKING BEHAVIOURS More women said they have at least one chronic disease (28.5%, compared to 16.4% of men) - Figure 14 This was lower than expected and suggests underdiagnosis and possibly also underreporting in the survey • Women with current/recent pregnancy reported low chronic disease burden (8%, not included in figure) • Most of the reported chronic disease burden was due to high BP and diabetes, especially in women, and • some of the other conditions reported such as heart disease, eye problems, and kidney disease, may be complications from HTN and diabetes. Another nutrition-related disease reported by 2.3% of men was gout • Many conditions were reported by just a few respondents, e.g. injuries, joint pain, headaches, mental health • conditions (nobody reported gastric ulcers, rheumatic fever or cancer) There were no differences in the proportion of people reporting HTN between PEN and non-PEN villages • FIGURE 14. CHRONIC DISEASES REPORTED BY SAMOA MEN AND WOMEN 16.4% Any chronic condition 28.5% 4.9% High blood pressure 13.9% 4.0% Diabetes 11.7% 1.1% Asthma/resp. disease 2.2% 1.4% Heart disease 1.1% 2.3% Gout 0.2% 0.7% Eye problem 1.6% 0.7% Skin problem 0.9% 0.5% Ear/ENT problem 0.5% 0.5% Men (554) Liver disease 0.5% 0.0% Women (555) Low blood pressure 1.1% 0.5% Arthritis/osteoarthritis 0.4% 0.4% Paralysis 0.4% 0.0% Mental health 0.4% 0.0% Kidney disease 0.4% Source: Cascade Study 2018 Household Survey, World Bank. Note: Women’s data are from the non-pregnant women group 33 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA Women reported slightly higher frequencies of clinic visits, especially women with current/recent pregnancies (most had received antenatal care at health facilities, except some women from Fagaloa which has active traditional birth attendants) - Figure 15. Although there were no differences between PEN and non-PEN villages in frequency of visits to healthcare facilities, there were differences between Upolu and Savai’i villages, with participants from Savai’i villages visiting healthcare facilities more frequently. FIGURE 15. FREQUENCY OF CLINIC VISITS BY SURVEY GROUP Men (554) 62% Non-pregnant women (555) 54% Reasons: Women with curr/recnt pregnancy (98) Preference of 45% traditional healers/fofos, distance to health facility and dislike of health workers 19% 18% 14% 15% 11% 12% 14% 8% 4% 6% 3% 3% 6% 5% 1% >1 per month Monthly Every Every <6 monthly <6 monthly 2-3 months 4-6 months as feeling fine despite need Source: Cascade Study 2018 Household Survey, World Bank. Note: Fofo = traditional masseur fofo=traditional masseur Reasons for infrequently visiting healthcare providers differed between PEN and non-PEN villages: A greater proportion of PEN villagers preferred using their local health center (80.3% vs. 56.5% non-PEN), • while non-PEN villagers preferred other government facilities and private doctors Non-PEN villagers were more likely to report preferences for traditional healers (5.8% vs. 3.9% in PEN), • disliking healthcare workers (1.1% vs. 0.5% in PEN), problems with transport, or the healthcare center being too far away These differences may be due to access: PEN villagers took a shorter time to reach their nearest health care • facility (30 minutes vs. 49 minutes non-PEN) and lower cost of transport (34 WST vs. 44 WST non-PEN).  octors were the most frequented chronic care providers, almost three-quarters of patients were in their D care, and about 1 in 10 also received care from nurses (Figure 16). Only 6% had a dietician involved in their care, and few got care from traditional healers or counsellors. • Among the chronic care patients, 68% had received written advice about how to manage their condition(s) • The chronic care patients spent on average WST 50 (US$ 20) per month on the treatment of the condition(s), • including transport, drugs, consultations. Half of the respondents reported an average weekly household income including any remittances of below WST 250. 34 STUDY FINDINGS ON CHRONIC CONDITIONS AND THEIR CARE FIGURE 16. CARE PROVIDERS TREATING PEOPLE’S CHRONIC CONDITIONS 2.7% 0.7% 0.3% 6.1% Doctor 7.8% Nurse Hospital based specialist 10.5% Dietician 72.0% Counsellor Traditional healer Other Source: Cascade Study 2018 Household Survey, World Bank. Note: Includes 296 responses of the 261 men, non-pregnant women and women with current/recent pregnancy who receive care for their chronic conditions (some citing multiple providers) WHY IS THE CLINICAL MANAGEMENT OF CHRON IC CONDITIONS OFTEN INTERRUPTED?  he reasons for stopping the treatment of chronic conditions fell into four categories (Figure 17 gives T household survey results): Problems accessing medications, including coordination within the health system - Timely medication refills 1.  are a key challenge for chronic patients, underlying reasons include the inconvenience, cost, availability and perceived importance The patient feeling better or having normal test results – This suggests poor understanding of the “silent 2.  conditions” of NCDs like HTN and diabetes 3. Not prioritizing health and wellbeing 4. Choosing alternative treatment options 35 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA FIGURE 17. REASONS FOR STOPPING THE MEDICATION AND TREATMENT INTERRUPTION Forgot to get refill/running out of drugs 30.5% Feeling better 30.1% No money for transport to collect drugs 23.7% Drugs too expensive 23.3% Forgetting to take the drugs 20.7% Refill too complicated/time-consuming/far 20.6% Tired of taking drugs 19.5% Unwanted side e ects 16.1% Drugs not available at facility 15.4% Travelling 13.5% Taking alternative or traditional medicines 13.3% Ill health 9.1% Feeling depressed 6.4% No food to take with the drugs 5.1% Time 4.6% Other 1.2% Sources: Cascade Study 2018 Household Survey, World Bank Convenient access to long-term medications is a significant barrier to treatment adherence. When the drugs are not available at the local health center, patients need to travel to Apia to get the medications. Sometimes records are not clear who had prescribed, and re-issuing or refills became more complicated. “But I went all the way to Apia it’s a long trip and far away more money to spend to get the medication from Apia but this is not a right thing to do. And they look stressful and feel unhappy.” (Female FG participant) “The bad thing we can’t take it [prescriptions] at any pharmacy, only at the hospital, but the pharmacy they can’t, just the hospital” (Female FG participant) Nurses acknowledged how patients could become confused because their record systems did not always allow them to understand what advice or treatment a patient had been given previously, which may have led to conflicting advice. They stressed that record keeping could be a major area for improvement. “The right answer to this is, we Nurses cannot know who has had a blood pressure taken, now there are a lot of times when the Doctor can not recall either during his visits. The only way forward with Sataua is by creating a file easier to track our patients.” (Nurse, Provider FG) Stopping long-term medications when symptoms have disappeared means treatment is interrupted although it should be maintained: There was a general lack of recognition that some NCDs requires long-term management – perhaps because NCDs are often only diagnosed in later life. 36 STUDY FINDINGS ON CHRONIC CONDITIONS AND THEIR CARE Not prioritizing health or putting other priorities first quickly impacts medication adherence. “… the reason why I missed my drug I have a lots of things and stuff to do at home and that’s why I forgot to take a pills.” (Male FG participant) “We forgot, most of the time when something emergency then I will just go and forgot to drink medicine.” (Female FG participant) Choosing alternative treatment options also meant that the prescribed medication was interrupted. While these treatments may well be helpful, they should often complement the medication rather than replace. “So the other idea I got from another person is, you know the Swedish from the sea, so gather that thing and put in a big teapot and boiled only water, and also the roots of every tree put it together and boiled and we drink it.” (Male FG participant) PATIENTS SUGGESTIONS FOR SUPPORTING THE MANAGEMENT OF CHRONIC CONDITIONS Clearly, sustaining the treatment of chronic conditions over time is challenging and patients require support. We asked the respondents about the best non-clinical contacts to support chronic disease management including medication adherence (Figure 18). • Close family members were by far the most favoured persons to support a chronic patient A greater proportion of participants from PEN villages suggested community health workers could play a role • compared to participants in non-PEN villages (10.7% vs. 6.9%), potentially reflecting their experience with the program, which engages community members in screening and care When asked about ways to remember taking the medication, the most frequent tools to remind were alarm clocks and scheduling the treatment for a set time, such as meal times. Some also mentioned SMS and social media as helpful tools, as well as wall calendars, radio reminders, and frequent visits to health services. Only few mentioned pill boxes as helpful tools for regular medication intake. FIGURE 18. PREFERRED PERSONS OUTSIDE CLINIC FOR SUPPORTING MEDICATION ADHERENCE 3.4% 0.5% 0.1% 6.8% Immediate family member 7.5% Community health worker Treatment peer/buddy Friend 81.7% Traditional healer Other Sources: Cascade Study 2018 Household Survey, World Bank 37 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA  In order to find solutions, we collected views on whether certain strategies could work to improve medication adherence and treatment success (Figure 19). Suggestions concerning health facility capacity to provide efficient, flexible and friendly NCD services • received a lot of support (pink bars, figure) Better patient information was also welcomed by many individuals, including more counselling at treatment • start (grey bars) Several strategies for supporting patients to maintain treatment and navigate challenges were also favoured, • among them phone reminders and community support groups (orange bars) Making drug pick-up more convenient also received positive feedback (blue bars) • Interestingly, participants from PEN villages were consistently more likely than those from non-PEN villages • to endorse the suggested strategies for adherence to treatment. This may be due to the better NCD awareness in PEN villages as the intervention had sensitized the community to the issues around NCD care. FIGURE 19. FAVOURED STRATEGIES TO IMPROVING TREATMENT ADHERENCE AND OUTCOMES Better information materials 43% Information/ education More information by sta 39% More counselling at treatment start 34% More reminders via phone 34% Community support groups 28% Treatment support More counselling when there are problems 27% Tracing when missed an appointment 26% Somebody to support treatment taking 22% Social media support groups 20% Support with alcohol and substance abuse 4% Drug pick-up at more convenient sites 33% Drug refill Have someone else pick up the drugs 21% Pick-up of >1 month of medication 20% Shorter waiting times 62% Service delivery Better sta attitude 62% Longer opening times 61% Faster service for NCD patients 44% Fewer (counselling) visits 26% Source: Cascade Study 2018 Household Survey, World Bank. PROVIDER SUGGESTIONS FOR IMPROVING CHRONIC CARE Suggestions for Improving Access to Healthcare and Medications: There were many similarities among the comments of patients and providers in terms of how healthcare and medication adherence for HTN/NCDs could be improved; in particular, the need for community involvement. Providers echoed patients comments that the women’s committees in the village could play a central role, but 38 STUDY FINDINGS ON CHRONIC CONDITIONS AND THEIR CARE described a multi-level approach that would include mass media, the village community, families, and patients themselves. “So these are the kind of messages [basic education about what hypertension is] that it’s good to put on the media, so that we are all saying the same message” (Nurse, Provider FG) “I think it’s effective [media campaigns] because if you hear the prostate cancer… there is quite many people talking about it now and many of them presented to hospital for screening. I think if we do that, it would be nice…” (Nurse, Provider FG) Providers expressed hope that the recent merger combining the Ministry of Health and the National Health Service and some recent directives from the Prime Minister would help to facilitate work by the women’s committees: “There is a network, my only belief, if the committees can be reunited it’s the easy way to do all the work. Because nurses always visit villages and other things. So all those services they measure the high blood pressure, diabetes, babies, vaccinations, all those things as we worked on in the past days.” (Nurse, Provider FG) “…. this is new re-structuring of the department to try and have Doctors embedded in the district hospital. There is also offices for Nurses duty 24hrs to cover, it’s unique in the villages to go back and strengthen the connections with the community and our workforce like the old days. Like the Women’s Committee, they can check blood pressures and they know their village information especially in the rural district. This connection was broken and now we are trying to reconnect.” (Nurse, Provider FG) One practice reflected the use of communities for client tracing: “We are doing that one for our maternity cases. Because we have our own appointment system… where we have all their phone numbers and their contact numbers and their addresses so if they didn’t come, we send the list to the communities and the community will trace them up - whether they are here or not.” (Nurse, Provider FG) Providers also expressed a desire for participant engagement and saw a role for the patient’s family. They described wanting to work “as a team” with patients and gave examples of where patient engagement was benefitting treatment adherence: “For example, say if the daughter is caring for her grandmother she advise her to drink the medication, secondly, another example the daughter in law in Savai’i needs help so she calls the hospital for assistance on which medication the grandmother should take making sure she is getting better.” (Nurse, Provider FG) Others were mentioning the use of notebooks by patients: “They told me to write down their blood pressure results, so I write it down, here is your blood pressure results today. So there are few people who got their own notebooks.” (Nurse, Provider FG) A conversation focused on the need for specialist NCD providers: “Need special people to check NCDs only. Because, every part are not good” (Nurse, Provider FG) “The main thing should do is to have an institution just for health education, health promotion itself. That’s all to improve. Let people work on primary health care to keep people from these things. This is all the time in the process of treatments and work on with all these things, but if all go and work on primary health care like those days, so there is no one needs to come in the hospital.” (Nurse, Provider FG) Some providers described how the process for NCDs could learn from midwifery by having specialist providers, stressing that the system for midwives was much better functioning because of the lower patient to provider ratio. 39 CONCLUSIONS V. Conclusions At 36% prevalence in adults, HTN is a very frequent condition in Samoa. There is a good degree of awareness in the population about the burden of HTN especially at higher age. The gaps in HTN service delivery and uptake are however very serious. This study demonstrates how bottlenecks across the continuum of HTN care multiply and as a result the HTN cascade tumbles. Due to large gaps in screening, diagnosis, effective clinical management and monitoring, Samoa’s hypertension control efforts have poor clinical outcomes. We estimate that only about 2.5% of all hypertensive men and 7.2% of hypertensive women have treatment-controlled BP. Many Samoans are fearful about getting diagnosed with HTN, as they associate the condition with cardiovascular emergencies and death. People often choose to ignore a positive screening result or the referral advice to obtain a full diagnosis. HTN treatment is considered burdensome by many patients in terms of drug costs and the inconveniences of accessing clinic services. Service satisfaction is quite low with long waiting times for consultations and unfriendly staff contributing to negative views. In line with other reports, this study revealed a series of issues in the service delivery system, in particular the primary health care facilities. The study found that limitations in the number and types of health care personnel creates bottlenecks in care provision, as does the concentration of doctors at hospital level23. This limits the reach and scale of NCD care, and stifles Universal Health Coverage. While there are many reasons for patients to fail their BP targets, the lack of anti-hypertensive drugs at clinics came out as a dominant factor. Also, stopping the medication once BP readings return to normal, or when symptoms disappear, is common and linked to poor understanding of HTN as a permanent condition requiring life-long treatment. There are several modifiable risk factors for HTN, including alcohol use and smoking in men, and high levels of physical inactivity in men and women. Still, nutrition is probably the most critical among the modifiable risk factor in Samoans across all age groups. All the evidence in this study points to overnutrition as an important factor for HTN risk, level of severity and treatment success. While people have knowledge on the importance of healthy eating, the issue of weight loss did not feature much in the focus group discussions and survey interviews. Equally, the medical records of HTN patients demonstrated the insufficient focus on body weight with only 13% of HTN patients having body weight data recorded at registration and regular body weight data lacking in most patient files. In a context where obesity is highly prevalent and normalized, hypertension control cannot succeed without changing attitudes on the contribution of overnutrition to disease burden. With 8 out of 10 adults still growing fruit or vegetables, and 6 out of 10 consuming them several times a day, messages about healthier nutrition behaviours might resonate with the population despite the general reliance on imported convenience foods. Although awareness about NCDs and how to prevent them exists in the population, behavior change still needs to happen. The stress arising from cultural obligations in the family and in public life need to be reflected on and addressed to facilitate healthier lifestyles. While the study was not designed to evaluate the legacy of PEN, we were able to collect some service coverage data and stakeholder perspectives concerning PEN. Although many PEN villagers had not actually participated in the PEN program because of not being aware of its implementation, the coverage of blood pressure screening was higher in PEN villages, especially among men, compared to non-PEN villages. This suggests that such community- based screening can help close the screening gap. The improved screening had translated to slightly higher diagnosis and treatment levels in PEN compared to non-PEN villages. Concomitantly, screen-positives in PEN 41 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA villages were better at following-up with a health care provider after being referred, which may be due to enhanced HTN risk perception, as well as strengthened referral links (a PEN objective). We had no evidence of better HTN patient outcomes in the PEN areas, but clinical management of HTN patients had also not been a focus of the PEN intervention. When recalling PEN messages about causes of HTN, most PEN villagers recalled learning about the importance of unhealthy food which is high in sugar, fat or salt. The significantly lower prevalence of obesity in PEN village men and their lower waist circumference are therefore worth mentioning. Stakeholder views about PEN were mixed, but there was agreement that PEN should receive sufficient investment to adequately cover human resources and equipment costs, and that PEN should be expanded to cover nutrition (e.g. link-up with METI), as well as patient care and treatment. The study also provided information on chronic conditions and their care in general, which can further inform the health system design to cater for today’s health problems. We found that 22% of adults suffer from a chronic condition and spend on average WST 50 (US$ 20) per month on the clinical management of the condition(s), including transport, drugs and consultations. In the FGDs, payment for treatment had been mentioned as a barrier to care seeking. While out-of-pocket (OOP) expenditure in Samoa is comparatively low in absolute and relative terms, the cost burden especially of chronic conditions on families remains important (in 2014, the OOP expenditure in Samoa represented only 5.9% of total health expenditure, in comparison, the OOP share in LMICs and PICs was respectively 56% and 13% in 2014). Patients perceived doctors are the most frequented chronic care providers, although many chronic care services could likely be delivered by nurses if they were available in sufficient numbers. The reasons for stopping the treatment of chronic conditions are similar to those found for HTN. As long as patients struggle with accessing and affording the medications, the risk of non-adherence is high. Also, a lack of understanding of what the medication does leads to treatment interruptions once patients feel better or have test results in the normal range. NCDs are a major cause of premature mortality in Samoa calling for a repurposing of PHC services. Unless there is drastic change, NCDs will continue to cut lives short, cause disability and burden the health sector with severe disease complications to manage and finance. Figure 20 illustrates the heavy contribution of NCD causes (in blue) to pre-mature mortality projected for Samoa for the year 2030, which is expected to occur in the absence of drastic change24. FIGURE 20. PROJECTED CAUSES OF PRE-MATURE MORTALITY IN SAMOA IN 2030 Annual % Change IHD HTN HD COPD Diabetes LRI 2000 to 2030 YLLs 100,000 + + -3% Oth NN Oth Cardio Hep -2% HIV STI CKD NN Preterm Oth NN Asthma CMP RHD NN Enceph -1% Stroke Oth Resp Falls Self Harm Med Treat Mech Drown 0% A Fib Aort An Urinary Endocrine Endocar F Body Lung C Colorect C Oth Neopla Alzheimer’s Cirr Hep B 1% Leukemia Cervix C Cirr Alc Cirr Hep C Stomach C Pancreas C Brain C Ovary C Epilepsy Oth Cirr Road Inj Violence 2% Liver C Ileus Gall Bile Congenital Skin Breast C Uterus C Prostrate C PUD Disaster Melanoma 3% Source: https://vizhub.healthdata.org/gbd-foresight/. Note: Age-standardized, ‘reference case’ scenario. 42 RECOMMENDATIONS VI. Recommendations Scale-up public awareness communications on nutrition, HTN and diabetes, using 1.  multiple media and channels, as well as policy interventions supporting primary prevention of NCDs such as taxation policies Why? What are the experiences? Implementation tips Elevated BP and raised blood If well done, information and  se a mix of communication •U glucose are by far the most sensitisation campaigns form the channels in awareness important chronic conditions in backbone of any public health campaigns, including Samoa, responsible for the most intervention. Better knowledge mass media and targeted NCD-related pre-mature deaths. increases risk perception, creates interpersonal communication Overnutrition and obesity are demand for services and supports  ddress pressures and •A driving factors for both conditions. personal action. Improved obligations specific to Samoa’s information and education levels culture, which cause stress and The population needs to have have the potential to ameliorate much higher awareness and contribute to the NCD epidemic outcomes across the entire knowledge about the risks of continuum of care, from uptake  armonise messaging, which •H untreated HTN, diabetes and of screening to compliance with should be evidence-informed, obesity. Higher awareness clinical management. and be linked to screening increases demand for screening service offers and health-seeking for NCD care. WHO’s NCDs ‘best buys’ provides evidence based lists  ohesive branding may help to •C Primary prevention is enhanced of interventions for reducing link the different components of if other proven policy alternatives risks due to tobacco, alcohol, a campaign, increase visibility such as taxation of unhealthy unhealthy diet and physical and create a platform for foods and tobacco are also used inactivity. participation to reduce NCD risks.  ontinue to promote taxation •C interventions, the endorsed policy for all PICs in tackling NCDs and curb unhealthy diets (with World Bank support on assessing policy impacts and evidence-based policy making) 45 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA Improve PHC-level chronic care provision by redesigning how services are 2.  delivered, making better use of outreach, triage, decentralisation and task shifting in providing people- centered care, drawing from the PEN Fa’a approach Why? What are the experiences? Implementation tips Samoan patients frequently Many countries need to re-  ask-shift in two specific areas: •T express low satisfaction with purpose their PHC services due a) Village women committees to services, mentioning long waiting to the rapidly increasing burden take on more functions of health times, lack of friendliness and of NCDs. Several principles apply promotion, patient follow up and difficult access to services as universally in PHC reform, among screening, under the guidance main barriers to service use. them: and supervision of rural district hospitals, and b) Nurses to take Demand for NCD screening is Better triage – Tracks for acute on more community outreach on insufficient and likely linked to versus chronic versus MCH clients screening and health promotion perceived quality of care. in facilities, with better service to ensure primary prevention, configuration for chronic patients case finding and linkage to care The provision of chronic care still relies heavily on doctors, while reducing waiting times.  stablish multidisciplinary care •E PHC-level and other health cadres Decentralisation – Providing teams in the rural hospitals with clearly defined Terms of are insufficiently used for chronic- services closer to home, Reference, which includes a care services. especially community—based primary care physician who can Suggestions concerning health case detection and support to effectively address the treatment facility capacity to provide patients on long-term treatments. gap efficient, flexible and friendly NCD Task shifting – Empowering  esignate PHC workers as care •D services received a lot of support nurses or community-based coordinators and care providers in the dialogues held during this health workers with skills to for NCD patients, who are close to study. support NCD care inside and the patients and work with them outside the health facility. to agree on a treatment plan  ssess patient flow at each •A People centered care – Services of the facilities and identify are provided to respond to the improvements so that chronic specific needs and preferences patients can receive care more of the beneficiaries of care. This efficiently and waiting times requires that patients have the can be reduced (consider education and support to make chronic track, outreach or NCD decisions and participate in their clinics, different opening hours, care. It is organized around the reassignment of roles and health needs and expectations of responsibilities in health cadres) patients rather than diseases. mprove behaviours of frontline •I staff through a “hospitality sector- type” training so patients are seen as clients and received with a more welcoming and supportive attitude  rain health personnel on the •T specific service needs of patients with largely asymptomatic conditions  astly, make needs-based •L investment in equipment and infrastructure supporting quality of care – for instance, aautomated (digital) monitors improve diagnosis and patient monitoring, and can contribute to team-based care and task sharing 46 RECOMMENDATIONS Ensure the patient-level medical record system is more effectively used to 3.  support frontline staff in providing chronic care to patients Why? What are the experiences? Implementation tips Too many patients are defaulting Patient-level information  nsure mobile phone numbers •E care without being traced, missing systems are a key component are kept updated in patient monitoring tests and failing for health systems to provide records as they are invaluable treatment targets, and health efficient and effective patient- for patient tracing and personnel does not have the centered care. Using specific appointment reminders for high- capacity to target retention efforts performance indicators, they risk patients to the patients in need. can show success and failure at  lan visit frequencies with the •P In the longer-term, the patient- different levels including patient, patient for drug refill, laboratory level information system must be clinic and system levels. To be monitoring, etc. to reduce loss- upgraded to an electronic medical useful to frontline staff, real- to-follow-up record system (EMR) to further time data dashboards, lists and exception reports drawn from the  enerate lists of patients with •G enhance patient-centered care missing monitoring data (BP, and close gaps in the continuum patient information system are required for ensuring patients are cholesterol, glucose, etc), of care. missing visits, prescription retained in care and receive the appropriate services. gaps, and other quality-related measures  s the future EMR functionalities •A are developed, ensure that any customisation of the software caters for SMS appointment reminders, patient lists for tracing lost-to-follow-up by facility, automated reports and other tools aiding patient retention mprove patient experience also •I by selectively decentralizing point-of-care tests to rural district hospitals, that can provide a result within minutes while the NCD patient is visiting, and can be recorded in the patient’s medical record for ongoing management (this should be limited to tests which can feasibly and sustainably be decentralized) 47 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA Scale up the PEN Fa’ a Samon program to more villages, and leverage the 4.  comparative advantages of village-based groups and individuals to improve the effectiveness of NCD screening, referral, diagnosis and treatment Why? What are the experiences? Implementation tips Samoa’s NCD strategy cannot Public health approaches  uild on the PEN Fa’a •B rely on the efforts of the health rely on community resources experience by expanding what facilities alone. especially in health promotion, worked well, and considering Comments of patients and prevention, linkage to care, a broadened intervention providers on how healthcare and and comprehensive support of package to address gaps in medication adherence for HTN patients. Community involvement the continuum of care and and other chronic conditions has been essential in many primary prevention (including could be improved was frequently other successful development promotion of screening, efficient about the need for better activities in Samoa, and the refill mechanisms for long-term community involvement. PEN Fa’a intervention also medications, outreach NCD successfully drew on community- clinics to remote areas, public Providers echoed patients’ based resources. Internationally, education on NCDs, healthy comments that the women’s faith-based organisations have eating/obesity prevention) committees in the village could also participated successfully play a central role, but described  arve out a significant role for •C in aspects of health promotion women’s committees given a multi-level approach that would and case detection activities. include mass media, the village the capacity of these groups Individuals can be trained as peer to complement health facility community, families, and patients educators and patient navigators themselves. efforts (see above on task to maximise retention in care of shifting) Published evidence on the NCD patients. capacity of women’s committee is  ssess potential roles for church •A in line with this25. groups and leaders in NCD prevention and care  omplete the work on adapting •C the CVD risk chart piloted in PEN Fa’a Samoa to reflect Samoa population features (weight, genetic factors) and become a tool which can be scaled up across all screening activities and help target care to individuals at greatest risk of CVD events 48 RECOMMENDATIONS Systematically address the bottlenecks and barriers to NCD treatment adherence: 5.  Improve access and availability of medications; Strengthen provider understanding of barriers to adherence; and Ensure the provision of patient- centered adherence support Why? What are the experiences? Implementation tips This assessment found that Research on treatment adherence mprove access to essential •I outcomes are poor among barriers usually finds that the medicines through facility- Samoan patients who are underlying causes of non- specific essential NCD drug list prescribed HTN treatment. adherence vary greatly. They and drug supply planning linking The qualitative work showed that can be as diverse as denial of coverage and treatment data barriers to treatment adherence the condition, prohibitive drug with drug forecasting are patient-specific and varied, costs, drug accessibility, poor  stablish accountability •E there is no solution which fits all. understanding of the regimen or mechanism for drug the condition (‘feeling better’), management and supply Therefore, the root causes of side effects, forgetfulness, non-adherence need to be planning while enhancing the depression, running out of compliance with standard better investigated by clinical drugs due to travel, etc. Unless personnel, at the individual level, operating procedures for daily the treating doctor or nurse monitoring of essential drugs and addressed through tailored understands the root cause of counselling and patient support. non-adherence, it is impossible to  nsure clinical NCD guidelines •E address the barriers successfully. are aligned with what WHO recommends and include simple, drug- and dose-specific guidance (in progress through World Bank support)  apacitate health personnel •C with better knowledge about continuity of care in NCDs, and how to minimise the known barriers at each stage of the continuum  hare practical tips on •S adherence promoting strategies with health personnel, which can be used with patients struggling to remain on treatment or fail treatment targets  roduce and disseminate a •P patient information leaflet on treatment adherence and how to overcome the different barriers  xplore and test new solutions •E to support patients better, such as adherence clubs, longer prescription cycles, social media chat groups 49 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA Ensure body weight monitoring and the management of overweight and obesity 6.  is an integral part of hypertension and diabetes care Why? What are the experiences? Implementation tips Food, its preparation and the Increased focus on bodyweight  CD care guidelines •N sharing of meals have large and BMI by health personnel and standard operating cultural and social importance in addresses the obesity problem in procedures must integrate Samoa. multiple ways. It communicates weight monitoring and active However, ultraprocessed to patients that body weight is an management and unhealthy food imports important parameter, is part of the  easure body weight at every •M combined with lifestyle changes clinical management and changes visit, communicate results and have precipitated a severe are being tracked. It provides time trends to patients in a epidemic of obesity, and lifestyle health staff with opportunities non-judgemental, supportive diseases which largely stem from to address body weight in way, and take clinical action as overnutrition. patient feedback and treatment needed counselling. It also enables health The patient data showed that staff to spot weight gain and the  nsure patients understand •E bodyweight is poorly measured, presence of severe obesity, which the link between obesity and recorded and tracked in HTN may require patient referral to a treatment success (high BMI patients. dietician. makes it harder to reach the Therefore, weight loss is not a target BP level) prominent part of the clinical  mphasize the multiple benefits •E management of HTN patients. of lower body weight on people’s risk to get diabetes, CVD, cancers, problems with joints, etc. ncrease Samoas’ capacity of •I dieticians and nutrition workers 50 CITATIONS AND NOTES Citations and Notes Global burden of disease, YYL metric: https://vizhub.healthdata.org/gbd-compare/  1 2 World Health Organisation/Samoa Ministry of Health 2014. Samoa NCD risk factors STEPS report. Apia: Ministry  of Health; 2014. 3 Linhart C, Naseri T, Lin S, Taylor R, Morrell S, McGarvey ST, Magliano DJ, Zimmet P. 2018. Continued increases in  blood pressure over two decades in Samoa (1991–2013); around one-third of the increase explained by rising obesity levels. BMC Public Health 18:1122. https://doi.org/10.1186/s12889-018-6016-2 4 Lin S, Naseri T, Linhart C, Morrell S, Taylor R, McGarvey ST, Magliano DJ, Zimmet P 2017. Trends in diabetes and  obesity in Samoa over 35 years, 1978–2013. Diabet. Med. 34, 654–661. 5 MOH 2018. National non-communicable disease policy (2019-2023), Apia, Ministry of Health  6 Bollars C, Naseri T, Thomsen R et al. 2018. Adapting the WHO package of essential noncommunicable disease  interventions, Samoa. Bulletin World Health Organ 2018;96:578–583 7 Institute for Health Policy Sri Lanka and Centre for Health Information Policy and Systems Research, Fiji. 19  December 2017 8 Kessaram T, Mckenzie J, Girin N, Roth A, Vivili P, Williams G, Hoy D. (2015). Non-communicable diseases and  risk factors in adult populations of several Pacific Islands: Results from the WHO STEPwise approach to surveillance. Australian and New Zealand journal of public health. 39. 10.1111/1753-6405.12398 9 Global burden of disease, YYL metric: https://vizhub.healthdata.org/gbd-compare/  10 World Health Organisation and Samoa Ministry of Health 2014. Samoa NCD risk factors STEPS report. Apia:  Ministry of Health; 2014. 11 Institute for Health Policy Sri Lanka and Centre for Health Information Policy and Systems Research, Fiji. 19  December 2017 12 World Bank. 2019. Hypertension Care in Ukraine : Breakpoints and Implications for Action. World Bank,  Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/31155 License: CC BY 3.0 IGO 13 Price AJ, Crampin AC, Amberbir A et al. 2018. Prevalence of obesity, hypertension, and diabetes, and cascade  of care in sub-Saharan Africa: a cross-sectional, population-based study in rural and urban Malawi. Lancet Diabetes Endocrinol 2018; 6: 208–22 14 Mutemwa M, Peer N, de Villiers A et al. 2018. Prevalence, detection, treatment, and control of hypertension in  human immunodeficiency virus (HIV)-infected patients attending HIV clinics in the Western Cape Province, South Africa. Medicine (2018) 97:35(e12121) 15 Strategy for the Development of Samoa 2017-2020: Accelerating Sustainable Development and Broadening  Opportunities for all, Dec 2016 51 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA 16 Ministry of Health Samoa 2008. Health sector plan 2008-2018. Apia (Samoa), 2008  Implementation tools: Package of essential noncommunicable (PEN) disease interventions for primary health  17 care in low-resource settings. Geneva: WHO; 2013. http://apps.who.int/iris/bitstream/ handle/10665/133525/9789241506557_eng.pdf?sequence=1 18 Bollars C, Naseri T, Thomsen R et al. 2018. Adapting the WHO package of essential noncommunicable disease  interventions, Samoa. Bull World Health Organ 2018;96:578–583 19 Samoa has a small urban population which is largely residing around the main hospital in the capital of Apia.  20 Bollars C, Naseri T, Thomsen R et al. 2018. Adapting the WHO package of essential noncommunicable disease  interventions, Samoa. Bull World Health Organ 2018;96:578–583 21 Our prevalence of elevated BP of 35.6% in the Cascade Study 2018 Household Survey was higher than the  prevalence reported from the 2013 STEPS survey (24.5%). This could partly be due to different age composition of the survey samples. While STEPS only included adults up to 64 years, the 2018 survey had 15% of its participants aged above 64 years (171/1109 non-pregnant adults). There could also have been differences in BP measurement methodology. We quality-assured the BP measurements by using well-calibrated equipment, by averaging two BP measurements and spacing them, and by taking measurements once the survey participants were well settled into the interview. The average interval between the two BP measurements was 4 minutes 13 seconds. We also recorded the recent intake of caffeinated drinks, as these could raise BP temporarily. About 4 in 10 survey participants had consumed such drinks shortly before the survey. However, we consider the measurement results as valid reflection of the BP levels of survey respondents. 22 The time window is measured from when the laboratory test results of the last visit were captured as there can  be a lag time in receiving some tests from the national laboratory 23 Baghirov R, Ah-Ching J & Bollars C 2019. Achieving UHC in Samoa through Revitalizing PHC and Reinvigorating  the Role of Village Women Groups, Health Systems & Reform, 5:1, 78-82, DOI: 10.1080/23288604.2018.1539062 24 As part of the Global Burden of Diseases, Injuries, and Risk Factors Study 2017 (GBD 2017), the authors  measured progress on 41 of 52 health-related SDG indicators and estimated the health-related SDG index for 195 countries and territories for the period 1990–2017 and projected indicators to 2030. To generate projections through 2030, they used a forecasting framework that drew estimates from the broader GBD study and used weighted averages of indicator-specific and country-specific annualized rates of change from 1990 to 2017 to inform future estimates. The methodology and findings of this analysis are further detailed in: GBD 2017 SDG Collaborators, 2018. Measuring progress from 1990 to 2017 and projecting attainment to 2030 of the health-related Sustainable Development Goals for 195 countries and territories: a systematic analysis for the Global Burden of Disease Study 2017. The Lancet. 8 Nov 2018; 392:2091–138. doi: http://dx.doi.org/10.1016/ S0140-6736(18)32281-5. 25 Baghirov R, Ah-Ching J & Bollars C 2019. Achieving UHC in Samoa through Revitalizing PHC and Reinvigorating  the Role of Village Women Groups, Health Systems & Reform, 5:1, 78-82, DOI: 10.1080/23288604.2018.1539062 52 Appendices CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA Appendix 1. Cascade Study 2018 Household Survey HOUSEHOLD SURVEY RESPONSE RATES • Response rate was lower in Savai’i Island villages (Neiafu and Vaisala) compared to Upolu Island villages (Mulifanua and Fagaloa). This was due to challenges in recruitment of survey participants – specifically the small population size of the selected villages (see footnote) and the proximity of the recruitment activities to the Christmas holidays (which affected access to space and participants). We did not experience any active refusal of participation - eligible participants were randomly selected from • household lists and approached for participation. Non-participation was always because the participant could not be located (had left the village temporarily, was working in a remote plantation, etc.) All participants have complete BP data. One participant from Fagaloa only had one measurement of BP (due to • research assistant error in recording the 2nd reading). This participant was included in analysis. Based on village size, it was necessary to recruit more than one participant per household in many cases. • Participants per household ranged from 1 – 9. There were no differences in mean number of participants per household by village, by island (Upolu vs. Savai’i), or by PEN vs. non-PEN. A unique household ID has been included in the dataset so that weighting may be applied at a later data if required. [Please note that the household ID is unique to this study and does not correspond to the household number in the Samoan census; therefore, this is not identifying information] TABLE 3. SURVEY RESPONSE RATES BY POPULATION AND VILLAGE PEN villages Non-PEN villages Upolu Island: Savai’I Island: Upolu Island: Savai’I Island: Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) Men aged 20+ years Number targeted 150 150 150 150 Number interviewed 153 141 154 106 Number with BP data 153 141 154 106 Response rate (BP data/targeted) 102.0 94.0 a 102.7 70.7a Non-pregnant women aged 20+ years Number targeted 150 150 150 150 Number interviewed 150 131 162 112 Number with BP data 150 131 162 112 Response rate (BP data/targeted) 100.0 87.3 a 108.0 74.7a Women with current/recent pregnancy Number targeted - - - - Number interviewed 36 16 40 6 Number with BP data 36 16 40 6 Response rate (preg w with BP data/targeted) N/A N/A N/A N/A 54 APPENDIX 1 TABLE 3. SURVEY RESPONSE RATES BY POPULATION AND VILLAGE cont/… PEN villages Non-PEN villages Upolu Island: Savai’I Island: Upolu Island: Savai’I Island: Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) Total survey population Number targeted 300 300 300 300 Number interviewed 300 272 312 218 Number with BP data 300 272 312 218 Response rate (BP data/targeted) 100.0 90.7 104.0 72.7 Households Number of respondents (median) 1.0 2.0 2.0 1.0 a The primary challenges to recruitment in Vaisala and Neifau were small village size (total population [including children] in Vaisala  <600; Neiafu ~900 (Census, 2011)) and the conduct of the survey in mid-December – preparations for Christmas holidays affected willingness to participate. DEMOGRAPHIC COMPOSITION OF SURVEY PARTICIPANTS Differences between PEN and non-PEN villages were explored for sociodemographic variables hypothesized to have an association with either HTN prevalence or treatment seeking behaviors (age, education, marital status, employment). The summary below focuses on differences among the total population. There were no differences in the age composition of PEN and non-PEN villages • A greater proportion of participants in PEN villages were married; non-pregnant women were responsible for this • difference (81.9% of women in PEN villages were married vs. 73.4% in non-PEN villages). There were no differences in marital status for men or recently pregnant/currently pregnant women. A greater proportion of participants in PEN villages had a college/university education compared to non-PEN • villages (41.8% vs. 35.3%), this was borderline statistically significant (p=0.05; driven by the non-pregnant women). There was, however no difference in the average number of years of education (11.4 vs. 11.3 years) in PEN and non-PEN villages, respectively. There were significant differences in employment status between PEN and non-PEN villages, but the differences • driving the association were in the less common categories (self-employed, retired, other), not unemployed vs. employed. TABLE 4. DEMOGRAPHIC CHARACTERISTICS BY POPULATION AND VILLAGE PEN villages Non-PEN villages p-value PEN Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) vs. non-PEN Men aged 20+ years Age group 20-39 years 55 (35.9) 60 (42.6) 68 (44.2) 38 (35.8) p=0.93 40-59 years 54 (35.3) 56 (39.7) 52 (33.8) 43 (40.6) 60-69 years 26 (17.0) 14 (9.9) 21 (13.6) 16 (15.1) 70+ years 18 (11.8) 11 (7.8) 13 (8.4) 9 (8.5) All ages 153 (100.0) 141 (100.0) 154 (100.0) 106 (100.0) Median age 51.4 43.1 44.0 48.3 Marital status Never married 40 (26.1) 43 (30.5) 45 (29.2) 29 (27.4) p=0.24 Currently married 106 (69.3) 93 (66.0) 102 (66.2) 65 (61.3) 55 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA TABLE 4. DEMOGRAPHIC CHARACTERISTICS BY POPULATION AND VILLAGE cont/… PEN villages Non-PEN villages p-value PEN Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) vs. non-PEN Men aged 20+ years cont/… Marital status cont/… Divorced/separated 2 (1.3) 1 (0.7) 0 (0.0) 2 (1.9) Widowed 2 (1.3) 3 (2.1) 4 (2.6) 1 (0.9) Cohabitating 3 (2.0) 1 (0.7) 3 (1.9) 9 (8.5) Ethnicity Samoan 153 (100.0) 141 (100.0) 153 (99.4) 106 (100.0) Part Samoan 0 (0.0) 0 (0.0) 1 (0.6) 0 (0.0) Other 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Denomination Catholic 19 (12.4) 0 (0.0) 27 (17.5) 3 (2.8) Methodist 11 (7.2) 105 (74.5) 0 (0.0) 0 (0.0) EFKS/Congregational 68 (44.4) 1 (0.7) 58 (37.7) 82 (77.4) Christian Church Seventh Day Adventist 3 (2.0) 0 (0.0) 0 (0.0) 4 (3.8) Latter Day Saints 39 (25.5) 35 (24.8) 49 (31.8) 10 (9.4) Assembly of God 5 (3.3) 0 (0.0) 18 (11.7) 7 (6.6) Othera 8 (5.2) 0 (0.0) 1 (0.6) 0 (0.0) Refused to Answer 0 (0.0) 0 (0.0) 1 (0.6) 0 (0.0) Education Less than full primary school 14 (9.2) 7 (5.0) 12 (7.8) 11 (10.4) p=0.87 Primary school completed 40 (26.1) 28 (19.9) 43 (27.9) 19 (17.9) Secondary school completed 53 (34.6) 46 (32.6) 50 (32.5) 32 (30.2) College/University completed 46 (30.1) 60 (42.6) 49 (31.8) 44 (41.5) p=0.84 Median education (years) b 12.0 12.0 12.0 12.0 Employment Unemployed looking for work 31 (20.3) 29 (20.6) 46 (29.9) 14 (13.2) p=0.10 Unemployed not looking for work 61 (39.9) 68 (48.2) 63 (40.9) 52 (49.1) Self-employed 10 (6.5) 1 (0.7) 13 (8.4) 3 (2.8) Employed 32 (20.9) 30 (21.3) 19 (12.3) 27 (25.5) Retired or unable to work 19 (12.4) 13 (9.2) 10 (6.5) 9 (8.5) Other 0 (0.0) 0 (0.0) 3 (1.9) 1 (0.9) Non-pregnant women aged 20+ years Age group 20-39 years 48 (32.0) 42 (32.1) 66 (40.7) 38 (33.9) p=0.32 40-59 years 67 (44.7) 61 (46.6) 62 (38.3) 42 (37.5) 60-69 years 26 (17.3) 15 (11.5) 23 (14.2) 20 (17.9) 70+ years 9 (6.0) 13 (9.9) 11 (6.8) 12 (10.7) All ages 150 (100.0) 131 (100.0) 162 (100.0) 112 (100.0) Median age (years) 48.6 46.4 44.6 48.4 56 APPENDIX 1 TABLE 4. DEMOGRAPHIC CHARACTERISTICS BY POPULATION AND VILLAGE cont/… PEN villages Non-PEN villages p-value PEN Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) vs. non-PEN Non-pregnant women aged 20+ years cont/… Marital Status Never married 16 (10.7) 13 (9.9) 26 (16.0) 17 (15.2) p=0.04 Currently married 122 (81.3) 108 (82.4) 117 (72.2) 84 (75.0) Divorced/separated 2 (1.4) 3 (2.3) 0 (0.0) 2 (1.8) Widowed 9 (6.0) 7 (5.3) 16 (9.9) 5 (4.5) Cohabitating 1 (0.7) 0 (0.0) 3 (1.9) 4 (3.6) Ethnicity Samoan 148 (98.7) 131 (100.0) 162 (100.0) 111 (99.1) Part Samoan 2 (1.3) 0 (0.0) 0 (0.0) 1 (0.9) Other 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Denomination Catholic 22 (14.7) 0 (0.0) 34 (21.0) 6 (5.4) Methodist 22 (14.7) 95 (72.5) 1 (0.6) 0 (0.0) EFKS/Congregational 51 (34.0) 1 (0.8) 62 (38.3) 85 (75.9) Christian Church Seventh Day Adventist 0 (0.0) 1 (0.8) 1 (0.6) 2 (1.8) Latter Day Saints 41 (27.3) 34 (26.0) 45 (27.8) 12 (10.7) Assembly of God 8 (5.3) 0 (0.0) 14 (8.6) 7 (6.3) Othera 5 (3.3) 0 (0.0) 4 (2.5) 0 (0.0) Refused to Answer 1 (0.7 0 (0.0 1 (0.6) 0 (0.0) Education Less than full primary school 55 (35.9) 60 (42.6) 68 (44.2) 38 (35.8) p<0.01 Primary school completed 54 (35.3) 56 (39.7) 52 (33.8) 43 (40.6) Secondary school completed 26 (17.0) 14 (9.9) 21 (13.6) 16 (15.1) College/University completed 18 (11.8) 11 (7.8) 13 (8.4) 9 (8.5) p=0.16 Median education (years) b 153 (100.0) 141 (100.0) 154 (100.0) 106 (100.0) Employment Unemployed looking for work 33 (22.0) 27 (20.6) 56 (34.6) 8 (7.1) p<0.01 Unemployed not looking for work 74 (49.3) 82 (62.6) 82 (50.6) 60 (53.6) Self-employed 4 (2.7) 0 (0.0) 5 (3.1) 2 (1.8) Employed 12 (8.0) 6 (4.6) 11 (6.8) 28 (25.0) Retired or unable to work 25 (16.7) 16 (12.2) 8 (4.9) 12 (10.7) Other 2 (1.3) 0 (0.0) 0 (0.0) 2 (1.8) Women with current/recent pregnancy Age group 20-39 years 31 (86.1) 13 (81.3) 40 (100.0) 5 (83.3) p=0.03c 40-59 years 5 (13.9) 3 (18.7) 0 (0.0) 1 (16.7) All ages 36 (100.0) 16 (100.0) 40 (100.0) 6 (100.0) Median age 27.9 28.6 28.4 30 4 57 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA TABLE 4. DEMOGRAPHIC CHARACTERISTICS BY POPULATION AND VILLAGE cont/… PEN villages Non-PEN villages p-value PEN Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) vs. non-PEN Women with current/recent pregnancy cont/… Marital status Never married 3 (8.3) 1 (6.3) 6 (15.0) 2 (33.3) p=0.34 Currently married 29 (80.6) 15 (93.8) 31 (77.5) 4 (66.7) Divorced/separated 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Widowed 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Cohabitating 4 (11.1) 0 (0.0) 3 (7.5) 0 (0.0) Ethnicity Samoan 36 (100.0) 16 (100.0) 38 (95.0) 6 (100.0) Part Samoan 0 (0.0) 0 (0.0) 2 (5.0) 0 (0.0) Other 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Denomination Catholic 5 (13.9) 0 (0.0) 13 (32.5) 0 (0.0) Methodist 4 (11.1) 14 (87.5) 0 (0.0) 0 (0.0) EFKS/Congregational 9 (25.0) 0 (0.0) 13 (32.5) 5 (83.3) Christian Church Seventh Day Adventist 1 (2.8) 0 (0.0) 0 (0.0) 0 (0.0) Latter Day Saints 14 (38.9) 2 (12.5) 10 (25.0) 0 (0.0) Assembly of God 2 (5.6) 0 (0.0) 3 (7.5) 1 (16.7) Other a 0 (0.0) 0 (0.0) 1 (2.5) 0 (0.0) Refused to Answer 1 (2.8) 0 (0.0) 0 (0.0) 0 (0.0 Education Less than full primary school 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) p=0.40 Primary school completed 9 (25.0) 1 (6.3) 6 (15.0) 0 (0.0) Secondary school completed 8 (22.2) 6 (37.5) 18 (45.0) 0 (0.0) College/University completed 19 (52.8) 9 (56.3) 16 (40.0) 16 (100.0) p=0.51 Median education (years) b 13.0 13.0 12.0 13.0 Employment Unemployed looking for work 12 (33.3) 2 (12.5) 15 (37.5) 1 (16.7) p=0.69 Unemployed not looking for work 20 (55.6) 13 (81.3) 22 (55.0) 4 (66.7) Self-employed 1 (2.8) 0 (0.0) 1 (2.5) 0 (0.0) Employed 3 (8.3) 1 (6.3) 1 (2.5) 1 (16.7) Retired or unable to work 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Other 0 (0.0) 0 (0.0) 1 (2.5) 0 (0.0) Total survey population Age group 20-39 years 134 (39.5) 115 (39.9) 174 (48.9) 81 (36.2) p=0.34 40-59 years 126 (37.2) 120 (41.7) 114 (32.0) 86 (38.4) 60-69 years 52 (15.3) 29 (10.1) 44 (12.4) 36 (16.1) 70+ years 27 (8.0) 24 (8.3) 24 (6.7) 21 (9.4) 58 APPENDIX 1 TABLE 4. DEMOGRAPHIC CHARACTERISTICS BY POPULATION AND VILLAGE cont/… PEN villages Non-PEN villages p-value PEN Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) vs. non-PEN Total survey population cont/… Age group cont/… All ages 339 (100.0) 288 (100.0) 356 (100.0) 224 (100.0) Median age 46.7 44.5 41.1 47.8 Marital status Never married 59 (17.4) 57 (19.8) 77 (21.6) 48 (21.4) p=0.03 Currently married 257 (75.8) 216 (75.0) 250 (70.2) 153 (68.3) Divorced/separated 4 (1.2) 4 (1.3) 0 (0.0) 4 (1.8) Widowed 11 (3.2) 10 (3.5) 20 (5.6) 6 (2.7) Cohabitating 8 (2.4) 1 (0.3) 9 (2.5) 13 (5.8) Ethnicity Samoan 337 (99.4) 288 (100.0) 353 (99.2) 223 (99.6) Part Samoan 2 (0.6) 0 (0.0) 3 (0.8) 1 (0.4) Other 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Denomination Catholic 46 (13.6) 0 (0.0) 74 (20.8) 9 (4.0) Methodist 37 (10.9) 214 (74.3) 1 (0.3) 0 (0.0) EFKS/Congregational 128 (37.8) 2 (0.7) 133 (37.4) 172 (76.8) Christian Church Seventh Day Adventist 4 (1.2) 1 (0.3) 1 (0.3) 6 (2.7) Latter Day Saints 94 (27.7) 71 (24.7) 104 (29.2) 22 (9.8) Assembly of God 15 (4.4) 0 (0.0) 35 (9.8) 15 (6.7) Othera 13 (3.8) 0 (0.0) 6 (1.7) 0 (0.0) Refused to Answer 2 (0.6) 0 (0.0) 2 (0.6) 0 (0.0) Education Less than full primary school 19 (5.6) 13 (4.5) 26 (7.3) 20 (8.9) p=0.05 Primary school completed 75 (22.1) 49 (17.0) 81 (22.8) 37 (16.5) Secondary school completed 114 (33.6) 95 (33.0) 140 (39.3) 71 (31.7) College/University completed 131 (38.6) 131 (45.5) 109 (30.6) 96 (42.9) Median education (years)b 12.0 12.0 12.0 12.0 p=0.39 Employment Unemployed looking for work 76 (22.4) 58 (20.1) 117 (32.9) 23 (10.3) p=0.01 Unemployed not looking for work 155 (45.7) 163 (56.6) 167 (46.9) 116 (51.8) Self-employed 15 (4.4) 1 (0.3) 19 (5.3) 5 (2.2) Employed 47 (13.9) 37 (12.8) 31 (8.7) 56 (25.0) Retired or unable to work 44 (13.0) 29 (10.1) 18 (5.1) 21 (9.4) Other 2 (0.6) 0 (0.0) 4 (1.1) 3 (1.3) a  he most commonly reported ‘Other’ response for religious denomination was Nazarene; T b Mean difference in years of education examined using a two-sided t-test; c P value based on a Fisher’s exact test due to small sample size. 59 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA HOUSEHOLD SURVEY RESPONSE RATES • Response rate was lower in Savai’i Island villages (Neiafu and Vaisala) compared to Upolu Island villages (Mulifanua and Fagaloa). This was due to challenges in recruitment of survey participants – specifically the small population size of the selected villages (see footnote) and the proximity of the recruitment activities to the Christmas holidays (which affected access to space and participants). We did not experience any active refusal of participation - eligible participants were randomly selected from • household lists and approached for participation. Non-participation was always because the participant could not be located (had left the village temporarily, was working in a remote plantation, etc.) All participants have complete BP data. One participant from Fagaloa only had one measurement of BP (due to • research assistant error in recording the 2nd reading). This participant was included in analysis. Based on village size, it was necessary to recruit more than one participant per household in many cases. • Participants per household ranged from 1 – 9. There were no differences in mean number of participants per household by village, by island (Upolu vs. Savai’i), or by PEN vs. non-PEN. A unique household ID has been included in the dataset so that weighting may be applied at a later data if required. [Please note that the household ID is unique to this study and does not correspond to the household number in the Samoan census; therefore, this is not identifying information] TABLE 5. SOCIO-ECONOMIC CHARACTERISTICS OF HOUSEHOLDS BY VILLAGE PEN villages Non-PEN villages p-value PEN Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) vs. non-PEN Average weekly household income including any remittances 100 WST 20 (5.9) 26 (9.0) 17 (4.8) 19 (8.5) p<0.01 100 - 250 WST 143 (42.1) 154 (53.5) 107 (30.1) 121 (54.0) Above 250 WST 171 (50.4) 106 (36.8) 223 (62.6) 82 (36.6) Don’t know 5 (1.5) 2 (0.7) 8 (2.2) 1 (0.4) Refused to answer 0 (0.0) 0 (0.0) 1 (0.3) 1 (0.4) Main source of household income a Employment 140 (41.3) 94 (32.6) 55 (15.4) 107 (47.8) Farming 37 (10.9) 51 (17.7) 48 (13.5) 19 (8.5) Fishing 52 (15.3) 10 (3.5) 14 (3.9) 23 (10.3) Remittances 71 (20.9) 94 (32.6) 68 (19.1) 56 (25.0) Handicrafts 2 (0.6) 0 (0.0) 83 (23.3) 1 (0.4) Selling Produce 17 (5.0) 24 (8.3) 66 (18.5) 9 (4.0) Small Business Owner 16 (4.7) 13 (4.5) 16 (4.5) 8 (3.6) Retirement Income 2 (0.6) 1 (0.3) 1 (0.3) 1 (0.4) Other 1 (0.3) 0 (0.0) 4 (1.1) 0 (0.0) Don’t Know 0 (0.0) 0 (0.0) 1 (0.3) 0 (0.0) Missing 1 (0.3) 1 (0.0) 0 (0.0) 0 (0.0) Ownership of selected household items None 30 (8.8) 25 (8.7) 17 (4.8) 16 (7.1) 1-2 items 119 (35.1) 141 (49.0) 194 (54.5) 100 (44.6) 3-4 items 162 (47.8) 117 (40.6) 137 (38.5) 95 (42.4) All 5 items 28 (8.3) 5 (1.7) 8 (2.2) 13 (5.8) p<0.01b Median number of items 3.0 2.0 2.0 2.0 60 APPENDIX 1 TABLE 5. SOCIO-ECONOMIC CHARACTERISTICS OF HOUSEHOLDS BY VILLAGE PEN villages Non-PEN villages p-value PEN Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) vs. non-PEN Average weekly household spend on all goods and services Median (WST) 100.0 100.0 150.0 100.0 Did not spend on goods and 0 (0.0) 1 (0.3) 1 (0.3) 1 (0.4) services Don’t knowc 23 (6.8) 9 (3.2) 7 (1.9) 16 (7.1) Refused to answer 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Percent spent on health services of total spend (calculated for last 12 months) Median (%)d 1.9 1.7 1.3 1.5 Did not spend on health 30 (8.8) 27 (9.4) 34 (9.6) 17 (7.6) services p=0.60e Don’t knowc 41 (12.1) 40 (13.9) 16 (4.5) 38 (16.9) Refused to answer 1 (0.3) 0 (0.0) 0 (0.0) 0 (0.0) a Employment was interpreted as meaning ‘formal employment’ rather than the other forms of self-employment listed; remittances was  extended to include ‘donations’ from family or church; ‘Other’ contains occupations such as village midwife, brick layer, etc. Fagaloa has a high proportion of kava growers as a part of this ongoing program: http://www.sobserver.ws/en/23_06_2018/local/34336/Im- proving-ava-production-in-Samoa.htm; b p value based on a t-test comparing mean number of items owned between PEN and non-PEN villages; c  ‘Don’t Know’ includes both those who responded “don’t know” and those who said “yes, but I don’t know how much” – the latter was the more frequent of the two responses; d Calculated only among n=926 participants who gave amounts for both expenditures and who reported spending anything on health  care services; e Two-sided t-test comparing mean % spend RESTING BLOOD PRESSURE BY POPULATION AND VILLAGE The two BP measurements taken on each participant were averaged for use in these analyses. There were no significant differences among the sample based on village participation in the PEN program. These analyses did not exclude participants who were taking medications to control BP. This should be taken into account in interpretation. TABLE 6. RESTING BLOOD PRESSURE BY POPULATION AND VILLAGE PEN villages Non-PEN villages Mulifanua Neiafu Fagaloa Vaisala p-value PEN vs. non-PEN Men aged 20+ yearss Mean SBP (95% CI) 20-39 years 127.3 (123.8, 130.7) 128.4 (125.1, 131.7) 130.7 (128.0, 133.5) 132.2 (128.9, 135.5) 40-59 years 139.1 (133.6, 144.5) 135.7 (131.5, 139.9) 135.8 (130.0, 141.6) 136.6 (132.2, 140.9) 60-69 years 142.2 (136.4, 147.9) 148.1 (129.0, 167.1) 130.4 (124.6, 136.1) 130.3 (123.7, 136.8) 70+ years 143.3 (133.2, 153.5) 145.4 (131.6, 159.2) 132.6 (120.0, 145.2) 143.1 (120.8, 165.4) All ages 135.9 (133.0, 138.7) 134.6 (131.5, 137.7) 132.6 (130.0, 135.1) 134.6 (131.8, 137.4) p=0.20a Mean DBP (95% CI) 20-39 years 78.5 (75.8, 81.2) 78.1 (75.0, 81.1) 83.0 (80.2, 85.8) 82.4 (78.7, 86.0) 40-59 years 86.7 (83.5, 89.9) 82.6 (79.7, 85.5) 88.0 (84.1, 91.9) 85.7 (83.0, 88.4) 60-69 years 83.0 (78.5, 87.5) 89.8 (78.7, 100.8) 82.9 (79.0, 86.8) 77.5 (71.7, 83.4) 70+ years 82.8 (76.1, 89.4) 89.0 (81.9, 96.0) 79.2 (73.4, 84.9) 78.3 (67.7, 89.0) All ages 82.7 (80.8, 84.5) 81.9 (79.7, 84.0) 84.4 (82.4, 86.3) 82.6 (80.6, 84.7) p=0.18a 61 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA TABLE 6. RESTING BLOOD PRESSURE BY POPULATION AND VILLAGE cont/… PEN villages Non-PEN villages p-value PEN Mulifanua Neiafu Fagaloa Vaisala vs. non-PEN Non-pregnant women 20+ years Mean SBP (95% CI) 20-39 years 113.6 (109.6, 117.6) 116.3 (111.6, 120.9) 114.7 (111.5, 117.9) 114.0 (110.4, 117.6) 40-59 years 128.1 (123.6, 132.7) 128.1 (123.3, 132.9) 124.7 (121.3, 128.1) 127.3 (122.1, 132.5) 60-69 years 137.7 (128.7, 146.6) 133.8 (117.9, 149.8) 132.9 (122.8, 143.0) 139.6 (130.1, 149.1) 70+ years 143.9 (126.5, 161.4) 137.7 (123.0, 152.4) 135.0 (120.6, 149.3) 138.9 (126.5, 151.3) All ages 126.1 (122.8, 129.4) 125.9 (122.3, 129.5) 122.5 (119.8, 125.2) 126.2 (122.7, 129.7) p=0.62a Mean DBP (95% CI) 20-39 years 77.3 (74.1, 80.4) 79.4 (76.4, 82.7) 80.1 (77.3, 82.9) 76.4 (73.0, 79.8) 40-59 years 86.1 (82.6, 89.5) 84.2 (81.0, 87.3) 84.4 (81.1, 87.7) 82.9 (79.1, 86.8) 60-69 years 81.8 (77.9, 85.8) 79.8 (69.7, 89.9) 82.7 (76.2, 89.3) 85.5 (79.2, 91.7) 70+ years 86.6 (76.1, 97.1) 81.8 (72.5, 91.2) 82.8 (72.3, 93.3) 83.4 (75.9, 90.9) All ages 82.6 (80.5, 84.6) 81.9 (79.7, 84.2) 82.3 (80.3, 84.3) 81.2 (78.9, 83.5) p=0.87a Women with current/ recent pregnancys Mean SBP (95% CI) 20-39 years 107.5 (103.4, 111.6) 115.9 (110.6, 121.2) 110.1 (107.0, 113.2) 110.7 (94.0, 127.4) 40-59 years 116.7 (100.4, 133.0) 122.7 (97.5, 147.8) - - All ages 108.8 (104.8, 112.7) 117.2 (112.2, 122.0) 110.1 (107.0, 113.2) 110.7 (94.0, 127.4) p=0.57a Mean DBP (95% CI) 20-39 years 71.9 (68.4, 75.4) 73.0 (67.7, 78.3) 75.1 (72.0, 78.2) 71.5 (63.3, 79.7) 40-59 years 70.6 (57.9, 83.3) 83.3 (54.6, 112.0) - - 60-69 years - - - - 70+ years - - - - All ages 71.7 (68.5, 74.9) 74.9 (69.7, 80.2) 75.1 (72.0, 78.2) 71.5 (63.3, 79.7) p=0.29a Total survey population Mean SBP (95% CI) 20-39 years 117.8 (115.2, 120.4) 122.6 (119.9, 125.2) 119.9 (117.8, 122.1) 122.3 (119.2, 125.4) 40-59 years 132.4 (128.9, 135.9) 131.5 (128.3, 134.7) 129.8 (126.4, 133.2) 131.7 (128.2, 135.2) 60-69 years 139.9 (134.7, 145.1) 140.7 (128.9, 152.6) 131.7 (126.0, 137.4) 135.5 (129.5, 141.4) 70+ years 143.5 (135.3, 151.8) 141.2 (131.7, 150.8) 133.7 (125.0, 142.4) 140.7 (130.0, 151.4) All ages 128.7 (126.5, 130.8) 129.7 (127.4, 132.0) 125.5 (123.6, 127.3) 129.7 (127.4, 132.1) p=0.22a Mean DBP (95% CI) 20-39 years 76.5 (74.7, 78.3) 78.0 (76.0, 80.1) 80.1 (78.4, 81.8) 78.9 (76.5, 81.3) 40-59 years 85.7 (83.4, 88.1) 83.4 (81.3, 85.5) 86.1 (83.5, 88.6) 84.3 (82.0, 86.5) 60-69 years 82.4 (79.5, 85.3) 84. (77.4, 91.9) 82.8 (79.1, 86.6) 81.9 (77.6, 86.3) 70+ years 84.0 (78.7, 89.3) 85.1 (79.3, 90.9) 80.8 (75.5, 86.1) 81.2 (75.5, 87.0) All ages 81.4 (80.1, 82.8) 81.5 (80.1, 83.0) 82.4 (81.1, 83.7) 81.7 (80.2, 83.2) p=0.19a a Two-sided independent samples t-tests were used to compare mean blood pressure for all ages between PEN and non-PEN villages. 62 APPENDIX 1 PREVALENCE OF HYPERTENSION BY POPULATION AND VILLAGE There were no significant differences in HTN prevalence based on village participation in the PEN program (hypertension = SBP ≥140 or DBP ≥ 90 mmHg or receiving medication for high BP). TABLE 7. PREVALENCE OF HYPERTENSION BY POPULATION AND VILLAGE PEN villages Non-PEN villages p-value PEN Mulifanua Neiafu Fagaloa Vaisala vs. non-PEN Men aged 20+ yearss % HTN, by age (95% CI) 20-39 years 9 (16.4) (7.8, 28.8) 13 (21.7) (12.1, 34.2) 130.7 (128.0, 133.5) 132.2 (128.9, 135.5) 40-59 years 26 (48.1) (34.3, 62.2) 23 (41.1) (28.1, 55.0) 135.8 (130.0, 141.6) 136.6 (132.2, 140.9) 60-69 years 17 (65.4) (44.3, 82.8) 8 (57.1) (28.9, 82.3) 130.4 (124.6, 136.1) 130.3 (123.7, 136.8) 70+ years 9 (50.0) (26.0, 74.0) 8 (72.7) (39.0, 94.0) 132.6 (120.0, 145.2) 143.1 (120.8, 165.4) All ages 61 (39.9) (32.1, 48.1) 52 (36.9) (28.9, 45.4) 132.6 (130.0, 135.1) 134.6 (131.8, 137.4) p=0.46a % HTN, by BMI (95% CI) < 18.5 - - - - 18.5 to <25.0 4 (15.4) (4.4, 34.9) 3 (11.5) (2.4, 30.2) 8 (40.0) (19.1, 63.9) 5 (55.6) (21.2, 86.3) 25.0 to < 30.0 19 (36.5) (23.6, 51.0) 16 (32.7) (19.9, 47.5) 10 (22.7) (11.5, 37.8) 14 (43.8) (26.4, 62.3) ≥30.0 38 (50.7) (38.9, 62.4) 33 (50.0) (37.4, 62.6) 45 (50.6) (39.8, 61.3) 26 (40.0) (28.0, 52.9) % HTN, by DM (95% CI) With diabetes 7 (100.0) (59.0, 100.0) 0 (0.0) (0.0, 97.5) 3 (30.0) (6.7, 65.2) 2 (50.0) (6.8, 93.2) diagnosis No diabetes 53 (36.8) (28.9, 45.2) 52 (37.4) (29.4, 46.0) 59 (41.3) (33.1, 49.8) 43 (42.2) (32.4, 52.3) diagnosis Non-pregnant women aged 20+ years Mean DBP (95% CI) 20-39 years 6 (12.5) (4.7, 25.2) 8 (19.0) (8.6, 34.1) 10 (15.2) (7.5, 26.1) 5 (13.2) (4.4, 28.1) 40-59 years 26 (38.8) (27.1, 51.5) 25 (41.0) (28.6, 54.3) 26 (41.9) (29.5, 55.2) 15 (35.7) (21.6, 52.0) 60-69 years 14 (53.8) (33.4, 73.4) 8 (53.3) (26.6, 78.7) 13 (56.5) (34.5, 76.8) 12 (60.0) (36.1, 80.9) 70+ years 7 (77.8) (40.0, 97.2) 9 (69.2) (38.6, 90.9) 6 (54.5) (23.4, 83.3) 11 (91.7) (61.5, 99.8) All ages 53 (35.3) (27.7, 43.5) 50 (38.2) (29.8, 47.1) 55 (34.0) (26.7, 41.8) 43 (38.4) (29.4, 48.1) p=0.86a % HTN, by BMI (95% CI) < 18.5 0 (0.0) (0.0, 97.5) 0 (0.0) (0.0, 97.5) - - 18.5 to <25.0 7 (50.0) (23.0, 77.0) 2 (28.6) (3.7, 71.0) 0 (0.0) (0.0, 36.9) 2 (25.0) (3.2, 65.1) 25.0 to < 30.0 5 (27.8) (9.7, 53.5) 7 (31.8) (13.9, 54.9) 7 (25.9) (11.1, 46.3) 8 (36.4) (17.2, 59.3) ≥30.0 40 (35.4) (26.6, 45.0) 41 (41.0) (31.3, 51.3) 48 (38.4) (29.8, 47.5) 33 (40.2) (29.6, 51.7) % HTN, by DM (95% CI) With diabetes 11 (68.8) (41.3, 89.0) 8 (80.0) (44.4, 97.5) 13 (59.1) (36.4, 79.3) 5 (62.5) (24.5, 91.5) diagnosis No diabetes 42 (31.3) (23.6, 39.9) 42 (34.7) (26.3, 43.9) 41 (29.5) (22.1, 37.8) 37 (35.9) (26.7, 46.0) diagnosis 63 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA TABLE 7. PREVALENCE OF HYPERTENSION BY POPULATION AND VILLAGE cont/… PEN villages Non-PEN villages p-value PEN Mulifanua Neiafu Fagaloa Vaisala vs. non-PEN Women with current/recent pregnancy % HTN, by age (95% CI) 20-39 years 1 (3.2) (0.1, 16.7) 1 (7.7) (0.2, 36.0) 2 (5.0) (0.6, 16.9) 0 (0.0) (0.0, 52.2) 40-59 years 0 (0.0) (0.0, 52.2) 1 (33.3) (0.8, 90.6) - 0 (0.0) (0.0, 97.5) All ages 1 (2.8) (0.1, 14.5) 2 (12.5) (1.6, 38.3) 2 (5.0) (0.6, 16.9) 0 (0.0) (0.0, 45.9) p=0.75a % HTN, by BMI (95% CI) < 18.5 - - - - 18.5 to <25.0 0 (0.0) (0.0, 52.2) 0 (0.0) (0.0, 84.2) 0 (0.0) (0.0, 84.2) - 25.0 to < 30.0 0 (0.0) (0.0, 41.0) 0 (0.0) (0.0, 84.2) 0 (0.0) (0.0, 33.6) 0 (0.0) (0.0, 97.5) ≥30.0 1 (4.2) (0.1, 21.1) 2 (16.7) (2.1, 48.4) 2 (6.9) (0.8, 22.8) 0 (0.0) (0.0, 52.2) % HTN, by DM (95% CI) With diabetes 0 (0.0) (0.0, 84.2) - 0 (0.0) (0.0, 97.5) - diagnosis No diabetes 1 (2.9) (0.1, 15.3) 2 (12.5) (1.6, 38.3) 2 (5.3) (0.6, 17.7) 0 (0.0) (0.0, 45.9) diagnosis Total survey population Mean DBP (95% CI) 20-39 years 16 (11.9) (7.0, 18.7) 22 (19.1) (12.4, 27.5) 37 (21.3) (15.4, 28.1) 17 (21.0) (12.7, 31.5) 40-59 years 52 (41.3) (32.6, 50.4) 49 (40.8) (32.0, 50.2) 49 (43.0) (33.7, 52.6) 37 (43.0) (32.4, 54.2) 60-69 years 31 (59.6) (45.1, 73.0) 16 (55.2) (35.7, 73.6) 22 (50.0) (34.6, 65.4) 17 (47.2) (30.4, 64.5) 70+ years 16 (59.3) (38.8, 77.6) 17 (70.8) (48.9, 87.4) 12 (50.0) (29.1, 70.9) 17 (81.0) (58.1, 94.6) All ages 115 (33.9) (28.9, 39.2) 104 (36.1) (30.6, 42.0) 120 (33.7) (28.8, 38.9) 88 (39.3) (32.8, 46.0) p=0.74a % HTN, by BMI (95% CI) < 18.5 0 (0.0) (0.0, 97.5) 0 (0.0) (0.0, 97.5) - - 18.5 to <25.0 11 (24.4) (12.9, 39.5) 5 (14.3) (4.8, 30.3) 8 (26.7) (12.3, 45.9) 7 (41.2) (18.4, 67.1) 25.0 to < 30.0 24 (31.2) (21.1, 42.7) 23 (31.5) (21.1, 43.4) 17 (21.2) (12.9, 31.8) 22 (40.0) (27.0, 54.1) ≥30.0 79 (37.3) (30.7, 44.2) 76 (42.7) (35.3, 50.3) 95 (39.1) (32.9, 45.5) 59 (38.8) (31.0, 47.0) % HTN, by DM (95% CI) With diabetes 18 (72.0) (50.6, 87.9) 8 (72.7) (39.0, 94.0) 16 (48.5) (30.8, 66.5) 7 (58.3) (27.7, 84.8) diagnosis No diabetes 96 (30.8) (25.7, 36.2) 96 (34.8) (29.2, 40.7) 102 (31.9) (26.8, 37.3) 80 (37.9) (31.3, 44.8) diagnosis Chi Squared tests were used to examine differences in hypertension between PEN and non-PEN villages. Significance testing was a undertaken for the ‘all ages’ category only. 64 APPENDIX 1 HISTORY OF BP MEASUREMENT TABLE 8. HISTORY OF BLOOD PRESSURE MEASUREMENT BY POPULATION AND VILLAGE PEN villages Non-PEN villages p-value PEN Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) vs. non-PEN Men aged 20+ yearss Ever measured BP [Q407] Yes 103 (67.3) 92 (65.2) 71 (46.1) 75 (70.8) p=0.02a No 50 (32.7) 49 (34.8) 83 (53.9) 31 (29.2) Don’t know 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) BP measurement in last 12 months [Q408] b 3 times or more 28 (27.2) 37 (40.2) 13 (18.3) 23 (21.7) p=0.03a 1-2 times 46 (44.7) 35 (38.0) 44 (62.0) 36 (34.0) Never 29 (28.2) 20 (21.7) 12 (16.9) 16 (15.1) Don’t know 0 (0.0) 0 (0.0) 2 (2.8) 0 (0.0) Place where last BP was measured [Q409] b My closest/ 57 (55.3) 79 (85.9) 19 (26.8) 57 (76.0) p=0.07 Government facility Another Government 12 (11.7) 7 (7.6) 16 (22.5) 7 (9.3) facility (not closest) Pharmacy 0 (0.0) 0 (0.0) 1 (1.4) 0 (0.0) Private doctor/clinic 7 (6.8) 3 (3.3) 12 (16.9) 0 (0.0) Village Women Com- 17 (16.5) 1 (1.1) 13 (18.3) 6 (8.0) mittee Local NGO/FBO 0 (0.0) 0 (0.0) 1 (1.4) 0 (0.0) Traditional healer 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Campaign 8 (7.8) 1 (1.1) 5 (7.0) 5 (4.7) (screening) day Measuring myself 1 (1.0) 0 (0.0) 0 (0.0) 0 (0.0) (self-test) Other 1 (1.0) 1 (1.1) 4 (5.6) 0 (0.0) Non-pregnant women aged 20+ years Ever measured BP [Q407] Yes 97 (64.7) 106 (80.9) 97 (59.9) 86 (76.8) p=0.16a No 53 (35.3) 25 (19.1) 65 (40.1) 26 (23.2) Don’t know 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) BP measurement in last 12 months [Q408] 3 times or more 30 (30.9) 51 (48.1) 28 (29.2) 32 (37.6) p=0.20a 1-2 times 46 (47.4) 39 (36.8) 50 (52.1) 39 (45.9) Never 21 (21.6) 16 (15.1) 16 (16.7) 14 (16.5) Don’t know 0 (0.0) 0 (0.0) 2 (2.1) 0 (0.0) 65 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA TABLE 8. HISTORY OF BLOOD PRESSURE MEASUREMENT BY POPULATION AND VILLAGE cont/… PEN villages Non-PEN villages p-value PEN Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) vs. non-PEN Non-pregnant women aged 20+ years cont/… Place where last BP was measuredb [Q409] My closest/ 49 (50.5) 87 (82.9) 34 (35.1) 74 (86.0) p=0.26a Government facility Another Government 13 (13.4) 5 (4.8) 22 (22.7) 2 (2.3) facility (not closest) Pharmacy 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Private doctor/clinic 9 (9.3) 2 (1.9) 16 (16.5) 2 (2.3) Village Women Com- 20 (20.6) 5 (4.8) 18 (18.6) 4 (4.7) mittee Local NGO/FBO 1 (1.0) 0 (0.0) 1 (1.0) 0 (0.0) Traditional healer 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Campaign 4 (4.1) 5 (4.8) 2 (2.1) 4 (4.7) (screening) day Measuring myself 1 (1.0) 1 (1.0) 1 (1.0) 0 (0.0) (self-test) Other 0 (0.0) 0 (0.0) 3 (1.9) 0 (0.0) Women with current/recent pregnancy Ever measured BP [Q407] Yes 26 (72.2) 15 (93.8) 28 (70.0) 5 (83.3) p=0.41a No 10 (27.8) 1 (6.3) 12 (30.0) 1 (16.7) Don’t know 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) BP measurement in last 12 months [Q408] 3 times or more 12 (46.2) 8 (53.3) 9 (32.1) 3 (60.0) p<0.01a 1-2 times 4 (15.4) 4 (26.7) 15 (53.6) 2 (40.0) Never 9 (34.6) 3 (20.0) 2 (7.1) 0 (0.0) Don’t know 1 (3.8) 0 (0.0) 2 (7.1) 0 (0.0) Place where last BP was measured [Q409] My closest/ 19 (73.1) 11 (73.3) 11 (39.3) 5 (100.0) p=0.20a Government facility Another Government 3 (11.5) 3 (20.0) 10 (35.7) 0 (0.0) facility (not closest) Pharmacy 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Private doctor/clinic 1 (3.8) 1 (6.7) 1 (3.6) 0 (0.0) Village Women 1 (3.8) 0 (0.0) 4 (14.3) 0 (0.0) Committee Local NGO/FBO 1 (3.8) 0 (0.0) 0 (0.0) 0 (0.0) Traditional healer 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Campaign 1 (3.8) 0 (0.0) 1 (3.6) 0 (0.0) (screening) day Measuring myself 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) (self-test) Other 0 (0.0) 0 (0.0) 1 (3.6) 0 (0.0) 66 APPENDIX 1 TABLE 8. HISTORY OF BLOOD PRESSURE MEASUREMENT BY POPULATION AND VILLAGE cont/… PEN villages Non-PEN villages p-value PEN Mulifanua N (%) Neiafu N (%) Fagaloa N (%) Vaisala N (%) vs. non-PEN Total survey population Ever measured BP [Q407] Yes 226 (66.7) 213 (74.0) 196 (55.1) 166 (74.1) p=0.01a No 113 (33.3) 75 (26.0) 160 (44.9) 58 (25.9) Don’t know 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) BP measurement in last 12 months [Q408] 3 times or more 70 (31.0) 96 (45.1) 50 (25.6) 58 (35.2) p<0.001a 1-2 times 96 (42.5) 78 (36.6) 109 (55.9) 77 (46.7) Never 59 (26.1) 39 (18.3) 30 (15.4) 30 (18.2) Don’t know 1 (0.4) 0 (0.0) 6 (3.1) 0 (0.0) Place where last BP was measured [Q409] My closest/ 125 (55.3) 177 (83.5) 64 (32.7) 136 (81.9) p<0.01a Government facility Another Government 28 (12.4) 15 (7.1) 48 (24.5) 9 (5.4) facility (not closest) Pharmacy 0 (0.0) 0 (0.0) 1 (0.5) 0 (0.0) Private doctor/clinic 17 (7.5) 6 (2.8) 29 (14.8) 2 (1.2) Village Women 38 (16.8) 6 (2.8) 35 (17.9) 10 (6.0) Committee Local NGO/FBO 2 (0.9) 0 (0.0) 2 (1.0) 0 (0.0) Traditional healer 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) Campaign 13 (5.8) 6 (2.8) 8 (4.1) 9 (5.4) (screening) day Measuring myself 2 (0.9) 1 (0.5) 1 (0.5) 0 (0.0) (self-test) Other 1 (0.4) 1 (0.5) 8 (4.1) 0 (0.0) a Chi Squared tests were used to examine differences in hypertension between PEN and non-PEN villages 67 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA Appendix 2. Patient File Review ELIGIBILITY CRITERIA Being on the Hypertension Register in one of the three health facilities included in the patient file review; a)  Patients not on hypertension drug therapy (non-pharma treatment) were therefore included b) Aged 20 years old or above c) Having been diagnosed with hypertension in 2017 or any year before. DATA VARIABLES COLLECTED File number Gender Year of birth Hypertension diagnosis: Date of diagnosis Hypertension diagnosis: Blood pressure at diagnosis (systolic) Hypertension diagnosis: Blood pressure at diagnosis (diastolic) Hypertension diagnosis: Body weight at diagnosis (kg) Hypertension treatment: Date of treatment start Current hypertension drugs: Thiazide-diuretic Current hypertension drugs: Calcium channel blocker Current hypertension drugs: ACE inhibitor Current hypertension drugs: Angiotensin receptor blocker Current hypertension drugs: Aspirin Current hypertension drugs: Statins and other lipid-lowering drugs Current hypertension drugs: Beta-blocker Diabetes Diagnosis: Type (I or II) Diabetes Treament: Only diet and exercise advice Diabetes Treament: Oral sugar-lowering drugs Diabetes Treament: Insulin only Diabetes Treament: Combination treatment Height Body weight (most recent) Blood pressure (penultimate): Date Blood pressure (penultimate): Result systolic Blood pressure (penultimate): Result diastolic Blood pressure (last): Date Blood pressure (last): Result systolic Blood pressure (last): Result diastolic Last laboratory (Sugar): Date Last laboratory (Sugar): Result Last laboratory (HbA1C): Date Last laboratory (HbA1C): Result Last laboratory (Cholesterol): Date Last laboratory (Cholesterol): Result Hospital admission (last): Date Hospital diagnosis: Comments 68 APPENDIX 3 Appendix 3. Focus Group Discussions Three FGDs were conducted in February 2019: (1) women with hypertension (Mulifanua; patients), (2) men with hypertension (Mulifanua; patients), (3) health facility staff (providers). PARTICIPANT SELECTION Participants for the patient FGDs were selected using data from the 2018 hypertension household survey. Mulifanua was selected as the site for the FGD based on it’s participation in the MOH-lead PEN program. FGDs were only conducted at this site due to weather restrictions (lack of ferries/planes between Upolu and Savai’i after a cyclone). For both men and women, a list was generated of participants who had reported a diagnosis of hypertension in the household survey. These individuals were listed to capture experiences of the health system. This list of participants was divided among those who had (1) chosen not the visit the health center for confirmation of the hypertension diagnosis; (2) visited the health center for diagnosis, but were not taking medication; and (3) initiated medication for hypertension. Equal numbers of participants were selected from each of these lists and invited to attend. Study staff approached potential participants by phone or in person (at their homes), explained the purpose of the FGD, and invited them to attend. Nine participants were invited for each FGD and all participants invited completed the FGD. MOH staff, under the supervision of Dr. Thompson, contacted potential participants for the health care provider FGD. A request was made by LPAMF staff to have nurses, clinicians, and MOH administrators of the PEN program present. The provider FGD comprised seven district health facility nursing staff – five female, two male and had representation from both Upolu and Savai’i facilities, including the two health facilities closest to Mulifanua. An attempt was made to organize a second FGD to engage clinicians, but it could not be implemented. CONDUCT OF THE FOCUS GROUP DISCUSSIONS The two patient FGDs took place at a community gathering space in Mulifanua, while the provider FGD was held in a conference room at the Ministry of Health. All three FGDs were facilitated by LPAMF staff who had received prior training in FGD facilitation; a lead facilitator was responsible for completing the FGD agenda, co-facilitator was responsible for ensuring adequate representation of each participant in the group, and two note takers captured non-verbal cues and recorded notes on the conversation. For the women’s FGD and the provider FGD, female staff served as facilitator and co-facilitator. As is customary in Samoa, a male facilitator lead the men’s FGD with support from a female co-facilitator. The FGD agendas were developed based on prompts provided by the Samoa-based World Bank team (with approval from the Ministry of Health Health Research Committee). Minor additions were made to the agenda to expand prompts under questions about PEN project. The two patient FGDs were conducted primarily in Samoan, with some English language used (as is usual in Samoa, particularly when discussing medical care/terminology). The provider FGD was conducted largely in English, with only some participants choosing to speak Samoan. All of the discussion was audio recorded with the participant’s informed consent and photographs of the outcomes of the brainstorming activities were captured. 69 CARE FOR HYPERTENSION AND OTHER CHRONIC CONDITIONS IN SAMOA TRANSCRIPTION, TRANSLATION AND REPORTING Using the audio files recorded during the FGDs, LPAMF staff transcribed the files verbatim. Transcripts were reviewed by a second member of staff for accuracy. English translations of Samoan language discussions were provided by the LPAMF senior research assistant and director who reviewed translations for accuracy. Dr. Hawley was responsible for preparing the summaries of each of the FGDs using a thematic analysis approach. Patient FGDs were combined for reporting due to agreement across the two groups in terms of major themes. The provider FGD was written up separately, but organized using similar headings so that responses could be contrasted with those of patients. Context: Shortly after the completion of the survey activities in Mulifanua in December, two participants in the study (who were well known to the community; the wife of one of the male FGD participants) passed away suddenly. The study team was told that the cause of death was hypertension-related. As a result, the women in this community were particularly attuned to the issue of hypertension and expressed anxiety about hypertension in an acute way that may not be generalizable outside of this village/this moment in time. 70 World Bank World Bank Health, Nutrition World Bank Samoa 1818 H Street, NW and Population Global Practice Level 7, Central Bank Building, Washington DC 20433 https://www.worldbank.org/en/topic/health Beach Rd, Apia USA Postal Address: PO Box 3999 https://www.worldbank.org/en/region/eap Tel : +685 24492 / 34340 Fax: +685 2422