© 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 licence 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: Hypertension and Type-2 Diabetes in Bangladesh: Continuum of Care Assessment and Opportunities for Action. Washington DC: World Bank 2019. 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. Cover Image: Rights Usage Permission ©Dreamstime.com/Derivative by Theo Hawkins HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION JUNE 2019 This page is for collation purposes TABLE OF CONTENTS 1.1 Aim of this report ........................................................................................................................................... 1 1.2 Global context.................................................................................................................................................. 2 1.3 Bangladesh health transition .................................................................................................................... 3 2.1 Hypertension cascade, breakpoints and outcomes .......................................................................... 5 2.2 Type-2 diabetes cascade, breakpoints and outcomes ..................................................................... 7 3.1 Policies, strategies and plans on NCD control.................................................................................. 11 3.2 Financing and resources for NCD control ......................................................................................... 12 3.3 Health infrastructure relevant to NCD control ................................................................................ 14 3.4 NCD interventions and costs .................................................................................................................. 16 4.1 Overview of selected care models implemented in Bangladesh .............................................. 21 4.2 Principles and examples of NCD and chronic care models implemented elsewhere....... 23 5.1 NCD policy and resource allocation ..................................................................................................... 39 5.2 NCD burden and risk factors .................................................................................................................. 39 5.3 Major NCD cascade breakpoints ........................................................................................................... 40 5.4 Integrated chronic care models............................................................................................................. 41 5.5 Surveillance, monitoring and evaluation ........................................................................................... 42 CITATION AND NOTES................................................................................................................................................ 63 ANNEXES v FIGURES ES 1 Schematic of continuum of care for chronic conditions ...................................................................xiii ES 2 Models of care as per NCD Control Operational Plan 2017‒22 ..................................................... xiv ES 4 Hypertension cascade Bangladesh ........................................................................................................... xvi ES 5 Diabetes cascade Bangladesh.................................................................................................................... xvii 1.1 NCDs as cause of all deaths by country group, % (2000, 2015) ....................................................... 2 2.1 Hypertension cascade for rural and urban residents aged 35 years and above ........................ 6 2.2 Diabetes cascade for rural and urban residents aged 35 years and above .................................. 8 3.1 Division of funding amongst the health service programmes in 2017‒22 ............................... 13 3.2 Reported medication use by diabetes patients (in percentages, 2014 data) ........................... 18 4.1 Percentage of health facilities providing diabetes and CVD services and respective medications in rural and urban Bangladesh ......................................................................................... 22 A 1.1 Disability-Adjusted Life Years by cause, Bangladesh, 2016 ............................................................ 45 A 1.2 Distribution of population reporting chronic illness in preceding 12 months by type of illness, HIES2016 .............................................................................................................................. 46 A 1.3 Prevalence of reported diabetes by sub-population, 2016 HIES .................................................. 46 A 1.4 Death attributable to risk factors, Bangladesh, 2016 ........................................................................ 47 A 3.1 Estimated total burden of hypertension in adults aged 18 years and above in Bangladesh Divisions ................................................................................................................................ 50 A 3.2 Estimated total burden of diabetes in adults aged 18 years and above in Bangladesh Divisions .............................................................................................................................................................. 50 A 4.1 Reported hospital admissions for hypertension by Bangladesh Division ................................. 51 A 4.2 Reported hospital admissions for diabetes by Bangladesh Division ........................................... 52 A 6.2 Diagnosis rates of diabetes across countries ........................................................................................ 55 A 7.1 Bangladesh NCD policy context compared to countries with similar GNI per capita ........... 56 A 7.2 Bangladesh NCD score card results in international comparison ................................................ 56 A 8.1 Bangladesh total health spending, 2014................................................................................................. 57 A 8.2 Allocations and level of expenditure of HPNSDP in FY 2015‒16.................................................. 57 A 8.3 Share of government funding across health implementation plans, 2017‒22........................ 58 A 9.1 Unit cost distribution for hypertension management, diabetes mellitus, stroke and acute ischemic heart disease .............................................................................................................. 59 A 9.2 Estimated cost of interventions to reduce NCD risk factors in three countries ...................... 59 A 11.1 Infographic: New model of NCD care in A) Urban and B) Rural Bangladesh ........................... 61 vi TABLES 3.1 Public sector health facilities visiting numbers per year ................................................................ 15 4.1 Highlighted hypertension and diabetes care models in Bangladesh across the care cascade .............................................................................................................................................. 24 4.2 Bangladesh’s new model for the continuum of care for hypertension and type-2 diabetes................................................................................................................................................ 29 A 2.1 Benchmarking Bangladesh with countries with similar GNI per capita ................................... 48 A 2.2 Benchmarking Bangladesh with countries in the South Asia Region ........................................ 48 A 2.3 Benchmarking Bangladesh with countries with similar GNI per capita ................................... 49 A 2.4 Benchmarking Bangladesh with countries in the South Asia Region ........................................ 49 A 5.1 Hypertension care: Overview of patient and provider barriers to the continuum of care .......................................................................................................................................... 53 A 5.2 Diabetes care: Overview of patient and provider barriers to the continuum of care .......................................................................................................................................... 54 A 8.1 Facility distribution in rural areas of Bangladesh ............................................................................. 58 A 10.1 Drug list for Diabetes Mellitus ................................................................................................................... 60 A 10.2 List of medicines for Hypertension ......................................................................................................... 60 vii This page is for collation purposes. ACRONYMS BADAS Bangladesh Diabetic Somiti (Diabetic Association of Bangladesh) BDHS Bangladesh Demographic and Health Survey BDT Bangladeshi Taka BIRDEM Bangladesh Institute of Research & Rehabilitation in Diabetes, Endocrine and Metabolic Disorders BP Blood pressure CBHC Community Based Health Care CC Community clinic CHW Community health worker CKD Chronic kidney disease COBRA-BPS Control of Blood Pressure and Risk Attenuation-rural Bangladesh, Pakistan, Sri Lanka COPD Chronic obstructive pulmonary disease CVD Cardiovascular disease DALY Disability-adjusted life years DGHS Directorate General of Health Services DHIS2 District Health Information System version 2 ESP Essential services package FBG Fasting blood glucose GBD Global burden of disease GoB Government of Bangladesh HIES Household Income and Expenditure Survey HPNSDP Health, Population and Nutrition Sector Development Plan HPNSP Health, Population and Nutrition Sector Programme HRH Human resources for health HTN Hypertension icddr,b International Centre for Diarrhoeal Disease Research, Bangladesh LIC Low income country LMIC Lower middle income country MCH Mother & child health mHealth Mobile health MNCAH Maternal, Neonatal, Child and Adolescent Health MO Medical officer MOHFW Ministry of Health and Family Welfare NCD Non-communicable disease NCDC Non-communicable disease control NHF National Heart Foundation NIPSOM National Institute of Preventive and Social Medicine OGTT Oral Glucose Tolerance Test OOP Out-of-pocket PEN Package of Essential Non-Communicable Disease PHC Primary Health Care ix PPP Purchasing Power Parity RADP Revised annual development programme RBG Random Blood Glucose SACMO Sub-assistant Community Medical Officer SLT Smokeless tobacco T2DM Type-2 diabetes mellitus UHFWC Union Health and Family Welfare Centres UZHC Upazila Health Complex WHO World Health Organisation x ACKNOWLEDGEMENTS This report was prepared under the overall guidance of Prof .David Wilson ( Program Director and Global Lead, Decision & Delivery Science, World Bank) by Bushra Binte Alam and Anna Koziel (Senior Health Specialists, World Bank), Nicole Fraser-Hurt, Zara Shubber, Adetayo Olabiran and Jil Billy Mamza(Consultants, World Bank). The team is deeply appreciative of the guidance received from. Prof Abul Kalam Azad, DG, Director General of Health Services(DGHS); Dr. Mohiuddin Osmani, Joint Chief, Ministry of Health and Family Welfare, Government of Bangladesh. The study team is grateful for the technical input of key stakeholders and resource persons. This includes the following experts and teams: Prof. Enayet Hussain, ADG, DGHS and Line Director, NCD OP; Prof. A K Azad Khan, President, and his management team at the Bangladesh Diabetic Association; National Prof. Brig. (Retd.)M.A. Malek, Founder and President, and his team at the National Heart Foundation, Prof. Liaquat Ali, Vice Chancellor, Bangladesh University of Health Sciences associated with Bangladesh Diabetic Association, Dr. Mostafa Zaman, Dr. Tara Kessaram and NCD team, WHO, Prof. M. Abul Faiz, former-DGHS , Mr. Muhammad Abdul Hannan Khan, Team Leader HISP Bangladesh Foundation, Dr. Aliya Naheed and NCD team, ICDDR,B, and Ms. Rie Ozaki, JICA . Special thanks go to Mr Shyfuddin Ahmed, Scientist at the ICDDR,B, for accompanying the study team to Munshiganj District for a field visit, and to the health facilities visited on this occasion. Dr. Mohibur Nirob, Medical Officer, DGHS, provided insights from practice and document translation. The team appreciates the contribution of Dr. Muhammod Abdus Sabur and Dr. Asib Nasim (Consultants ,WB) who supported in collection of data and visits to different sites. The team also acknowledges the valuable information provided by Tekabe Ayalew Belay (Program Leader, World Bank). The authors would also like to thank the World Bank peer reviewers Mickey Chopra (Lead Health Specialist & Global Solutions Lead) and Marelize Görgens (Senior Monitoring & Evaluation Specialist) for their comments. xi “Self-management of chronic diseases can be challenging for patients, as it likely needs daily and lifelong changes to lifestyle, and therefore requires tailored education and additional and ongoing support.” EXECUTIVE SUMMARY Bangladesh needs to transition from a disease-specific public health care system designed for infectious diseases and mother & child health (MCH) to an integrated primary health care (PHC) system with strong out-patient care and community health promotion components. The non-communicable disease (NCD) burden is very large and new solutions are needed to reduce NCD risk factors, prevent or slow the progression to disease, and manage complications sustainably. Bangladesh’s NCD Control (NCDC) Operational Plan 2017‒22 provides a framework to instigate change to better cater for NCD prevention, care and support. This analysis was carried out to provide input into the dialogue on how to improve NCD integrated care and outcomes in Bangladesh. The objectives were to a) Assess the hypertension (HTN) and type-2 diabetes burden in Bangladesh, including benchmarking, b) Review Bangladesh’s policy and health care delivery environment for NCDs, c) Determine the continuum of care for hypertension and diabetes, d) Assess models of NCD and chronic care in Bangladesh as well as other countries to identify common principles and promising practices, and e) Identify opportunities for strengthening primary level care for HTN and diabetes. The methodology consisted of literature review, consultations with experts, data triangulation, benchmarking against comparable economies, and synthesis of the evidence. For the continuum of care, the cascade framework was used to assess the stages from screening and diagnosis to treatment initiation, adherence and disease control (Fig. ES 1). Figure ES 1 Schematic of continuum of care for chronic conditions Source: Author’s illustration based on the concept of the chronic care cascade. xiii HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION NCD BURDEN IN BANGLADESH AND BENCHMARKING In 2016, NCDs were The NCD burden in Bangladesh is very large. In 2016, NCDs were responsible for almost responsible for almost two-thirds (63%) of Disability-adjusted life years (DALYs) in Bangladesh, while communicable, maternal, neonatal two-thirds (63%) of and nutritional diseases accounted for 27% and injuries for 11% of Disability-adjusted life DALYs. Ischemic heart disease (IHD), stroke and type 2 diabetes are years (DALYs) in the main contributors to the NCD burden. In the 2016 Household Bangladesh, while Income and Expenditure Survey (HIES), HTN, chronic heart disease communicable, maternal, and diabetes were among the most reported illnesses with rising levels neonatal and nutritional of reported diabetes in females between 2010 and 2016. diseases accounted for 27% and injuries for 11% Bangladesh has achieved great success in reducing all-cause mortality of DALYs. and pre-mature death and improving life expectancy. However, benchmarking across countries with similar GNI per capita income levels revealed poor rankings for NCD-attributable premature death for Bangladesh (rank 11 of 12 for all NCD mortality). Poor rankings were also observed for premature death due to cerebrovascular disease, chronic kidney disease and chronic obstructive pulmonary disease (all rank 12), diabetes (rank 11) and ischaemic heart disease (rank 9). Furthermore, the 12-country benchmarking analysis demonstrates the high share of all-cause death and disability attributable to major NCD risk factors such as high systolic blood pressure, high fasting plasma glucose and tobacco smoking. NCD POLICY, FINANCING AND SERVICE DELIVERY Bangladesh scores well internationally on NCD policy.1 The government of Bangladesh has initiated many NCD-related policies or programmes. For example, the country has defined a NCD-specific Operational Plan and NCD clinical guidelines and has increased excise taxes and prices for tobacco demand reduction. However, NCD interventions “lack proper planning, implementation and monitoring”2 according to a recent NCD policy review. In order to address the challenges and bottlenecks in public sector NCD service delivery, the Operational Plan proposes the implementation of NCD models of care (Fig. ES 2, see also Fig. A ll.1). Figure ES 2 Models of care as per NCD Control Operational Plan 2017‒22 (left: urban model, right: rural model) Source: Author’s drawings based on NCDC Operational Plan 2017‒22. xiv EXECUTIVE SUMMARY While the models of care are promising, the public sector NCD The annual point-of- budget is relatively small (<5% of public health sector funding) service payments that compared to the NCD burden (63% of DALYs). The resulting per an individual with capita NCD allocation of US$0.08 is also very low compared to the diabetes makes each estimated cost of implementing WHO’s recommended set of “best buy” interventions in Low and Middle Income Countries (US$ 1.50 per year represents 27% of person and year). The Bangladesh Copenhagen Project, in its the annual assessment of the benefit of reducing NCD burden through targeted consumption investment, found HTN medication had an exceptionally high level expenditure average in of return on investment (BDT 37 benefit for every BDT spent).3 Bangladesh and Diabetes treatment also showed positive returns, albeit at a much illustrates the lower ratio, at BDT 3 benefit per BDT spent.4 potential of NCDs impoverishing Due to insufficient public sector delivery, out-of-pocket expenditure patients and their for NCD care is large in Bangladesh. Chronic care patients face recurrent expenditures to manage their disease, and disease families. complications can lead to expensive hospital stays. Annual payments for anti-diabetes medication are on average BDT 35,385 (US$ 429), an expense 22 times higher than average medication costs for non-diabetic patients.5 A diabetic’s annual point-of-service payments averages at BDT 49,538 (US$ 600). This is 27% of the 2016 HIES reported annual consumption expenditure average of BDT 185,000 and illustrates the potential of NCDs impoverishing patients and their families. Regarding service delivery for NCDs, key challenges include the health system’s orientation towards MCH and communicable diseases at PHC level at the expense of NCDs, poor NCD screening, the lack of availability of a tier-specific essential services package to patients, the weak patient referral system, and the inadequate coordination of NCD control across health and non- health sectors and between state and non-state actors. These and other factors lead to a sub- optimal continuum of NCD care. CONTINUUM OF CARE CASCADE FOR HYPERTENSION AND DIABETES Hypertension Cascade A published meta-analysis assessing the prevalence of hypertension across diverse age groups and populations in Bangladesh estimated overall HTN prevalence at 14%.6 There is a significantly higher HTN burden in urban settings and among females. Using data from Bangladesh’s Demographic and Health Survey (DHS), this analysis estimated that in 2016, about 4.1 million adults aged 35+ years had stage 27 HTN regardless of treatment status, with rural females most affected. The HTN continuum of care for adults aged 35+ years in Bangladesh highlights significant breakpoints (Fig. ES 3). xv HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION Figure ES 3 Hypertension cascade Bangladesh Sources: BDHS 2011 and 2016 UN population estimates. The 2013 National HTN Guidelines8 focus on the PHC level and provide much relevant guidance. However, the guidelines do not provide sufficient information on HTN screening, such as screening strategies and targeting of specific populations. Furthermore, despite HTN treatment adherence being a major challenge, insufficient guidance is provided on treatment adherence support. The analysis estimated that identifying 50% (or 7.1 M) of undiagnosed hypertensive people aged 35+ years by 2022 (a NCD Plan target) would cost about BDT 1140 M (US$ 13.8 M). Treating the currently diagnosed 7.3 M hypertensives aged 35+ translates into a cost of approximately BDT 7800 M (US$ 95 M). Diabetes cascade Bangladesh is among the top 10 countries with the highest number of people living with diabetes, and has a relatively low diagnosis rate.9 A meta-analysis including studies conducted across a diversity of age groups and populations estimated the prevalence Bangladesh is among of diabetes at 6% in Bangladesh.10 A study conducted at a tertiary diabetes the top 10 countries hospital in Dhaka estimated that only 20% of DALYs forgone due to diabetes with the highest were due to diabetes episodes and 80% were due to diabetes complications. number of people The diabetes continuum of care for adults aged 35+ years in Bangladesh living with diabetes. also highlights significant breakpoints along the cascade (Fig. ES 4). The 2013 National diabetes Guidelines for the PHC level11have deficiencies regarding these breakpoints in the continuum of care. The guidelines don’t provide much directive on diabetes screening and risk factor scoring. Information on the choice of diagnostic tests is scarce (HbA1C testing is included for gestational and type-1 diabetes but not for type-2 diabetes). The guidelines are also deficient in respect to glucose monitoring during treatment, drug refill at PHCs, and patient self-monitoring. xvi EXECUTIVE SUMMARY Figure ES 4 Diabetes cascade Bangladesh Sources: BDHS 2011 and 2016 UN population estimates. This analysis estimated that finding 50% of the currently undiagnosed 3.8 M diabetic adults aged 35+ years would cost approximately BDT 780 M (US$ 9.5 M). Treating the currently diagnosed 2.33 M diabetic adults aged 35+ years would cost approximately BDT 9860 M (US$ 119 M). MODELS FOR CHRONIC CARE The Government of Bangladesh has defined models for NCD care for urban and rural settings (Fig. ES 2, Fig. A 11.1). This review also scanned the literature on models of chronic care with the aim to identify international good practices and principles of such models of care. PHC-level care models often share some of the following five features: 1. Simplification, standardisation and monitoring of healthcare delivery processes 2. Decentralisation 3. Task-shifting 4. Peer and community involvement in programming 5. Support for self-management Simplifying healthcare management aims to strike a balance between While there are what is desirable and what is currently feasible in resource-limited settings. currently no It allows a wider range of providers to be involved in patient management national standard while minimising risks to patients. It can also help patients directly with operating self-management and treatment adherence (for example if the treatment procedures for the regimen is simplified). Standardising health care management helps management of maintain patient safety and high-quality healthcare service provision. NCD in Bangladesh, Indeed, a key objective of WHO’s Package of Essential NCD (PEN) the country is in the Interventions for Primary Health Care is “standardisation of diagnostic and investigation procedures and drug prescription” as well as “formulation of process of referral criteria for further assessment or hospitalisation”. While there are developing these currently no national standard operating procedures for the with the support of management of NCD in Bangladesh, the country is in the process of WHO. developing these with the support of WHO. These will be based on simplified algorithms and the use of standard drug prescriptions based on generic formulations. The use of standardised indicators, electronic records and a unique client xvii HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION identifier number will help with referrals, linking client data in all facilities, easier tracking, better understanding of defaulting and better data-driven management processes. Bangladesh has strongly advocated for a decentralised health system, however, there are several barriers to full decentralisation including administrative issues, institutional mandates and decision-making roles. Practical applications of decentralisation could include for instance drug dispensing via adherence clubs with the aim to decongest clinics and facilitate drug refill for patients. Bangladesh’s new NCD care model for rural areas—where access to diabetes and cardiovascular disease services is particularly poor in Bangladesh—is an attempt to decentralise NCD service delivery. Furthermore, many providers organise community outreach or health ‘camps’ and the country has been piloting and scaling up a wide variety of decentralised models of care for NCDs. Peer and community involvement in service delivery is an important feature of PHC-level care models. Community participation can serve to raise awareness, design appropriate interventions and thus increase acceptance, advocate for political action, provide informal and formal support to patients, mobilise resources, and support sustainability of programmes. Peer support care models, where peers are used to provide social and emotional support, assistance with disease management, and linkage to clinical care and community Considering the huge resources, have been widely applied in health and are starting to gain burden of NCDs and the momentum in NCD care. Considering the huge burden of NCDs and the critical shortages of critical shortages of healthcare staff in rural areas in Bangladesh, there is healthcare staff in rural a need to tap into the community itself to support NCD programmes. areas in Bangladesh, This could include using community volunteers or peers to help with there is a need to tap into NCD prevention, screening and client navigation through the health the community itself to system. support NCD programs. Task-shifting and involvement of a full range of health providers This could include using can be indispensable in a context of shortages and/or unequal community volunteers or distribution of healthcare staff. Task shifting has been shown to be peers to help with NCD effective in improving process outcomes (such as successfully screening prevention, screening for HTN and diabetes) and health outcomes (such as achieving HTN and and client navigation diabetes control) and achieving treatment concordance with doctors. through the health Enablers of task-shifting include appropriate training and provision of system. standard algorithms and protocols while the principal barriers are linked to legislative prescribing restrictions and availability of medicines. Bangladesh’s new NCD care model includes task-shifting of NCD screening to healthcare workers in community clinics. Further task-shifting could involve trained lay workers or volunteers, drug sellers, pharmacists or informal village doctors performing screening, or the expansion of the prescribing list of community healthcare workers if feasible. The active participation of patients in their own treatments through self-management has considerable scope. Interventions to improve self-management can include education, peer support, identification of barriers to treatment adherence, problem-solving to overcome barriers, provision of self-monitoring tools, reminders, goal setting and incentives. However, self- management of chronic diseases can be challenging for patients, as it likely needs daily and lifelong changes to lifestyle, and therefore requires tailored education (regarding content, delivery and frequency) and ongoing support. A 2017 systematic review assessing the effectiveness of self-management interventions for diabetes found that diabetics who xviii EXECUTIVE SUMMARY participated in self-management interventions improved their diabetes knowledge, glucose control, body-mass index, lipid profile and quality of life.12 Elsewhere, chronic care passports have been found to lead to better diabetes and HTN control and better quality service provision. While some self-management is already happening in Bangladesh (including BIRDEM diabetic clients), there are opportunities to promote such approaches further - for example by using different, culturally sensitive patient education material, and by improving access to monitoring tools and peer support. Among the case studies illustrating these five elements of care models, the report also describes South Africa’s Integrated Chronic Disease Management (ICDM) model. While this model speaks to all five elements, it represents a more wide-ranging re-design of service delivery. Important aspects are patient flow in a health facility and early triaging; the use of a chronic care patient record and simplified evidence-based PHC guidelines; task-shifting; decentralisation of health promotion, tracing and support services to outreach teams; and “assisted” self-management within the community. Such a model however requires a radical overhaul of PHC-level service delivery, including appointment systems, a designated chronic patient ‘track’ in facilities and consultation area, and pre-dispensed medication among others. Early evaluation findings based on health care worker reports suggest that South Africa’s ICDM model has improved patient care and interaction, level of knowledge, teamwork and work satisfaction. OPPORTUNITIES FOR STRENGTHENING PRIMARY LEVEL CARE FOR HTN AND DIABETES Based on the review of the evidence and cascade analyses, the report provides the following key recommendations: 1. Coordinate NCD control - An effective coordination mechanism for NCD control is urgently required, with a strong mandate for coordination, measurable targets and clear reporting processes. The coordinating body should have cross-sectoral reach with pluralistic provision. 2. Adequately finance NCD control - It is advisable that the Government consider alternate financing mechanisms. Households need to be protected from impoverishment due to NCD costs. 3. Incorporate NCD needs in human resources for health strategy - Taking into consideration the increasing burden of NCDs, a long-term human resources strategy is required which responds to the changing service needs, and adequate budget allocations need to be made. 4. Address NCD risk factors - Bangladesh needs to raise awareness and risk perception in the population through the implementation of tested communication strategies and promotion of healthier lifestyles to reduce exposure. A substantive focus on preventative and promotive care is urgent, especially in the rapidly urbanising areas. Stronger and additional tax policies should be considered. 5. Mainstream NCD prevention and management - Existing programmes and initiatives, such as MCH, community health and UPHCP, should integrate NCD prevention and care into their activities. 6. Find NCD cases through improved screening capacity - Bangladesh needs to train its health workers to better identify NCD risks and set screening targets at local level to xix HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION help meet, by 2022, the end-of-plan national screening targets for diabetes and hypertension. Protocols and guidelines should be explicit on screening and linkage to care, and be accompanied by job aids for community health workers. 7. Focus on treatment adherence - Essential drugs need to be made available at PHC level at all times as this is a vital part of improving treatment adherence. Guidance on hypertension and diabetes care needs to be explicit about treatment monitoring and adherence support. Health care professionals need to be educated on the guidelines and on the importance to focus on high-risk patients. Assisted self-monitoring and mHealth technology should be promoted. 8. Fully involve the Community Clinics for NCD care - It is vital that the CCs can lead the NCD case finding, linkage to care and management of uncomplicated cases, while facilitating referral as per their mandate. The CCs' outreach function needs to be appropriately resourced for health promotion and NCD screening at scale. Low cost, low technology tools could be considered, such as the “string test for obesity”. 9. Harness existing health work force/structures to address NCD burden - Bangladesh should consider task shifting within its health cadres in order to accommodate the changing needs for long-term monitoring and treatments, and involve paramedics and SACMOs where possible. MCH services could provide nutritional education and support to pregnant women beyond the usual nutritional advice relevant to pregnancy itself. 10. Expand human capacity - Bangladesh should adopt good practices from elsewhere on harnessing community support for the care continuum, the mobilisation of volunteers and lay workers, and the use of peer approaches. 11. Pilot an integrated Chronic Care Passport – Akin to BADAS’ diabetic book, such a patient-held passport is recommended in order to facilitate medication refill at community clinics and support self-monitoring of patients. Such a passport would also help referrals and continuity of care and likely improve the quality of care. 12. Build in NCDs into existing surveillance systems - Bangladesh should incorporate NCDs into existing disease surveillance systems. This will enable the Government to better assess trends and its progress towards SDG Target 3.4 (reducing the premature NCD mortality by one-third by 2030). 13. Strengthen NCD M&E for better decision-making – Bangladesh should roll out the Unique Patient Identifier as a prerequisite for a powerful M&E system, and set NCD service delivery targets to measure progress. The country should document the experiences with the new MOH NCDC model and share the learning. xx INTRODUCTION 1.1 Aim of this report Bangladesh faces a health transition challenge as its health outcomes improve and it transitions from a health system dominated by infectious disease and maternal and child health challenges to one confronted increasingly by non-communicable diseases (NCDs). The country’s NCD statistics highlight the urgent need to re-configure service delivery, The new NCD make more strategic investments, and find new solutions to reduce NCD Operational Plan risk factors, prevent disease sustainably, and manage complications. The provides a 5-year window new NCD Operational Plan provides a 5-year window to redress the to redress the NCD situation by moving the public health care system from a disease- situation by moving the specific one designed for infectious disease and mother & child health public health care system (MCH) to an integrated care system emphasising integrated primary from a disease-specific health care (PHC), out-patient care and community health promotion. one designed for This report aims to provide a review of the NCD situation by using the infectious disease and “cascade” framework—drawing on the indicators diagnosed, treated, mother & child health controlled—to determine the continuum of care for hypertension and (MCH) to an integrated type-2 diabetes mellitus in Bangladesh (section 2). It then care system emphasising summarises the country’s policy and health care delivery integrated primary environment for NCDs (section 3) followed by an assessment of health care (PHC), out- models of NCD and chronic care in Bangladesh and elsewhere patient care and (section 4). Based on the evidence reviewed and analyses conducted, the community health report then provides conclusions and recommendations (section 5). promotion. The methodology consisted of literature review, consultations with experts, data triangulation, benchmarking, and synthesis of the published evidence. For the continuum of care, the cascade framework was used to assess the stages from screening to diagnosis, treatment initiation, adherence and disease control. PHC-level models from other countries were selected opportunistically to capture promising practices. The objective of this technical assistance is to provide input into the dialogue on how to improve NCD integrated care and outcomes in Bangladesh. The support is provided within the World Bank’s assistance to countries to strengthen NCD care through data-driven resource allocation and decision-making. 1 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION 1.2 Global context NCDs now account for more than half of the global burden of disease, and cardiovascular diseases (CVD) alone are responsible for half of NCD-attributable deaths of which the majority occur in low- and middle-income countries (LMICs).13 Death and disease due to NCDs matter in their own right, but they also act as important barriers to poverty reduction and sustainable development. The global call for action was strongly made in the 2011 High-level Meeting of the UN General Assembly on the Prevention and Control of NCDs.14 This Political Declaration was followed by two NCD-specific targets in 2015 in Sustainable Development Goal 3 “Ensuring healthy lives and promote well-being for all at all ages”, with targets to i) Reduce by one third premature mortality from NCDs through prevention and treatment and promote mental health and well-being (target 3.4), and ii) Support the research and development of vaccines and medicines for the communicable and non-communicable diseases that primarily affect developing countries and provide access to affordable essential medicines and vaccines (target 3.b).15 Despite these high-level endorsements, progress in addressing NCDs is uneven and does not match the growing need. NCDs as cause of death has between 2000 and 2015 increased significantly in country groups Bangladesh belongs to: the lower middle income countries (LMICs), the International Development Association (IDA)-eligible countries, and countries in the South Asia region (Fig. 1.1). Figure 1.1 NCDs as cause of all deaths by country group, % (2000, 2015) Source: World Development Indicators, indicator ‘cause of death, by NCD (% of total)’. Note: Bangladesh is part of the Lower middle income level countries, in IDA country group, and in South Asia region. Since many health systems in lower income economies have been focusing on infectious and childhood diseases, there is often low capacity to confront the emerging disease pattern dominated by NCDs. The implications of the poorly addressed NCD epidemics are vast: There is a social burden which is multifaceted and ranges from prolonged disability of NCD cases, to diminished resources within families due to out-of-pocket payments for services and drugs, reduced productivity due to illness, and a tremendous demand on health systems for acute and long-term NCD care. This translates into a staggering economic burden. It has been estimated that during 2011–25, cumulative economic losses due to NCDs under a “business as usual” scenario in LMICs amounts to US$ 7 trillion.16 This far outweighs the annual US$ 11.2 billion 2 INTRODUCTION cost of implementing a set of high-impact interventions to reduce the NCD burden.17 The economic case for NCD investments is therefore clear. The 2013 global survey on national capacity for the prevention and control of NCDs18 identified a number of missed opportunities and made two overarching recommendations to address the NCD challenge: 1) To strengthen NCD governance, prevention, health care, and surveillance and monitoring; and 2) To support NCD-related research and evaluation of the impact of interventions and policies. The World Health Organisation (WHO) provided in its 2014 NCD report several key recommendations to guide the national and global NCD response.19 It emphasised the importance for better tracking of financial resources for NCDs. Also, it called on programmes to prioritise cost-effective population-wide and individual interventions, including the “best buys” identified in 2011.20,21 Supporting implementation research and its translation into practice was also highlighted, and the necessity to produce high-quality case-studies to better understand why progress has not been made. 1.3 Bangladesh health transition Bangladesh has a population of 162.95 million.22 Currently, about 65% of In the 2016 the population live in rural areas but the country is rapidly urbanising. The Household Income demographic and epidemiological transition leads to an ageing population and Expenditure and shifting disease burden from infectious, communicable diseases to Survey (HIES) , chronic NCDs (Annex 1, Fig. A 1.1). In 2016, NCDs were responsible for hypertension, 62.7% of Disability-adjusted life years (DALYs) while communicable, chronic heart maternal, neonatal and nutritional diseases accounted for 26.8% and disease and injuries for 10.6% of DALYs respectively.23 Ischemic heart disease (IHD), diabetes were stroke and type 2 diabetes are the main contributors to the NCD burden. In among the most the 2016 Household Income and Expenditure Survey (HIES)24, reported illnesses hypertension, chronic heart disease and diabetes were among the most with rising levels of reported illnesses with rising levels of reported diabetes in females between 2010 and 2016 (Fig. A 1.2, A 1.3). reported diabetes in females between Benchmarking Bangladesh health indicators with comparison countries: 2010 and 2016.  Bangladesh has overall achieved great success in reducing mortality, pre-mature death and improving life expectancy. This is especially compelling when comparing across countries with similar per capita income levels (Annex 2, Table A 2.1), but even within the South Asia Region, Bangladesh compares well (Table A 2.2): Life expectancy at birth has increased to 73 years (2016) from 58 years in 1994.25 Large gains have been made in child survival, and maternal mortality has seen important improvements.26  However, Bangladesh is ranking poorly in comparisons on premature death due to NCDs. Among 12 countries with similar per capita income levels, Bangladesh takes rank 11 for pre-mature death due to all NCDs combined, rank 9 for IHD, rank 11 for diabetes mellitus, and rank 12 for cerebrovascular disease, chronic kidney disease (CKD) and chronic obstructive pulmonary disease (COPD) (Table A 2.3). Benchmarking within the South Asia Region, the poor rankings on pre-mature mortality remain for overall NCD burden, cerebrovascular disease, and COPD, but not for diabetes, IHD and CKD (Table A 2.4). Most of Bangladesh’s premature deaths from IHD, stroke, COPD and diabetes are attributable to risk factors (Annex 1, Fig. A 1.4). High systolic blood pressure is estimated to be responsible for 3 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION 18% of all deaths in Bangladesh, and smoking for an estimated 17%. High fasting plasma glucose may be responsible for about 11% of deaths. The Global Burden of Disease (GBD) analysis for Bangladesh notes that the greatest disease risk factors are now NCD risk factors: “Overall, the three risk factors that account for the most disease burden in Bangladesh are tobacco smoking, household air pollution from solid fuels, and dietary risks”.27 Regarding smoking, Bangladesh has actually had progress in tobacco control in recent years. Tobacco excise taxes have been raised, but, at 62%, fall short of the WHO-recommended 70% minimum.28 Nicotine replacement therapy and cessation services are partially available, but patient costs are not covered. Although it is recommended that all public places are completely smoke-free, restaurants and public transport are not yet smoke-free, and funds for smoke-free enforcement are insufficient. Bangladesh has exceptionally high use of ‘smokeless tobacco’ (SLT); a quarter of adults use SLT.29,30 SLT products are cheap and readily available, and misconceptions regarding its deleterious health effects, lack of The country needs the comprehensive tobacco control regulations and weak enforcement of right service delivery existing regulations are contributors.31 One WHO study found that systems to reduce tobacco was a major risk factor in Bangladesh and cost the economy population exposure to about $44 million annually.32 behavioural risk factors, slow the progression to For many people in Bangladesh, NCD risks are already set in childhood:33 22% of babies are born with low birthweight, 36% of disease in exposed people, children under five are stunted, and 33% are underweight.34 This, and address compounded by lifestyle factors such as air pollution, poor diet and complications to prevent nutrition, high salt intake, tobacco use and insufficient physical activity, disability and premature precipitates NCDs often at a relatively young age. The underlying social death. determinants like poverty, low education and urbanisation are complex to change, and the interplay of the NCD determinants can have long-lasting, transgenerational impacts. Many of these underlying factors are of course outside the direct influence of the health sector and rely on a “comprehensive government” approach. In summary, Bangladesh’s health transition with the rising NCD burden and the associated costs requires a reorientation of the health care system. The country needs the right service delivery systems to reduce population exposure to behavioural risk factors, slow the progression to disease in exposed people, and address complications to prevent disability and premature death. "One WHO study found that tobacco was a major risk factor in Bangladesh and cost the economy about $44 million annually.” 4 THE CONTINUUM OF CARE FOR HYPERTENSION AND DIABETES This analysis used the cascade framework and applied it to male and female adults in urban and rural areas, respectively. It focused on hypertension as a tracer condition and type-2 diabetes as a cost-driving NCD, in order to devise the continuum of care for chronic NCDs in Bangladesh. The simple concept of the continuum or cascade of care provides a useful lens to assess the delivery of chronic care, and help re-orientate health systems for improved chronic care outcomes. The continuum or cascade of care asks four questions about a patient’s health journ ey and trajectory:  First, is a patient diagnosed if he/she has a health condition?  Second, is the patient linked to appropriate health care?  Third, is the patient adherent to the required care regimen?  Fourth, does the patient achieve disease control? Failure at each stage of the cascade precludes success at the next, which Failure at each stage means the cascade of care may tumble rapidly with the breakpoints at of the cascade each stage adding to the final outcome. For hypertension and diabetes precludes success at care in Bangladesh, cascades are presented separately for adult female the next, which means and male residents in rural and urban areas. The most comprehensive national data came from the 2011 Demographic and Health Survey (DHS) the cascade of care which measured blood pressure (BP) and blood glucose in a subset of may tumble rapidly survey participants aged 35 years and above. These population-based with the breakpoints biomarker data from 2011 were applied to 2016 population estimates in at each stage adding order to generate estimations along the continua of care. to the final outcome. 2.1 Hypertension cascade, breakpoints and outcomes Hypertension prevalence has been reported from various studies conducted in Bangladesh.35 A recent meta-analysis reported an overall hypertension prevalence of 14.0% (95% CI: 11.0%– 16.0%), across a diversity of age groups and populations.36 2016 estimates of adults with hypertension in each of the eight Divisions are provided in Annex 3, suggesting that the hypertension burden varies by a factor 5 across Divisions, and diabetes burden by a factor 4. The only national data source providing all the key data for the cascade analysis was the 2011 DHS in which blood pressure measurements were done in a sample of participants aged 35 years and above. Hypertension prevalence in these DHS participants aged 35+ was 26%37. Based on this, we estimated that in 2016, 14.36 million adults aged 35+ had hypertension and another 15.35 million were pre-hypertensive38 - see Fig. ES 3 for the national cascade, and Fig. 2.1 for 5 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION rural and urban female and male residents. The prevalence of hypertension was highest in females living in urban (40.2%) and rural (29.4%) areas, followed by urban males (25.2%), and lowest in rural males (17.6%). Overall, 51% had been diagnosed. Therefore, about 7.25 million adults aged 35+ years are undiagnosed hypertensives. The absolute gap to diagnosis was largest in rural females (2.5 million undiagnosed, or a relative gap of 46% of all hypertensives) and lowest in urban males (1.2 million undiagnosed, or 50% of hypertensives). Treatment among those diagnosed was high based on self-reported taking of hypertension medicine, however, only 38.2% (urban males) to 49.2% (rural males) actually had treatment success by achieving normal BP. This means that only 2.8 million adults aged 35+ years have their hypertension controlled, and over 12 million have uncontrolled hypertension. Other analyses of the same data showed that the risk of hypertension is significantly associated with older age, being female, higher educational attainment, higher wealth, urban residency, higher body mass index (BMI) and having diabetes.39 Community and facility based NCD screening in Jessore District found 27% hypertension prevalence (Nov. 2013 – Jan. 2016, 14,023 participants).40 Overall, 48% were aware of being hypertensive, which is similar to the levels observed in the 2011 DHS. A study among university students in Dhaka showed that hypertension is also an issue among younger people. It reported a 6.5% prevalence (12.1% in males, 3.4% in females).41 Figure 2.1 Hypertension cascade for rural and urban residents aged 35 years and above Sources: DHS 2011; Rahman et al 2017; 2016 UN population estimates; World Bank data for rural residency. In the same 2011 DHS data, one in 10 females had stage 2 hypertension.42 It can be estimated that in 2016, about 4.1 million adults aged 35+ years had stage 2 hypertension either on treatment or without treatment, with the largest share among rural women (40% of all stage 2). Hypertension is the most important risk factor for stroke and heart attacks (see Fig. A 1.4 for deaths attributable to risks). But hypertension also leads in itself to hospital admissions. In a 12- month period up to end October 2017, the District Health Information System (DHIS2) counted 6 CONTINUUM OF CARE FOR HYPERTENSION AND TYPE-2 DIABETES 22,805 hospital admissions for primary hypertension and 8,322 admissions for hypertensive disease. Annex 4 shows the hospital in-patient admission rates for each division (the health information system does not collect outpatient data disaggregated by reason for clinic visit/ disease). In order to improve the continuum of care for hypertension, an understanding of the main barriers on the patient and the provider sides is helpful to plan relevant interventions (see Annex 5 for an overview). Based on the literature, chief barriers to diagnosis are the low priority given to case finding at community level (see also Box 2.1 on screening), combined with insufficient population awareness of the risk and symptoms of hypertension, whereas barriers to treatment and treatment adherence are the recurrent drug costs, the weak capacity for supporting patients in chronic care at PHC level and poor risk perception regarding life-threatening complications of hypertension. Box 2.1 What do the National Hypertension Guidelines say regarding the cascade stages? WHO in collaboration with the Directorate General of Health Services (DGHS) and the National Heart Foundation published hypertension guidelines in 2013.43 Hypertension treatment is mostly provided by general practitioners and specialist physicians, including cardiologists, through government and private healthcare services. This guideline wants to be relevant and applicable to the primary care setting, where primary care physicians often work with limited resources. The guideline places emphasis on the affordability of the treatment for hypertensive patients. It promotes the importance of lifestyle modifications (as health- supporting and as a way to limit medication need), and encourages the use of the cheapest generic drugs if equivalent. Given the low cost and availability, it also endorses beta-blockers as first line therapy for resource constrained settings in Bangladesh. The guideline provides specific information about stepping up and stepping down treatment depending on a patient’s response to therapy. There are also instructions regarding follow up and target BP levels. If a patient fails to achieve the BP target after 3–6 months of treatment, the guideline advises referral to a secondary or tertiary center for further evaluation. The guideline does not provide operational detail on screening for hypertension, and provides little guidance on treatment adherence support. 2.2 Type-2 diabetes cascade, breakpoints and outcomes The prevalence of diabetes has risen rapidly from about 0.7% in 1980.44 Globally, Bangladesh Diabetes prevalence in the 2011 DHS participants aged 35+ years with is among the top 10 glucose measurement was 11.0%. The meta-analysis by Islam et al. countries for number (2016) reported an overall diabetes prevalence of 6.0% (95% CI: 5.0%– 7.0%) across a variety of studies on different age groups and populations, of diabetes cases, and and the 2017 Diabetes Atlas reports 6.8% prevalence for Bangladeshi has a relatively low adults aged 18 years and above.45 Globally, Bangladesh is among the top diagnosis rate in 10 countries for number of diabetes cases, and has a relatively low international diagnosis rate in international comparison (rank 45 of 221, see Annex 6). comparison. Again, the only national data source providing key data for the diabetes cascade analysis was the 2011 DHS. No population level data were available on the coverage of diabetes screening. Based on our estimations, 6.13 million adults aged 35+ years had diabetes46 in 2016, and 13.73 million were pre-diabetic (Fig. ES 4)47. The prevalence of diabetes was highest in urban females 7 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION (17.3%) followed by urban males (14.9%), and virtually the same in rural females (9.5%) and males (9.3%) (Fig. 2.2). The lower diabetes prevalence in rural areas is consistent across studies, however, representative epidemiological data is scant and the ranges found across studies are quite large.48 Based on the same DHS 2011 data, about 3.8 million adults 35+ years are undiagnosed diabetics. The absolute gap to diagnosis was largest in rural males (1.2 million undiagnosed, or a relative gap of 73% undiagnosed diabetics) and lowest in urban males (0.72 million undiagnosed, or 49% of diabetics). Undiagnosed diabetes prevalence in adults 35+ years was therefore 6.7%. This matches results from a survey reporting a 6.6% prevalence of undiagnosed diabetes in adults aged 20+ years (urban 12.2%, rural 2.6%).49 Figure 2.2 Diabetes cascade for rural and urban residents aged 35 years and above Sources: DHS 2011 (with secondary data analysis); 2016 UN population estimates; World Bank data for rural residency. Similar to hypertension, treatment among those diagnosed with diabetes was high based on self- reported taking of medicine and treatment control was poor with only 29.2% (rural males) to 42.9% (rural females) having treatment success by achieving normal fasting plasma glucose while taking medication. This means that only 12.3% of adults aged …significant inequality 35+ years have their diabetes controlled, and close to 5.4 million in diabetes are diabetic without controlling their fasting plasma glucose. Other management was analyses of the same data suggested that participants who had a lower found between poor education and lower economic condition were less likely to be aware of and wealthy their diabetes.50 Poor management was observed among non-educated and low-income groups. Furthermore, significant inequality in diabetes households. management was found between poor and wealthy households: 18.2% vs. 63.2% (awareness), 15.8% vs. 56.6% (treatment), and 8.2% vs. 18.4% (control). In the survey described by Alam et al. (2016), urban residence, higher age, overweight and abdominal obesity were significant predictors of diabetes. The 2010 STEPS survey asked about the type of treatment people were taking: 21% of diabetics on treatment reported receiving insulin and 8 CONTINUUM OF CARE FOR HYPERTENSION AND TYPE-2 DIABETES 61% oral anti diabetic drugs.51 In another study, 46% reported using metformin, 41% reported insulin, and 39% reported sulfonylurea use.52 Undiagnosed as well as treated but poorly controlled diabetes can lead to hospital admissions. In the 12 months to October 2017, the DHIS2 system53 counted 5,070 hospital admissions for type-2 diabetes and 937 admissions for diabetes-related complications. Annex 4 shows the rates for each division. A study conducted at a tertiary diabetes hospital in Dhaka estimated that of 473.43 DALYs forgone due to diabetes, only 20.5% were due to diabetes itself while the rest (79.5%) was due to its complications. The study also estimated that the most number of DALYs forgone were incurred by diabetes with CVDs (38%), followed by diabetes with retinopathy (20%), and diabetes with nephropathy (13%).54 The barriers across the diabetes care continuum were also reviewed and summarised (see Annex 5). Limited availability of diabetes services is most marked in rural and peri-urban slum areas where patients report that effective diabetes services are scarce.55 Diabetes is often an incidental finding when people seek medical care for other problems such as TB, or as a result of severe complications of diabetes. Likewise, continuity of diabetes care is poor in peri-urban slum areas with high costs, meaning that people delay follow-up visits. In rural areas, patients at community clinics (CCs) need to travel to the local sub-district hospital for basic diabetes care, and to a specialist hospital of the Diabetic Association of Bangladesh (BADAS) for comprehensive services. For diabetes cases, disease-specific knowledge and attitudes are Patients’ non- important factors throughout the continuum. Several studies report that adherence to males show higher levels of relevant knowledge, but females showed treatment is better attitudes and practice towards diabetes compared to correlated with poor males.56,57,58,59 Diabetes knowledge is strongly associated with education and income.60 In turn, better disease control i.e., normal fasting plasma understanding of the glucose (FPG) while on medication is seen in persons with higher disease management education and higher wealth (2011 BDHS). One study found that about due to lack of 80% of patients monitored their blood sugar at least once every 3 adequate education months.61 In Fatema’s report (2017), 38% of diabetes patients measured from health staff. their blood sugar once in a month either at home or at the hospital and 37% checked their blood sugar at least once in 3 months.62 Only 3% tested their blood sugar once in a year, and 2% never got checked. Difficulties in attending regular or monthly check-ups can stem from lack of family support for such monitoring visits. Patients’ non-adherence to treatment is correlated with poor understanding of the disease management due to lack of adequate education from health staff. The below quotes, selected from a qualitative study carried out among diabetic out-patients at the Bangladesh Institute of Health Science hospital in Dhaka, highlight this point: “I have diabetes, I take five different medicines, but I am not sure which ones are specifically for diabetes and the dose of each medicine” “ … nobody ever informed me about the side effects of the medicines and what I should do if I have any problem” (Diabetes patients)63 In one multi-site study in Bangladesh, 72% of prescriptions did not have directions on how to take the medication and 62% did not specify the duration of treatment.64 An assessment among medical officers working in BADAS hospitals found that physicians with certificate course on 9 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION diabetology, and junior physicians had significantly better prescribing practices than physicians without the course and Senior Medical Officers/consultants, respectively.65 Medication costs are another chief reason for non-adherence, as these can put strains on families, as is highlighted in the below quote from a diabetic patient who took part in another qualitative study, conducted in urban centres. “I only take medicines when I can afford them; it is too much to pay for them all the time … I have to think of my family” (Diabetes patient)66 Box 2.2 What do the National Diabetes Guidelines say regarding the cascade stages? The 2013 National Guidelines for the Management of Diabetes Mellitus at Primary Health Care by DGHS, BIRDEM and WHO67 is published in Bangla to guide diabetes management at primary care level. Regarding screening, the document provides the risk factors for T2DM and gestational diabetes, but no detail on targeting of screening or screening strategies is given. Three tests are mentioned for the primary care level (OGTT, FPG, RBG). However, more implementation guidance could be provided and HbA1C testing included as a diagnostic method for T2DM (it is included for type-1 and gestational diabetes). No risk factor scoring is referred to. Diagnosis of gestational diabetes and type-1 diabetes in young people is well covered. Guidance on treatment maintenance and refill is not given, although antidiabetic drugs should be supplied at the PHC level. Another area which is not well addressed is glucose monitoring, and how patients can be supported for self-monitoring. Acute, delayed and chronic complications of diabetes and how to prevent these, and the need for clinical consultation along with laboratory tests, are however mentioned in the guideline. The 2003 Guidelines for care of type-2 diabetes mellitus in Bangladesh are more widely disseminated. They provide comprehensive guidance across the continuum of care, including targeting and frequency of screening, diagnostic tests and case definitions (with flowchart), initiation of non-pharmacological and pharmacological treatments, follow-up visits during treatment consolidation, the use of patient self-monitoring data by the physician, and criteria for referral. Section 3 summarises the policy and care provision environment relevant to the delivery of hypertension and diabetes services. “Medication costs are another chief reason for non-adherence, as these can put strains on families…” 10 NCD SERVICE PROVISION AND COSTS ACROSS THE CONTINUUM OF CARE 3.1 Policies, strategies and plans on NCD control To address the NCD epidemic and shifting demands on the health sector, the Government of Bangladesh and several non-government organisations (NGOs) have published a number of policy documents, strategies and plans. This has provided a conducive policy environment for NCD control in Bangladesh, but has not necessarily been harnessed for NCD control activities (Box 3.1). A recent comprehensive NCD policy analysis68 states that “although the government has initiated many NCD-related policies or programmes, they lacked proper planning, implementation and monitoring” and concludes that the growing burden of NCDs has not been effectively addressed (the analysis covered 51 NCD-related policy documents that have been published since Bangladesh’s independence in 1971). Box 3.1 Bangladesh's NCD policy context in international comparison Using the 2017 Global NCD Monitor data to compare across 12 similar economies:69 We compared the WHO-reported data on NCD policies, targets, guidelines and action plans, as well as on targeted policies for demand-reduction (tobacco, alcohol, salt, unhealthy diet) across 12 countries with similar GNI per capita. Bangladesh came out with the highest overall score among this country group, suggesting a comparatively strong NCD policy environment. The scores, however, showed gaps in addressing unhealthy diet and food marketing in Bangladesh, among other gaps (Annex 7). Using the 2014 ‘NCD scorecard’ data to compare across countries in different income groups:70 The NCD scorecard study assessed countries’ performance on tackling NCDs across the four domains of governance (which included national plans and leadership), action on reducing risk factors, conducting research and surveillance, and responding to NCDs through the health system. The study scored Bangladesh as a medium performer with regards to governance and a poor performer across the remaining three domains. The health systems response for NCDs in particular was far better developed in upper-middle and high income countries, presumably due to a longer history of NCD burden and management (Annex 7). The current Operational Plan for NCDs within the 4th Health, Population and Nutrition Sector Programme (2017‒22) relates that during the 3rd sector programme (2011‒16), “control of NCDs was given priority but with inadequate resource allocation...[and] most of the resources were used in purchasing equipment, training of manpower and arsenicosis and emergency related activities”. 11 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION Four NCDs—CVDs, cancer, diabetes mellitus, and arsenicosis—are being treated as major public health concerns. Key activities were the 2010 STEPS survey, the piloting of the Package of Essential NCD (PEN) interventions in 2013 at Satkhira in Debhata Upazilla Health Complex (UZHC), establishment of Health Promotion Model villages for NCD control in a few areas, and the model school initiatives in 91 schools. Other measures were the creation of health literacy by courtyard meetings, facility based education, use of mass media, and establishment of the “NCD corner” in health facilities. …the better coordination The 2014 mid-term review made clear recommendations on how of non-state-actors in NCD NCDs need to be better addressed in a comprehensive manner and with effective regulatory mechanisms. The current NCD Operational control with the public Plan (Box 3.2) aims to address the barriers to effective action. There sector is critical to achieve is awareness that the better coordination of non-state-actors in NCD objectives, as is a NCD control with the public sector is critical to achieve these stronger focus of the public objectives, as is a stronger focus of the public sector on prevention sector on prevention and and health promotion, with investment in intensive care units and health promotion... tertiary care largely left to the burgeoning private sector. Box 3.2 Objectives of the NCD Control Operational Plan 2017‒22 The current Operational Plan for NCDs has as its objective to reduce mortality and morbidity from NCDs in Bangladesh through strengthening of evidence-based measures for control of risk factors, and health service delivery options for early detection and management. Its specific objectives are: a. To promote development and implementation of effective, integrated, sustainable, and evidence-based public policies for chronic NCDs, their risk factors, and determinants b. To develop and strengthen capacity for surveillance of chronic NCDs, their consequences, their risk factors, and the impact of public health interventions c. To foster, support, and promote social and economic conditions that address the determinants of chronic NCDs and empower people to increase control over their health and to adopt healthy behaviours d. To strengthen the capacity and competencies of the health system for the integrated early detection and management of chronic NCDs and their risk factors 3.2 Financing and resources for NCD control It is estimated that total health spending in Bangladesh in 2014 was $14.7 billion (2015 Purchasing Power Parity (PPP)) (Annex 8 Fig. A 8.1). Government contribution was about 22%, while Out-of-pocket (OOP) expenditure by households was about 65% and development assistance for health was about 12% of all spending.71 In fiscal year 2015‒16, the total allocation under the revised annual development programme (RADP) for the DGHS was BDT 237,590.00 lakh. Among all the 19 operational plans of the Health, Population and Nutrition Sector Development Plan (HPNSDP) 2011‒16, the Non- Communicable Diseases Control (NCDC) Operational Plan was allocated 1.73% of the total budget and 65% of this allocation to NCDs was spent over the Plan period (Fig. A 8.2 for allocations and level of expenditure).72 12 NCD SERVICE PROVISION AND COSTS ACROSS THE CONTINUUM OF CARE Budgets for the current HPNSP period of 2017‒22 foresee funding for the provision of health services (61.5% of total), health system strengthening (37.0%), but also governance and stewardship, training, education, and some research, which all have relevance to NCD control.73 In this 5-year budget, the NCDC Operational Plan itself obtains 4.2% of the total funding or BDT 111,827.27 lakh. (Fig. 3.1). This translates to BDT 6.8 per capita (or 8.2 U.S. cents). Depending on the operational plan, there are different sources of funding. The NCDC operational plan is funded by 65% from government sources and 35% other sources (Fig. A 8.3). Figure 3.1 Division of funding amongst the health service programmes in 2017‒22 Source: Operational plans ,Programme Implementation Plan, 4th Health, Population and Nutrition Sector Program 2017‒22. Notes: MNCAH=Maternal, Neonatal, Child and Adolescent Health; HSM=Hospital Services Management; CCSDP=Clinical Contraception Services Delivery Program; MCRAH-Maternal, Child, Reproductive and Adolescent Health; CDC- Communicable Disease Control; AMC=Alternative Medical Care; LHEP=Lifestyle and Health Education & Promotion; CBHC=Community Based Health Care; TB-L&ASP=Tuberculosis- Leprosy and AIDS/ STD Program; FP-FSD=Family Planning-Field Services Delivery; NCDC=Non Communicable Diseases Control; NNS=National Nutrition Services; IEC=Information, Education & Communication; NEC=National Eye Care. As mentioned above, the total amount allocated for the implementation of the NCDC program in 2017‒22 is BTD 111,827.27 lakh. The bulk of this budget is for supplies and services costs (BDT 101,378.45 lakh), including vaccines and medications (20.6% of the total budget), conferences/workshops (16.3%), training (12%) and survey costs (5.2%). Advertising and publicity costs are 8% of the total budget estimates, printing and binding another 5.6%, and research expenses are 3% (NCDC Program 2017‒22). It is worth noting that NCD-related health services are also benefiting from support through other operational plans like the Community Based Health Care (CBHC) Operational Programme and the Urban Primary Health Care Services Delivery Project, a project of Ministry of Local Government and Rural Development. In those programmes and projects, some elements of NCD control accrue, from capacity building, education, training or facilities’ development to the provision of health services. The cross cutting issues like air pollution, poverty, mother and child or gender issues are being tackled in all operational programmes. Furthermore, it is important to state that an estimated two-thirds of health spending comes from households through out-of-pocket (OOP) payments (section 3.4. for further detail). This 13 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION OOP contribution amounted to $9.6 billion in 2014 (2015 PPP)74—see Annex 8, Fig. A 8.1. Public sector contribution to health was about 22%, and development assistance close to 12%.75 3.3 Health infrastructure relevant to NCD control The Ministry of Health and Family Welfare (MOHFW) is responsible for health service delivery through various implementing authorities. Different directorates are responsible for 29 health operational programmes. Box 3.3 provides an overview of the public sector health care system. Community Clinics (CCs), one of the governmental flagship programmes, are providing a limited amount of services related to NCDs. The role of CCs is to provide basic curative care but also some health promotion and prevention (as per CBHC 2017‒22). Diagnosis and disease management for hypertension and diabetes is carried out by medical doctors which limits these activities to occur at the Sub-district level facilities and above only. Box 3.3 Public sector health system in Bangladesh The health care infrastructure under the DGHS has six levels: national, divisional, district, upazila (subdistrict), union, and ward. The national level service delivery system is organised around public health functions, and postgraduate medical education and training and specialised treatment for patients. Tertiary health care is provided at national or divisional level, secondary care is provided in district or general hospitals, and primary care is provided at upazila level and below down to CCs. Primary care: Nationally, there were 13,336 CCs in operation by 31 July 2016. On average, there is one clinic per 7,000‒10,000 citizens (CBHC 2017‒22). From July 2015, all the activities of CCs are being carried out based on the DGHS’s CBHC Operational Plan. Emergency and complicated cases are referred to higher level facilities. There are about 4.3 Government community health workers per 10,000 population and they are distributed fairly evenly whereas those supported by NGOs are concentrated in certain areas. At the union level, there are 51 primary-care hospitals with 830 beds and 3,083 health facilities for outpatient services only (3,134 PHC level facilities at the union level in total). Overall there are 482 primary-care government hospitals at the Upazila level, which have 19,508 hospital beds (see Annex 8 Table A 8.1. for facility distribution and beds in rural areas). In total there are 16,968 public health facilities including all hospitals and ambulatory care centers at primary care level. Higher-level care: There are 131 public hospitals and facilities (secondary and tertiary under DGHS), out of which there are 14 hospitals and 117 secondary and tertiary facilities, with a total of 29,426 beds. The specialised public sector hospitals are located in urban settings, and there are also 35 urban dispensaries and 23 school health clinics in some of the bigger cities. Since the establishment of CCs in 2009, the number of CC clients has been increasing and is now reaching approximately ten million visits every month. Women and children are the ones seeking help most frequently (80% of all visits between 2009–16, CBHC 2017‒22). Table 3.1 summarises the visiting frequency in the public health sector. Within the public sector, the highest number of service use is ambulatory visits at outpatient departments (OPDs), yet the reporting within the DHIS2 does not provide disease-specific statistics from OPDs which could be used for analysis of NCD caseloads. 14 NCD SERVICE PROVISION AND COSTS ACROSS THE CONTINUUM OF CARE Table 3.1 Public sector health facilities visiting numbers per year APPROXIMATE NUMBER VISITS PER 1,000 FACILITY TYPE OF VISITS PER YEAR POPULATION Community clinics 120 million 736 Ambulatory care at outpatient departments 178.7 million 1,097 Emergency departments of different types of facilities 7.4 million 45 District hospitals under DGHS 5.71 million 35 Tertiary level hospitals 620,000 3.8 Sources: Health Bulletin 2016; CBHC 2017‒11. Notes: Bangladesh also has a large private health care sector with 2,630 private hospitals and clinics, including 40 private medical colleges (compared with 120 secondary and tertiary public sector hospitals and 482 UZHCs in the public sector);76 The number of registered private hospitals and clinics in 2016 was 4,596 and that of registered private diagnostic centers was 9,741, with a total of 78,426 beds in hospitals and clinics. The country also hosts the world’s largest and most dynamic NGOs, The health workforce in supporting health care delivery in a very significant way (see section 4 Bangladesh is for some NGOs working on hypertension and diabetes). The traditional substantially smaller and community health workers in rural areas are a legacy of village- than in other based care systems, but also reflect the growth of informal markets. countries...well below These workers fill a void in healthcare service delivery due to the WHO’s minimum inadequate numbers of and incentives for more formally trained public recommended numbers sector health workers to serve in those communities. Each village in of doctors, midwives Bangladesh has “village doctors,” and in the village and sub-district markets, drug vendors (often the same person) at unregistered drug and nurses required to retail outlets. provide basic services. Human resources in the health care sector are overall relatively constrained, creating challenges for the health system and patients themselves. The health workforce in Bangladesh is substantially smaller than in other countries, which at 0.58 per 1000 population is well below WHO’s minimum recommended numbers of doctors, midwives and nurses required to provide basic services (2.28 per 1000 population).77 In 2011–12, Bangladesh also had about 219,000 CHWs (i.e., 13.7 per 10,000). Of these workers, about 56,000 were government workers (23,500 family welfare assistants, 19,300 health assistants, and 12,991 community health care providers) and about 163,000 were supported by NGOs (over 100,000 supported by BRAC).78 The urban health care system is dominated by the private sector while the urban health services is controlled by local government institutes (City Corporations and Municipalities). Local government bodies have executed several projects with development partners to strengthen urban health care especially to reach the poor (Box 3.4). Box 3.4 Urban health care initiatives The Local Government Division has over the years implemented three health projects: The Urban Primary Health Care Project (1998‒2005), the Second Urban Primary Health Care Project (2005–2011)79, and, evolving from these projects, the Urban Primary Health Care Services Delivery Project (July 2012 to June 2017) with the financial support of the Asian Development Bank, Swedish International Development Cooperation Agency and the United Nations Population Fund. MOHFW provides services to the urban population through their outdoor dispensaries and secondary and tertiary hospitals. In total, there are around 4,000 satellite centers to reach the urban population. A total of 35 urban dispensaries under the DGHS are providing outdoor patient services including an expanded programme on immunisation (EPI) and MCH interventions to the urban population. 15 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION 3.4 NCD interventions and costs Controlling hypertension Key issues in hypertension Diagnosis and control of raised blood pressure is a cornerstone of the prevention of stroke, IHD and peripheral vascular disease control include the efficiency and is cost-effective in all regions of the world.80 Key issues in of identifying those who hypertension control include the efficiency of identifying those would most benefit from who would most benefit from treatment, how to improve access treatment, how to improve to medications and treatment adherence, and how best to deliver access to medications and medications. One trend has been to evaluate the overall NCD risk treatment adherence, and of a patient rather than just BP, and several studies have shown how best to deliver that it is more cost-effective to determine overall CVD risk medications. instead of focusing on BP (or cholesterol) levels alone.81 Controlling diabetes Early detection of diabetes confers opportunities to slow disease progression, address comorbidities like hypertension, and lower the costs of managing advanced diabetes and its complications. Diabetes meets key criteria for screening with reliable tests to identify elevated glucose, a precursor phase (prediabetes), and interventions to delay onset and manage the disease.82 Given that interventions to prevent diabetes among Given that interventions to persons with prediabetes are cost-effective,83 screening should prevent diabetes among always also take into account the prediabetes range of results.84 persons with prediabetes are Equally, regular screening to detect and treat diabetes cost-effective, screening complications in their early stages is generally recommended, should always also take into before irreversible clinical manifestations and disability set in.85 Since there is a preclinical phase before each complication sets account the prediabetes range in, diagnosed diabetes cases should undergo regular eye, foot, of results. and urine checks. The following sections give broad estimates of costs, using best available data from Bangladesh or elsewhere. The clinical heterogeneity of hypertensive and diabetic patients makes overall case management costs highly variable. Screening and diagnosis costs Hypertension: The average screening cost is approximately US$ 3.90 for LMICs (BDT 322).86 This includes the cost of performing a NCD risk assessment and BP measurement at the primary care level. Diabetes: The average cost of screening is approximately US$5 (BDT 411).87 This estimate is for glucose screening at PHC level, and includes documentation, setting up referrals and organising screening events. Relevance for Banladesh: Finding 50% of the currently undiagnosed 7.1 million hypertensive people aged 35+ years would cost about US$ 13.8 million (BDT 1140 million), and finding 50% of the currently undiagnosed 3.8 million diabetic adults aged 35+ years would cost about US$ 9.5 million (BDT 780 million). Patients cover a large part of the screening and diagnosis costs out of their own pockets. A single blood sugar test can cost a patient BDT 120.88 Test costs can be higher in urban areas than rural 16 NCD SERVICE PROVISION AND COSTS ACROSS THE CONTINUUM OF CARE settings.89 Total diagnosis costs to the patient, including the investigations, consultation fee and indirect costs, amount to an average costs of BDT 1700 (US$ 20).90 Treatment costs Hypertension: Using WHO data, it has been estimated that hypertension treatment would cost about US$ 13 (BDT 1,070) per patient per year.91 Diabetes: For Bangladesh, IDF estimates treatment costs of approximately US$ 51 (BDT 4,200) per patient per year.92 Costs can however be much higher depending on the treatment regimen (Annex 9 Fig. A 9.1 illustrates the wide range of costs from a systematic review of LIC/LMIC data). In one study, the average annual cost of diabetes care was estimated at US$ 25,900 (US$314, direct cost US$283 and indirect cost US$31).93 Treating the currently diagnosed 7.3 million hypertensive adults aged 35+ years translates into a cost of approximately US$ 95 million (BDT 7800 million). Treating the currently diagnosed 2.33 million diabetic adults aged 35+ years would cost approximately US$ 119 million (BDT 9860 million). The Bangladesh Copenhagen Project assessed the benefit of averting The Bangladesh NCD burden through targeted investment. It found hypertension Copenhagen Project medication highly beneficial at BDT 37 benefit for every BDT spent. assessed the benefit of Diabetes treatment also showed positive returns, albeit at a much averting NCD burden lower ratio, of BDT 3 per BDT spent. through targeted investment. It found Much of the long-term treatment and monitoring costs are borne by hypertension patients (e.g., BDT 2,486 annual payment per person for glucose testing strips). These high costs repeatedly lead patients to delay treatment medication highly seeking, prevent them from accessing comprehensive and continued beneficial at BDT 37 care, and severely hinder their ability to manage their illness benefit for every BDT effectively.94 Often, patients require several different medications. In spent. Diabetes one Bangladesh study, diabetic patients took on average 3.6 different treatment also showed medications.95 Figure 3.2 shows the medication patterns of diabetic positive returns... patients at the OPD of the Bangladesh Institute of Health Science Hospital. Complications can also quickly lead to high care costs. The average costs of retinopathy alone were reported to be US$140 per patient in Bangladesh in 2012.96 According to studies in India, up to 30% of people with newly diagnosed diabetes already had evidence of retinopathy, nephropathy, and CVD at the time of diagnosis.97 Diabetes is of course also associated with substantial indirect costs of lost productivity attributable to absenteeism and suboptimal work performance.98 17 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION Figure 3.2 Reported medication use by diabetes patients (in percentages, 2014 data) Source: Islam SMS et al. (2017). Healthcare use and expenditure for diabetes in Bangladesh. BMJ Global Health 2017;2:e000033. The medicines for hypertension and diabetes listed in the NCD Control Operational Plan are provided in Annex 10. Costs for diabetes care from a study comparing patients with and without diabetes are summarised in Box 3.5. Box 3.5 Average excess costs for people needing diabetes care Annual payments for medicines: BDT 35,385 (US$ 429) or 22 times higher than non- diabetic patients Annual out-patient care: BDT 5,075 (US$ 62) or 50% higher than non-diabetic patients Annual in-patient care: BDT 6,592 (US$80) or twice as high as non-diabetic patients Total annual point-of-service payments: BDT 49,538 (US$ 600) or six times higher than non-diabetic patients. This is also 27% of the HIES 2016-reported annual consumption expenditure average of BDT 185,000. Sources: Islam SMS et al. 2017, HIES 2016. For low-income groups, direct costs associated with diabetic care are therefore a dramatic drain on the annual household income (Box 3.5). This highlights the need to address financing and equity issues. Future treatment costs for hypertension and diabetes will further escalate with population growth and aging unless effective prevention and health promotion interventions are scaled-up. 18 NCD SERVICE PROVISION AND COSTS ACROSS THE CONTINUUM OF CARE Costs of preventive and health promotion interventions According to WHO,99 population-based measures that address tobacco and harmful alcohol use, as well as unhealthy diet and physical inactivity, account for a very small fraction of the total price tag of intervening against NCDs. Promoting public awareness about diet and physical activity costs approximately US$ 0.038 per In the Bangladesh person per year, interventions to reduce people’s salt intake cost about Copenhagen Project, US$ 0.030, and replacing trans-fat with polyunsaturated fat through smokeless tobacco population-level interventions cost about US$ 0.009. Figure A 9.2 in cessation and tobacco Annex 9 presents cost estimates for these three interventions in China, tax interventions India and the Russian Federation, which are higher than WHO’s global showed positive returns cost estimates but still regarded as “best buys” in these countries. In of BDT 9 and 8 per BDT the Bangladesh Copenhagen Project, smokeless tobacco cessation and spent, respectively. tobacco tax interventions showed positive returns of BDT 9 and 8 per BDT spent, respectively. The combination of population-based and individual-based best buy interventions for NCDs and their underlying risk factors amounts to an annual per capita investment of about US$1.50 in LMICs based on WHO estimates.100 Section 4 provides an overview of care models in Bangladesh focusing on diabetes and hypertension. It highlights some innovative and expanded models of care, both in Bangladesh and beyond. 19 “For low-income groups, direct costs associated with diabetic care are a dramatic drain on the annual household income. This highlights the need to address financing and equity issues.” NCD MODELS OF HEALTHCARE SERVICE DELIVERY 4.1 Overview of selected care models implemented in Bangladesh Available NCD services in Bangladesh are primarily concentrated in urban areas and provided by the private sector. In Dhaka, NCD service provision is for the most part through private for-profit and non-profit organisations, through general and specialised tertiary hospitals, such as The Bangladesh Institute of Research and Rehabilitation in Diabetes, Endocrine and Metabolic Disorders (BIRDEM)—the largest institute of the Bangladesh Diabetic Samiti (BADAS, previously known as Diabetic Association of Bangladesh)—and The National Heart Foundation of Bangladesh (NHFB). From the public sector, the National Institute of Cardiovascular Disease, a tertiary hospital that provides cardiovascular specialist care, has an outpatient department, including a hypertension clinic. Moving outside Dhaka, the Ministry of Health and Family Welfare (MoHFW) is scaling up ‘NCD corners’, which were initially piloted in three Upazila (sub-district) UZHCs in 2012 and are now planned to be available in 300/491 UZHCs; there is one UZHC in each Upazila. A NCD service capacity assessment carried out in an UZHC with a newly established NCD corner reported that the “NCD corner was not able to function adequately” due to stock-outs of a substantial number of essential NCD medications and inadequate record taking of consultations and referrals among others101. Box 4.1 Key findings from the 2014 Bangladesh Health Facility Survey The Bangladesh Health Facility Survey of 2014, funded by the Government of Bangladesh and USAID, assessed the health care system capacity to appropriately diagnose and treat NCDs.102 The survey reported that less than one in five (18%) health facilities offer services for diabetes, increasing to around one in four (26%) if excluding the community clinics. For CVD diagnosis, care and treatment services, this is provided by 16% of all health facilities and 28% of facilities if excluding community clinics. Unsurprisingly, higher level facilities are more likely to provide diabetes or CVD services with almost all district hospitals (95%) and 81% of UZHCs providing such services. In addition, more than three quarters of facilities do not have guidelines for the diagnosis and management of diabetes and CVDs and 80% or more (depending on condition) of health facilities don’t have any staff member recently trained (defined as in the last two years) to provide services for either diabetes or CVDs. The majority of facilities are ill-prepared to provide quality services for diabetes or CVDs and, in particular, lack the essential medicines for treating diabetes and CVDs, especially in rural areas (see Figure 4.1). 21 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION Figure 4.1 Percentage of health facilities providing diabetes and CVD services and respective medications in rural and urban Bangladesh Sources: National Institute of Population Research and Training (NIPORT), Associates for Community and Population Research (ACPR), and ICF International. 2016; Bangladesh Health Facility Survey 2014. Dhaka, Bangladesh: NIPORT, ACPR, and ICF International. Despite these shortfalls in NCD service delivery, a NCD-related PHC level situation assessment conducted in 2015 confirmed the potential that is in the country to address the NCD burden.103 For instance, the human resources including doctors, nurses, Despite shortfalls in NCD technologists and field level workers, are present (albeit not at service delivery, a NCD- sufficient levels). It however also demonstrated that detection related PHC level situation and referral of cases from the community level by health assessment conducted in 2015 workers is weak and training limited.104 confirmed the potential that is in the country to address the To help improve this situation, the Ministry of Health is piloting, through its NCDC project 2017–22, a community-based, primary NCD burden. healthcare model that utilises the country’s CCs. This model of care is based on WHO’s Package for Essential NCD (PEN, Box 4.2) and includes screening of NCDs at the CC, a referral to the UZHC for screen positive patients, diagnosis and treatment initiation at the UZHC, and treatment maintenance back at the CC. The aim is to reach 20 (200) UZHCs and 200 (2000) CCs by 2020 (2022). It would allow further involvement of the community level and non-physicians to screen for risk factors for CVD and participate in managing uncomplicated NCDs according to established protocols, under the supervision of medical officers. In some rural areas NCD services are also provided by small community-based intervention pilots. Table 4.1 provides an overview of hypertension and diabetes care models in Bangladesh by care cascade stage. 22 NCD MODELS OF HEALTHCARE SERVICE DELIVERY Box 4.2 WHO’s Package for Essential NCDs (PEN) WHO’s Package for Essential NCDs (PEN) is a conceptual framework that aims to strengthen equity and efficiency of primary health care in low-resource settings. The package includes a prioritised list of cost-effective interventions that can be applied to an acceptable quality even in resource-poor settings. Furthermore, the package includes core medicines and technologies, risk prediction tools, tools for planning, implementation and assessment of capacity, and guidance and protocols on how interventions should be implemented. It also addresses technical and operational issues on how the essential NCD interventions can be integrated into primary care and evaluated for impact. According to the WHO, PEN “should not be considered as yet another package of basic services” but, rather, an important "first step for integration of NCD into PHC and for reforms that need to cut across the established boundaries of the building blocks of national health systems”. WHO’s PEN is the minimum standard for NCDs to strengthen national capacity to integrate and scale up care of heart disease, stroke, cardiovascular risk, diabetes, cancer, asthma and chronic obstructive pulmonary disease in primary health care in low-resource settings. The package has already been implemented in a number of countries, including Bangladesh, Sri Lanka, Bahrain, Kuwait, Indonesia, Palestine, Malawi, Bhutan, the Philippines and the Democratic People’s Republic of Korea, among others. 4.2 Principles and examples of NCD and chronic care models implemented elsewhere The response to major infectious diseases with long-term treatment needs (such as HIV and TB) has provided some important lessons for the long-term management of chronic diseases. While there are certainly lessons to be learned from NCD management in high-income settings, the rapid scale-up of health services in resource-constrained settings where health systems are generally much weaker than high-income settings and there are critical shortages of health professionals, requires a “public health approach”. In HIV, this approach, described by WHO in 2003105, prioritised large-scale access to treatment over maximising individualised care, the usual hallmark of chronic disease care in high-income settings. The main principles of the successful public health approach in generalised HIV epidemics, particularly in Sub-Saharan Africa, are simplification and standardisation of healthcare delivery processes, decentralisation, task-shifting, peer and community involvement in programming and support for self- management.106 Some of these elements have been applied to other chronic diseases - including NCDs - in Bangladesh and other resource-constrained settings, through innovative service delivery models of care. 23 24 Table 4.1 Highlighted hypertension and diabetes care models in Bangladesh across the care cascade HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION INSTITUTE/ FUNDING; 1 TARGET GENERAL MODEL/ PATIENT SCREENING/ ADHERENCE/ PROJECT TARGET POP DISEASE SCALE PAYMENT DIAGNOSIS LINKAGE/REFERRAL TREATMENT MONITORING OTHER NCDC Government; HTN & Community-based “Free/subsidised Screen in CC by Those with +ve screen or Initiated in Unclear Future plan is to project of rural DM PHC model. No treatment for trained CHW. complications referred up to UZHC and use multi- DGHS baseline on current NCDS at least for Some home UZHC for diagnosis & refill planned purpose scale (new the poor screening in treatment (no formal to be volunteer/ programme). (identified) will be selected pilot referral/ follow up currently). available workers to work Programme target= provided.” areas. CC to be However, the country is from CC in communities 20 (200) UZHC and provided BP working on establishing “a to motivate 200 (2000) CC by machine, strong referral linkage with people to 2020 (2022) glucometer, urine feedback…from CC to UZHC attend sugar strip. and further referral to CC/UZHC for Diagnosis by MO DH/MCH/ Specialised screening, at UZHC, which institutes” diagnosis & are to be management of provided X-ray, NCD as ECG, glucometer, appropriate. BP machine, measuring tape, weighing scale NCD corner Government; HTN & Community-based As above At NCD corner in If complications, advise At NCD Unclear rural DM PHC model, based on UZHC patients to go to district corner in PEN intervention. 1st hospital or above. Currently, UZHC. piloted in 2012 in 3 no formal referral/ follow up. UZHCs in the SW Patients who attend UZHC districts of Khulna are often not triaged at Division. Now scaled- UZHC and thus queue to see up plan in 300 UZHC a GP who subsequently in country refers patient to NCD corner, also manned by a GP. Table 4.1 Highlighted hypertension and diabetes care models in Bangladesh across the care cascade (continued) INSTITUTE/ FUNDING; 1 TARGET GENERAL MODEL/ PATIENT SCREENING/ ADHERENCE/ PROJECT TARGET POP DISEASE SCALE PAYMENT DIAGNOSIS LINKAGE/REFERRAL TREATMENT MONITORING OTHER BRAC NGO; rural HTN & Community-based Unclear Screen at home Screen +ve patients at home At hospital by At regular home visit by DM model. Pilot in 11 by trained CHW; referred for diagnosis and MO CHW. Importance of a sub-districts in 7 diagnosis in management in hospital. healthy lifestyle emphasised. districts hospital by MO Unclear exact referral mechanism COBRA-BPS Research HTN Home community- Subsidies, in the Screen at home Stepped-up referral from the By trained Adherence to lifestyle advice Currently being institute; rural based model utilising form of a travel by CHW. CHW to a trained GP using a public and through HHE. Follow-up at evaluated for CHW. Currently voucher for clinic Diagnosis checklist; HTN care private home by CHW impact as part piloted in some areas visit and a free confirmed by coordinators facilitate the providers in of a RCT, of Tangail and supply of antiHTN trained GP in tracking of referrals in HTN triage including in Munshiganj Districts medications, designated HTN government clinics counters Pakistan and Sri provided to low triage counter in using HTN Lanka. income health facility checklist individuals with poorly controlled HTN. BIRDEM Private, non- All, but 700-bed GH, largest 30% of the On-site and Patients referred up to & On-site by On site DM & nutrition The hospital profit; urban DM and institute of BADAS. hospital’s beds during diabetes down from BIRDEM, MO. education available. All has the largest NCD MODELS OF HEALTHCARE SERVICE DELIVERY endocrin All expected GH are free, including camps & street particularly from/to affiliated Pharmacy patients have DM book. OP turnover in e services, including OP investigation & screening associations. This is usually also available the world disorder and IP services, treatment, for initiatives in done in form of on-site. under a single s priority providing poorest patients Dhaka during DM recommending patient to roof comprehensive special days attend without formal primary, secondary referral & follow up and tertiary care. 25 26 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION Table 4.1 Highlighted hypertension and diabetes care models in Bangladesh across the care cascade (continued) INSTITUTE/ FUNDING; 1 TARGET GENERAL PATIENT SCREENING/ ADHERENCE/ PROJECT TARGET POP DISEASE MODEL/SCALE PAYMENT DIAGNOSIS LINKAGE/REFERRAL TREATMENT MONITORING OTHER The Private; urban HTN & Subscription-based Monthly Unclear DM patients with SMS Patients able to transfer data The programme Telemedicin DM membership subscription fees complications/ in-need of in- prescriptions from home monitoring was able to e Reference organisation that range from patient services are referred provided devices (glucometers & BP increase client Centre links DM patients US$0.60 to to BADAS. Unclear for HTN. monitors). Home visits by adherence to Limited with healthcare US$20.00. qualified providers also treatment plans (TRCL)/ providers (qualified available (US$10–20/ by 62% and AMCARE doctors & nurses) month). Patients also receive reduce doctor/ through a call centre. relevant information & hospital visits medical alerts via SMS. from 5–6 per Patients have direct access year to 1–2 per to their patient file. year. National Private, non- CVD, 300-bed specialist 30% of the On-site and Patients referred up to & After acute Heart profit; urban including hospital, provides hospital’s beds during free down from hospital, hospital Foundation HTN invasive and non- are free for hypertension particularly from/to affiliated episode and Hospital and invasive cardiac poorest patients camps associations. This is usually in HTN clinic Research diagnostics and done in form of Institute investigations. recommending patient to (part of Includes OP, ED & IP attend without formal NHFB) specialist services, referral & follow up including a cardiac surgery unit. In 2015, the provided services for 88343 OP and 21704 IP. Sources: Key informant interviews and site visits; 4th HPNSP NCDC Operational Plan 2017–22; BADAS website and Annual report 2016; NHF website and Annual Report 2016 and 2015; Bangladesh Health Watch Report 2016; World Bank NCD report 2013; Jafar et al, 2017. Multicomponent intervention versus usual care for management of hypertension in rural Bangladesh, Pakistan and Sri Lanka: study protocol for a cluster randomized controlled trial. 18:272. DOI 10.1186/s13063-017-2018-0. BADAS The Diabetic Association of Bangladesh; BIRDEM Bangladesh Institute of Research & Rehabilitation in Diabetes, Endocrine and Metabolic Disorders; BP blood pressure; CC Community Clinic; CHW community health worker; COBRA-BPS Control of Blood Pressure and Risk Attenuation-rural Bangladesh, Pakistan, Sri Lanka, Feasibility Study; CVD cardiovascular disease; DGHS Directorate General of Health Services; DH District Hospital; DM type 2 diabetes mellitus; ECG electrocardiogram; ED Emergency Department; GH General Hospital; GP General Practitioner; HTN hypertension; IP inpatient; MCH Maternal and Child Health; MO Medical Officer; NCD Non-communicable diseases; NCDC Non-Communicable Disease Control; NGO non-governmental organisation; NHFB National Heart Foundation Bangladesh; OP outpatient; PEN Package for Essential NCDs; PHC primary health care; RCT Randomised controlled trial; UZHC Upazila Health Complex. NCD MODELS OF HEALTHCARE SERVICE DELIVERY The following section briefly describes the underlying principles of the aforementioned strategies and provides some examples of their use in chronic diseases, primarily focusing on NCDs in other low- and middle-income countries. As will be highlighted, these principles can be complimentary and NCD programmes will often use more than one of these principles in their programme design. At the same time, an approach can be used across the different principles. For instance, mobile health or mHealth solutions relate to the principles of self-monitoring, In Bangladesh, decentralisation and task shifting. Interventions using mHealth standardisation, technology can enhance or replace the in-person, face-to-face and decentralisation and group-based service delivery, and has been found to improve limited task-shifting outcomes, be cost effective, and culturally relevant. According to (primary prevention Mallow et al. (2015), "mHealth technology has been used to improve and screening) and self- outcomes include seeking out health information via the web, access to management has been appointment scheduling and medication refills, secure messaging, the predominant computerised interventions to manage a chronic condition, use of a strategy. personal health record, use of remote monitoring devices, and seeking support from others with similar health concerns through social networks".107Evidence on the potential of mHealth applications especially in diabetes care is accumulating rapidly, for instance: A telephone-delivered intervention after discharge from diabetes rehabilitation in Germany demonstrated improvements in patients' level of activity and health status after 12- months.108 A two-way text messaging program in Utah, USA that provided behavioral coaching, education, and testing reminders to diabetes patients who were failing to achieve their self- management goals, saw high satisfaction among patients and frequent use in a sub-set of them.109 mHealth interventions may however not always be superior to traditional approaches. For instance, mHealth mobile phone messaging and participatory community mobilisation were both tested in a large trial involving 96 villages in Bangladesh. The mHealth intervention increased knowledge and awareness of T2DM and its risk factors but had no detectable impact on disease outcomes. In contrast, the community mobilisation using a participatory learning and action approach not only increased knowledge and awareness of disease, but also significantly reduced population prevalence of T2DM and intermediate hyperglycaemia, and T2DM incidence among an intermediate hyperglycaemic cohort.110 In Bangladesh, standardisation, decentralisation and limited task-shifting (primary prevention and screening) and self-management has been the predominant strategy (Table 4.1 highlights the key components of some of the models). In upcoming sections, we provide examples of models that have used a more extensive task-shifting strategy as well as peer support, community involvement and self-management strategies. Simplification, standardisation and monitoring of healthcare delivery processes Simplifying healthcare management approaches aim to strike a balance between what is desirable and what is currently feasible in resource-limited settings.111 It allows a wider range of providers to be involved in patient management while minimising risks to patients. They can also help patients directly with self-management and treatment adherence (for example if treatment regimen is simplified). Standardisation is “the process of developing, agreeing upon and implementing uniform technical specifications, criteria, methods, processes, designs or 27 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION practices that can increase compatibility, interoperability, safety, repeatability and quality112.” Simplifying decision-making and defining clear and simple standards based on evidence-based interventions, including standard operating procedures (SOPs), help maintain patient safety and high-quality healthcare service provision. One of the key objectives of WHO’s PEN Interventions (Box 4.2) is “standardisation of diagnostic and investigation procedures and drug prescription” as well as “formulation of referral criteria for further assessment or hospitalisation”113. Relevance for Bangladesh: While there are currently no national standard operating procedures for the management of NCD in Bangladesh114, the country is currently in the process of developing these with the support of WHO. These will be based on simplified algorithms and the use of standard drug prescriptions based on generic formulations. Decentralisation including “Care closer to home” In health service delivery decentralisation encompasses the transfer of service delivery from higher levels, such as hospitals, to lower levels of the healthcare system, which are often closer to patients (“Care closer to home”), saving patients time and money and allowing individuals living in rural and remote locations to access care. A 2017 systematic review of primary care models for NCD interventions in Sub-Saharan Africa found that 10/12 studies incorporated the use of rural locations for implementation of their clinics.115 Box 4.3 provides a case study from Sri Lanka, which implemented a decentralised PHC NCD model. This model is similar to Bangladesh’s new NCDC care model (Table 4.2 provides an overview of Bangladesh’s model by tier of health care and cascade stage while Figure A 11.1 Annex 11 provides infographics on the NCDC care model in urban and rural settings of Bangladesh). Box 4.3 Decentralising primary NCD care: Sri Lanka’s Healthy Lifestyle Centres The development of Sri Lanka’s Healthy Lifestyle Centres (HLCs) was based on a decentralised primary care model using the lessons learnt from a WHO PEN intervention pilot, as well as two other pilots.116 The HLCs were initiated in 2011 to address the lack of NCD screening in Sri Lanka’s primary health care. They were set up within primary medical care institutions (PMCUs)—the lowest level of curative institutions offering only outpatient services provided by one medical officer and one health assistant and/or a dispenser—and the Healthy Life Centre was expected to be open on at least one weekday morning, serving 20 adults aged between 40 and 65 years per session. Services are provided in the HLCs using standardised protocols for NCD and risk factor screening, lifestyle management of screen positive patients, referral to the specialised medical clinics conducted at the same PMCU or another institution for those with a 10-year CVD risk higher than 30%, and group health education. Implementation of the Sri Lanka programme highlights a few lessons that can be drawn. Firstly, the importance of carrying out monitoring and using that data for programming, including to redesign the programme as needed, is highlighted. The project noted low participation in some areas and of men overall in the programme, which led the country to expand the opening hours of the HLC and consider expanding it further to include weekends and public holidays. The country also initiated “outreach” screening in geographical locations where participation was low and in workplace settings to target those previously not reached by the programme. Secondly, to improve health system capacity for NCD provision, the country is considering the development of a separate community-level cadre dedicated to NCD-related activities that would work in their communities to encourage people to attend for screening, follow-up NCD patients in the community, and provide ongoing health and lifestyle education to the population. 28 NCD MODELS OF HEALTHCARE SERVICE DELIVERY Table 4.2 Bangladesh’s new model for the continuum of care for hypertension and type-2 diabetes PRIMARY SCREENING TREATMENT TREATMENT PROVIDER PREVENTION (BP, NCD RISK) DIAGNOSIS INITIATION MONITORING Outreach Education on Clinical screening Patient self- risks & of CVD risk factors monitoring healthy life including BG & BP style Partial determination of 10-year CVD risk & refer Community Education on Clinical screening No but policy is for refills Patient self- Clinic risks & of CVD risk factors to be obtained once monitoring healthy life including BG & BP patients are initiated on Regular BP & BG style Partial treatment check at facility determination of (unclear frequency) 10-year CVD risk & refer Union Education on Clinical screening Both HTN Yes, by doctor (available Patient self- risks & healthy of CVD risk factors (serial BP) & only in few unions), using monitoring life style including BG & BP DM2 (FPG/ both biomedical (generic Partial OGTT) drug or insulin determination of formulations) and lifestyle 10-year CVD risk & advice. refer Manage conditions and I&R complications Upazila Education on Clinical screening Both HTN Yes, by doctor, using both Screening, Health risks & healthy of CVD risk factors (serial BP) & biomedical (generic drug identification & Complex life style including BG & BP DM2 (FPG/ or insulin formulations) referral/ Laboratory OGTT) and lifestyle advice. management of screening complications Full determination Laboratory follow-up of determination of of HTN and DM 10-year CVD risk cases Manage conditions Down-referral to CC and I&R once disease complications controlled District Education on Clinical screening Both HTN Yes, by doctor, using both Screening, Hospital risks & healthy of CVD risk factors (serial BP) & biomedical (generic drug identification & life style including BG & BP DM2 (FPG/ or insulin formulations) management or up- Laboratory OGTT) and lifestyle advice. referral of screening complications Full determination Laboratory follow-up of determination of of HTN and DM 10-year CVD risk cases Manage conditions Down-referral to CC and I&R once disease complications controlled 29 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION Table 4.2 Bangladesh’s new model for the continuum of care for hypertension and type-2 diabetes (continued) PRIMARY SCREENING TREATMENT TREATMENT PROVIDER PREVENTION (BP, NCD RISK) DIAGNOSIS INITIATION MONITORING Urban: Education on Screening of CVD Both HTN Yes, by doctor, using both Screening, CRHCC risks & healthy risk factors (serial BP) & biomedical (generic drug identification & life style including BG & BP DM2 (FPG/ or insulin formulations) management or up- (unclear if includes OGTT) and lifestyle advice. referral of laboratory or complications clinical only) Laboratory follow-up Full determination of HTN and DM of determination of cases 10-year CVD risk Manage conditions and I&R complications Urban: Education on Screening of CVD Both HTN Yes, by doctor, using both Screening, PHCC risks & healthy risk factors (serial BP) & biomedical (generic drug identification & life style including BG & BP DM2 (FPG/ or insulin formulations) management or up- (unclear if includes OGTT) and lifestyle advice. referral of laboratory or complications clinical only) Laboratory follow-up Full determination of HTN and DM of determination of cases 10-year CVD risk Manage conditions and I&R complications Private Yes, in selected Yes, more so if Yes, more so if doctor Yes, in selected pharmacies pharmacies upon doctor associated with pharmacy pharmacies upon client request associated patient request with pharmacy Sources: NCDC Operational Plan 2017‒22; In-country observations and key informant interviews. Notes: BG=blood glucose; BP=blood pressure; CC=community clinic; CVD=cardiovascular disease; CRHCC= Comprehensive Reproductive Health Care Centre; DM2=diabetes mellitus type 2; FPG=fasting plasma glucose; HTN= hypertension; I&R=Identify and refer; NCD=non-communicable disease; OGTT=Oral Glucose Tolerance Test; PHCC= Primary Health Care Clinic. Relevance for Bangladesh: The new NCDC care model is an attempt to decentralise NCD service delivery, based on a decentralised primary care model whereby the first point of contact for NCD patients with the healthcare system is at the CC, the lowest level tier of the healthcare system. In addition, many healthcare providers, such as BIRDEM, organise community outreach or health ‘camps’ whereby health services are brought to the people. The country is also introducing a new cadre of community healthcare staff, the Multipurpose Health Volunteers, who would be able to carry out education, screening and follow-up of patients in the community itself with referral of screen positive patients to healthcare facilities. The approach used is similar to Sri Lanka’s experience and some lessons that can be drawn for Bangladesh have been highlighted above. Task-shifting and involvement of a full range of healthcare providers In a context of shortages and/or unequal distribution of healthcare staff in low- and middle- income countries, many countries have introduced task-shifting strategies for a variety of healthcare services.117 Task-shifting refers to the reallocation or shifting of tasks from the 30 NCD MODELS OF HEALTHCARE SERVICE DELIVERY responsibility of one cadre (a qualified person such as a doctor) to another, usually lower, cadre (such as a nurse or even a person with no health qualification). This practice has been particularly prevalent and successful in MCH,118 and, as previously mentioned, in the context of HIV/AIDS, where the impressive scale up of antiretroviral therapy coverage among people living with HIV can at least be partly attributed to task-shifting of treatment initiation to nurses.119 Training, the use of clearly defined SOPs and appropriate supervision and mentorship helps maintain patient safety and service quality. In recent years, task-shifting approaches have expanded to NCDs in low- and middle-income countries. A 2014 systematic review assessing the effectiveness, cost-effectiveness and barriers of task-shifting to non-physician healthcare workers for the management and prevention of NCDs in LMICs identified 22 published studies, including seven studies for hypertension and CDVs and five studies for diabetes120. The systematic review found that task-shifting was effective in improving process outcomes (such as successfully screening for hypertension and diabetes) and health outcomes (such as achieving hypertension and diabetes control) and achieving treatment concordance with doctors. Two studies, both conducted in India, reported on the cost-effectiveness of task-shifting strategies for oral cancer and anxiety and depression, finding the intervention to be cost-effective and cost-saving respectively. Task-shifting enablers were appropriate training and provision of standard algorithms and protocols while the main identified barriers were linked to legislative prescribing restrictions and availability of medicines. Boxes 4.4, 4.5 and 4.6 provide task-shifting case studies from Kenya, Rwanda and Indonesia respectively. Box 4.4 Task shifting for patient management and medication adherence clubs: Kenya Five years after launching a NCD programme in Kibera slum in Nairobi, Kenya, Médecins Sans Frontiéres—in response to an overwhelming NCD patient volume load for clinical officers— began to shift the management of stable patients with five NCDs to nurses, as an extension of its task-shifting approach already used for HIV and TB121. The nurses underwent a week-long training on the five NCDs (hypertension, diabetes, epilepsy, sickle cell disease and asthma) as well as how to use structured clinical decision support protocols. A retrospective review of routinely collected clinic data from the two PHC facilities carried out in 2014 found that nurses’ adherence to NCD protocols and process indicators was 69% for routine screening questions, 81% for routine laboratory monitoring and >90% for weight checks, BP monitoring and review of laboratory results. Seventeen (2%) of consultations were referred back to clinical officers for uncontrolled disease, medication side effects, and NCD-associated complications. The programme also used medication adherence clubs for stable NCD patients, who are able to collect their chronic medication every three months through a club, rather than through individual clinic appointments122. These clubs also provide an opportunity for monitoring, health education and peer support. Furthermore, by having the clubs at different times (including Saturday) and avoiding the need to queue at the clinic, the saved opportunity cost for patients can be considerable. 31 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION Box 4.5 Nurse-led NCD care: Rwanda’s integrated model for rural settings In Rwanda, an integrated, nurse-led NCD care model implemented in rural areas has been providing comprehensive care services for a wide range of NCDs including heart failure, chronic cancer pain, hypertension, diabetes, and chronic respiratory diseases123. Services are led by nurses and are provided on disease specific clinic days. Training, on-going mentorship from specialist doctors and standardised diagnosis and treatment protocols are provided to nurses. Nurses working in health centres are also provided mentorship from district hospital nurses. The nurses use point-of-care diagnostics including HbA1c and are able to perform simplified echocardiography to inform initial management of heart failure. Standardised forms are used to record clinical information and kept in individual patient files that are later captured in an electronic medical records system. Patients are also offered group education sessions and socioeconomic support and those identified as high-risk are provided support from community health workers. By Deptember 30, 2014, 3367 patients had been enrolled in three district hospitals and seven health centres with 30% and 16% having hypertension and diabetes respectively. The outcomes assessments for hypertension and diabetes are ongoing. Box 4.6 Village health workers for community awareness and screening: Indonesia Indonesia implemented a primary care level diabetes and hypertension screening and treatment programme based on the WHO PEN programme in 2011124. The country used trained village health care volunteers to conduct ‘Posbindu’ (community engagement, community-based awareness, monitoring, and screening activities) for diabetes and hypertension, in various community areas including the homes of the volunteers and designated Posbindu facilities. Posbindu activities are carried out on the same day monthly whereby the village volunteers carry out general health history check-ups, take weight and height measurements, and screen for hypertension and diabetes mellitus using a BP monitor and a glucometer for random capillary blood glucose for any attending person aged 15 years and older. Persons who are screened positive are referred to PHC facilities where doctors or nurses confirm the diagnosis and provide treatment. Individuals may also present directly to the PHC facility for screening and diagnosis. An economic analysis of the community screening strategy found that the programme averts 0.04 DALYs per individual screened for diabetes or hypertension and is cost-saving (14.22 million Indonesian rupiah savings for every DALY averted, around US$ 1052). The study also reported that a more targeted screening policy (screening of those aged 40 years and older as opposed to screening for 15 years old and above as is the current practice) and the use of a more accurate screening test for diabetes (fasting versus random capillary glucose) would be even more cost-effective. Relevance for Bangladesh: The new NCDC care model is focused on task-shifting of NCD screening to community healthcare workers working in CCs. Diagnosis and treatment initiation remains the responsibility of physicians. There is a critical shortage of nurses in Bangladesh and task-shifting to nurses would not be feasible. Community healthcare workers are also already overstretched. The country may consider further task-shifting of screening to trained lay workers or volunteers, drug sellers or pharmacists or informal village doctors as well as expanding the prescribing list of community healthcare workers if feasible. Peer and community involvement in service delivery Peer and community involvement is critical for large-scale roll out of healthcare programmes.125 At its strongest and most transformative, community participation encompasses empowerment of the community the programme serves, where the community is involved in considering options, making decisions, and taking collective action.126 Community participation can serve to 32 NCD MODELS OF HEALTHCARE SERVICE DELIVERY raise awareness, design appropriate interventions and thus increase acceptance, advocate for political action, provide informal and formal support to patients, mobilise resources, and support sustainability of programmes. Peer support care models, where peers are used to provide social and emotional support, assistance with disease management, and linkage to clinical care and community resources, have been widely applied in health and are starting to gain wider momentum in NCD care. Boxes 4.7 and 4.8 provide case studies from NCD peer support care models in Cambodia, Cameroon, South Africa, Thailand and Uganda. Box 4.7 Peer support care models for diabetes management Peers for Progress, a peer support programme for diabetes management of the American Academy of Family Physicians Foundation, was implemented in Cameroon, South Africa, Thailand and Uganda127. Each country team developed its own programme based on the local context. However, the programmes shared the same general approach with peers providing: 1) assistance in daily self-management; 2) social and emotional support; 3) linkage to clinical care; and 4) ongoing support. Depending on country context, the support involved individual and group meetings, text messaging, home visits, exercise sessions, cooking classes, and accompanying patients to clinical care as required. Peers were based in diabetes clinics (Cameroon and Uganda), non-health community organisations (South Africa), and a health volunteer service (Thailand). In Cameroon, the programme observed significantly decreases in participants’ average body mass index (from 28.6 to 25.5 kg/m 2), systolic (from 142 to 124.4 mmHg) and diastolic (from 84.4 to 77.7 mmHg) blood pressure, and HbA1c (from 9.6% to 6.7%) in only six months. Before the project the number of diabetics with a HbA1c greater than 8% was 85/96; six months later only 4 had a HbA1c greater than 8% and 65 had a HbA1c less than 7%. In Thailand, evaluation data were available for fifty-three adults with type 2 diabetes; after six months the study reported a significant decline in average HbA1c (from 8.6% to 7.9%) and a borderline significant decline in body mass index (from 26.7 to 25.2, p=0.07). In Uganda, the project also observed a decline in participants’ average diastolic blood pressure (from 85.4 to 76.3 mmHg) and average HbA1c (from 11.1% to 8.3%) while the number of diabetics with HbA1c values less than 7% almost doubled (from 17% to 32%). While the evaluation results are impressive, it must be noted that sample sizes were relatively small. 33 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION Box 4.8 Peer educators for diabetes screening, education and patient navigation: Cambodia The MoPoTsyo Patient Information Centre, a Cambodian NGO, has been implementing an innovative care model for diabetes using peer educators.128 129 The NGO created networks of ‘informed’ and ‘empowered’ patients organised around a team of peer educators that also have diabetes. To become a peer educator, a person obtains six weeks of training, after which they receive basic equipment and supplies and are able to use their homes as a ‘Patient Information Centre’ for weekly patient meetings and education sessions. Peer educators screen adult patients in the community using urine strips and those with a positive urine screen are further screened using a glucometer. Newly identified diabetic patients receive classes at the peer educator home with an emphasis on self-monitoring of urine glucose levels (strips are provided monthly) and lifestyle adaptation, including better nutrition and daily exercise. The NGO also supplies medication to its networks through a Revolving Drug Fund whereby the NGO procures a set of diabetes and hypertension medicines on the international market to sell to patients at a fixed price via contracted pharmacies. Medical consultations and laboratory services are provided by a diabetes specialist contracted by the NGO in district hospitals twice a month. During the consultation days, the peer educator is responsible for crowd control and registers patients and takes blood pressure, blood glucose and weight measurements. Other services, such as retinopathy screening, are organised via external private, non-profit clinics where a modest fee applies for those not eligible for a Health Equity Fund. The role of the peer educator here is to guide the patients through the available services including when and where to access them. Peer educator travel costs are reimbursed and, for each activity (screening, monitoring and education sessions), an incentive is provided. The peer educator is also eligible for a bonus based upon the evaluation of their work through the NGO’s monitoring and evaluation system. Results after 18-months of programme implementation showed that peer educators had reached more than 80% of adults in the pilot district with 474 confirmed as having diabetes. Of these, 67% were newly diagnosed. Almost 90% of patients reported having increased their activity level after the programme and 84% had reduced the amount of white rice in their diet. A review of patient self-management books showed that 71% had recorded more than two urine glucose results in the past month (this average was depressed by one new peer educator area where rates were only 5% compared to more than 80% in the remaining seven areas). The study also documented short-term improvements in blood glucose and BP control as well as low referrals for medical consultations, expenditures and drop-out rates.130 Relevance for Bangladesh: Considering the huge burden of NCDs and the critical shortages of healthcare staff in rural areas, there is an opportunity to tap into the community itself to help with NCD programming. This could include using trained community volunteers or peers to help with NCD prevention, screening and client navigation through the health system. Self-management Due to the large burden of NCDs and its chronicity, requiring lifelong management, experts have argued that there is a strong need to educate and support patients towards full self-management of their chronic condition131. The term self-management in relation to diseases was first coined by Thomas Creer who described self-management as the active participation of patients in their own treatments132. While there is a wide variety of definitions and conceptualisations for disease self-management today the term more broadly refers to the day-to-day management of chronic conditions by individuals over the course of an illness133. 34 NCD MODELS OF HEALTHCARE SERVICE DELIVERY …self-management of chronic Self-management is a key component of the Chronic Care Model (CCM)—perhaps the most well-known and influential chronic diseases can be challenging disease management model134. Interventions to improve self- for patients, as it requires management can include education, peer support, identification daily and lifelong changes to of barriers to adherence to treatment (lifestyle and/or lifestyle, and therefore biomedical), problem-solving to overcome barriers, provision of requires tailored education self-monitoring tools, reminders, goal setting and incentives. In (both content, delivery and one of its most basic forms, the provision of educational material frequency) and additional and such as leaflets to patients that describe their disease and ongoing support. management can already decrease information asymmetry between provider and patient and help the patient on the path to empowerment and self- management of his or her disease. However, self-management of chronic diseases can be challenging for patients, as it requires daily and lifelong changes to lifestyle, and therefore requires tailored education (both content, delivery and frequency) and additional and ongoing support. In Bangladesh, for example, patients with type 2 diabetes who had received professional health education and care in an urban tertiary health facility were still found to have limited knowledge of the risk factors, causes, and management of their condition135. There is a wide literature on the use and effectiveness of self-management strategies in NCDs, particularly from high-income countries. In 1996, the Center for the Advancement of Health reported that “there is substantial evidence (from over 400 studies of self-management) that programmes providing counselling, education, information feedback, and other supports to patients with chronic conditions are associated with improved outcomes136.” Evaluation evidence of the self-management component of the Chronic Care Model mentioned above points to three key interventions that are effective: distribution of educational material; patient motivational counselling; and patient educational sessions137.One simple tool that patients can be given to support their empowerment and self-management is a booklet that includes tailored and appropriate information on their condition and treatment, their care plan, and specific patient goals. The Chronic Care Passport, which has been successfully piloted in 13 countries in Latin America and the Caribbean, is one such example.138 The passport, a patient held card used by patients with chronic conditions including diabetes and hypertension, promotes integration between different levels of care and includes a care plan to be followed by providers and clients along with goals and meal plans. Preliminary results from the demonstration projects have found that that the use of these passports is associated with better diabetes (% of patients with HbA1c<7% increased from 32% to 40%) and hypertension control (from 45% to 49%). It has also been associated with better quality service provision; there was a reported increase in the proportion of patients who had received nutritional advice (12% to 52%), a foot exam (28% to 68%) or an eye exam (21% to 61%) while the proportion of patients meeting 3 or more quality of care measures, increased from 12% to 56%. A similar patient-held ‘passport’ for diabetes has been used in other contexts for NCD management and was also observed to be in use in the BIRDEM institute in Dhaka. 35 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION A 2017 systematic review of peer-reviewed articles assessing the effectiveness of self- management interventions for type 2 diabetes identified 14 studies with a wide-range of self- management interventions139. While all had some form of education … people with type 2 intervention involved, how these were delivered, by who, the material diabetes who used, duration and frequency, and their content differed. Overall, however, the review found that people with type 2 diabetes who participate in self- participated in self-management interventions improved their management knowledge of diabetes (3 studies; all found positive statistically interventions improve significant results), blood glucose control (8/9 studies reporting their knowledge of varying improvements in HbA1c levels, ranging from 0.7% to 1.7%), diabetes…, blood BMI (2 studies both showing significant improvements), lipid profile glucose control…, BMI..., (3/4 studies reporting statistically significant improvements) and lipid profile… and quality of life (all 3 studies reporting on quality of life reporting quality of life… significant improvement). Relevance for Bangladesh: Despite the high prevalence of NCDs in Bangladesh, NCD knowledge among Bangladeshis is still relatively poor. Self-management of NCDs, in particular diabetes, can be complex and challenging for patients. The country may consider additional approaches to support patients and their families to manage and control their disease, for example using different, culturally sensitive education material, monitoring tools and peer support, among others. Approaches should be evaluated to ensure that they are appropriate for the population served. Comprehensive change towards delivery system redesign The above five principles are often combined when health services get reoriented in response to changing health care needs. One model which speaks to all the principles is the Integrated Chronic Disease Management (ICDM) model140, however, the model represents a more wide- ranging re-design of service delivery or a “diagonal health system strengthening” intervention. Box 4.9 Delivery system redesign: Integrated Chronic Disease Management in South Africa In South Africa, an integrated approach to the management of chronic diseases derived from Wagner’s Chronic Care Model141 and WHO’s improved care for chronic conditions framework142 was developed and modified according to local conditions.143 The main aim was to attain efficient, integrated services for chronic communicable (HIV, TB) and chronic non- communicable (hypertension, diabetes, COPD, asthma, epilepsy, mental health) diseases. This Box 4.9 Delivery system redesign: Integrated Chronic Disease Management in South Africa (continued) required a complex intervention directed at each of the health system building blocks and along the continuum of care.144 Since the introduction of the ICDM model in South Africa meant a radical overhaul of PHC- level service delivery, several preparatory steps were required, such as pre-implementation preparedness at province, district and facility level, baseline assessments, and facility re- organisation. In the latter, key elements were: Re-organisation of patient flow; Introduction of appointment scheduling; Scheduling of nurses; Pre-appointment retrieval of clinical records; Creation of a designated consultation area for chronic patients; A single administrative point; Integration of clinical records; and Pre-dispensing of chronic medication. Much consideration was given to improving patient flow for chronic patients with two levels of triage. Firstly, a designated waiting area for chronic patients was created (with MCH and acute patients 36 NCD MODELS OF HEALTHCARE SERVICE DELIVERY triaged elsewhere). An additional vital signs station for chronic patients was conveniently located between the waiting area and consulting room. Secondly, after completing the vital signs, the chronic patients were further triaged and directed appropriately, into a) Repeat medication with normal vital signs, b) Repeat medication with abnormal vital signs, c) Six month full examination, and d) Doctor referral. A chronic patient record was designed, and easy-to-read evidence-based PHC clinical guidelines were made available (simplification and standardisation principle). At facility level, implementation was the responsibility of the “ICDM model champion” instead of the head of the facility, and lower level cadre such as adherence counsellors and pharmacist assistants assumed certain functions of higher level cadre (task shifting principle). Also, the guidelines offered a practical and acceptable tool to help expand the scope of practice of non-physician clinicians to include NCD care.145 Health promotion and patient tracing and support services were moved to ward-based outreach teams consisting of professional nurses, enrolled nurses and CHWs, as well as integrated school health teams (decentralisation principle). One of the main aims of the ICDM model is also to empower health care clients to take responsibility for their own health, through “assisted” self- management within the community with CHW support and point of care testing of BP and glucose (self-management and community involvement principles). The ICDM illustrates well the shift in the management of chronic disease patients from a disease-centred to a patient-centred care approach in which patients are informed and educated to participate in their own care. Early evaluation findings suggest that implementation experiences have been positive, with health staff noting improved patient care, better interaction with patients, improved level of knowledge and better teamwork coupled with an improved level of satisfaction with the work environment at the clinic.146 Although a third of the professional nurses perceived that the ICDM would increase their workload, this did not affect their experience and attitudes negatively towards the model. Relevance for Bangladesh: NCD corners have been established in UZHCs, however, tracks for MCH, acute and chronic patients as in the ICDM, and explicit triaging of patients, could further help to improve patient flow and appropriate care. The vital sign station along the track may improve compliance with routine measurement of weight, BP and glucose. Appointment scheduling combined with staff scheduling would help maximise resources, but appointment scheduling requires good data systems and strong cooperation by patients to honour the appointments. Pre-dispensing is another area with potential efficiency gains. 37 This page is for collation purposes. CONCLUSIONS AND RECOMMENDATIONS 5.1 NCD policy and resource allocation While Bangladesh has made important strides towards a policy environment which is conducive to NCD control, our review confirms While Bangladesh has that further focus on NCDs is required. Positive change includes made important strides policies on tobacco taxes, cigarette packaging and alcohol advertising, towards a policy while policies addressing dietary risks are especially weak. Within the environment which is Health Sector Programme, it is commendable that NCD control has its conducive to NCD own operational plan and budget and that diabetes and CVDs are control, our review classified as priority diseases of major public health concern. However, confirms that further the dedicated NCDC budget is small (4.2% of health sector programme, focus on NCDs is or BDT 6.8/US cents 8.2 per capita) compared to the NCD burden (63% required.. of DALYs), even if several other health plans’ budgets contribute to overall health sector capacity to address NCDs. The per capita NCD allocation is also incommensurately small compared to the estimated cost of implementing the full set of “best buy” interventions in LMIC settings (US$ 1.50 per person and year). As a consequence of low government spending, OOP expenditure for NCD care is very large, likely exceeding the OOP across all diseases (which is 66% of all health spend). Human resources in the health care sector are comparatively few compared to the recommended WHO standard of 2.28 doctors, midwives and nurses per 1000 population. RECOMMENDATIONS 1. Coordinate NCD control - An effective coordination mechanism for NCD control is urgently required, with a strong mandate for coordination, measurable targets and clear reporting processes. The coordinating body should have cross-sectoral reach with pluralistic provision. 2. Adequately finance NCD response - It is advisable that the Government consider alternate financing mechanisms. Households need to be protected from impoverishment due to NCD costs. 3. Incorporate NCD needs in Human Resources for Health Strategy - Taking into consideration the increasing burden of NCDs, a long term human resources strategy is required which responds to the changing service needs, and adequate budget allocations need to be made. 5.2 NCD burden and risk factors Our benchmarking of Bangladesh’s NCD health indicators confirms the poor performance of its health system in preventing premature death due to NCDs, especially IHD, stroke, diabetes, 39 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION COPD, CKD and hypertension. A large proportion of the NCD burden is attributable to common risk factors, of which the main ones are high blood pressure, smoking and high fasting plasma glucose. RECOMMENDATIONS 4. Address NCD risk factors - Bangladesh needs to raise awareness and risk perception in the population through the implementation of tested communication strategies and promotion of healthier lifestyles to reduce exposure. The health system needs to ensure a substantive focus on preventative and promotive care in the future, and a particular geographic focus on the rapidly urbanising areas. Stronger and additional tax policies should be considered. 5. Mainstream NCD prevention and management - Existing programmes and initiatives, such as MCH, community health and UPHCP, should integrate NCD prevention and care into their activities. 5.3 Major NCD cascade breakpoints Our cascade analysis in adults aged 35+ years using hypertension as a tracer condition and blood glucose as an indicator of diabetes demonstrates the very large burden of undiagnosed NCDs. Rural residents and males are especially prone to low diagnosis rates. International comparison confirms the low diagnosis rate in Bangladesh with the country Screening and its strategic in the lowest quartile of the diabetes diagnosis rate. There are targeting for yield and cost- also an approximately 15 million adults aged 35+ years with pre- effectiveness is not prominent hypertension and about 14 million with pre-diabetes, whereby in policy and practice in the the two conditions will overlap in many of them. Screening and public health system, resulting its strategic targeting for yield and cost-effectiveness is not in underdiagnoses, late case prominent in policy and practice in the public health system, detection, disability and excess resulting in underdiagnoses, late case detection, disability and costs to the care system and excess costs to the care system and the patients. The NCD plan the patients. suggests that 50% of adults should be screened by 2022. Data on linkage to care and treatment initiation after a person’s diagnosis was not covered in the literature available to the study team. This is surprising given the challenges arising by the weak referral systems and complex patient pathways in the multifaceted health care system of Bangladesh. As a proxy, the 2011 BDHS was used to determine the breakpoint between diagnosis (“aware” of the condition) and getting onto treatment. Since the BDHS used “treatment” and “taking medicine” interchangeably despite diabetes and hypertension having non- pharmacological and pharmacological treatment approaches, data interpretation was difficult. Our cascade analysis in adults aged 35+ years suggested that treatment initiation upon diagnosis may not be a major breakpoint, but data was limited to self-reported behaviours of 2011 DHS participants. We could however conclude that an estimated 950,000 hypertensive females (19% of diagnosed) and 370,000 hypertensive males (15% of diagnosed) who had been diagnosed were not on medication. For diabetes, we could estimate that about 140,000 diagnosed diabetic females (11%) and about 100,000 diagnosed diabetic males (9%) were not on medication. The cascade analysis demonstrates the low effectiveness of treatment among individuals reporting taking medication for hypertension or diabetes. Only 45% of people taking anti- hypertensive drugs had normal BP levels, and rates were lower in males especially in urban areas (only 38% hypertension control). Diabetes treatment had even lower effectiveness with only 36% of diabetics on medication having a normal glucose reading, again with males showing 40 CONCLUSIONS AND RECOMMENDATIONS lower success (only 29% of rural males achieving the glucose target level). Patient barriers to adherence are multiple, but long-term medication costs especially for diabetes are clearly preventing patients from complying with treatment regimens. Diabetes complications caused by ineffective treatment increase the use and cost of health care very significantly. Study data confirm the large additional expenditure of diabetes cases for hospital admissions. Though type- 1 diabetes is not a focus of this report, it is important to note that it occurs earlier in life than type-2; therefore, if poorly controlled, cases have more years lived with disability and are at risk for more years of life lost. This is an important consideration for policy and practice. RECOMMENDATIONS 6. Finding NCD cases - Bangladesh needs to train its health workers to better identify NCD risks and set screening targets at local level to help meet, by 2022, the end-of-plan national screening targets for diabetes and hypertension. Local action plans should outline how the population (especially men and rural populations) can be mobilised and reached with routine and campaign-based screening including solid referral mechanisms for screen-positives.Protocols and guidelines should be explicit on screening and linkage to care, and be accompanied by job aids for community health workers. 7. Focus on treatment adherence - Essential drugs need to be made available at PHC level at all times as this is a vital part of improving treatment adherence. Hypertension and diabetes guidelines/ protocols/ SOPs need to be explicit on treatment monitoring and treatment adherence, and present practical guidance on proven adherence support activities. These guidance documents should promote the “differentiated care” approach of focusing health workers efforts on non-adherent and unstable patients while providing easy refill options and spaced appointments for stable patients, and tracing cases lost to follow-up to facilitate re-entry into care. Health care professionals need to be educated on the guidelines and on the importance to focus on patients failing their treatment. Patients need to receive information, education and be empowered towards assisted self-monitoring with increasing use of mHealth technology where appropriate. 5.4 Integrated chronic care models The data on Bangladesh’s demographic and health transition reviewed for this analysis emphasise the need to re-orient the health care system towards chronic care and its integration in practice. The data show the different patient typologies, with Given the emphasis on many people being at a pre-disease stage in need of effective maternal health in counselling to facilitate healthier lifestyles, while others are either Bangladesh, the stable on medication and achieve normal test values, or they struggle to achieve disease control targets. Some patients are on engagement of this multiple medications, often for both diabetes and hypertension. population with the PHC level also provides an The term “differentiated care” is increasingly used for chronic opportunity for the disease management, acknowledging the different needs of development of counselling patients.147 The concept also puts emphasis on patient-centered mechanisms toward care and self-management, which are both recognised in curbing the risk of diabetes Bangladesh as important characteristics of patient care. through weight loss and Furthermore, differentiated care takes into account setting increased physical activity specificities such as rural versus urban living contexts. integrated in MCH services. This review identified principles shared between different integrated care models, including: Simplification, standardisation and monitoring of healthcare delivery processes; Decentralisation; Task-shifting and involvement of a full range of healthcare 41 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH: CONTINUUM OF CARE ASSESSMENT AND OPPORTUNITIES FOR ACTION providers; Peer and community involvement in service delivery; and Self-management. Given the emphasis on maternal health in Bangladesh, the engagement of this population with the PHC level also provides an opportunity for the development of counselling mechanisms toward curbing the risk of diabetes through weight loss and increased physical activity integrated in MCH services. RECOMMENDATIONS 8. Fully involve the Community Clinics for NCD care - Being at the heart of the government’s rural NCD model, it is vital that the CCs can lead the local NCD case finding, linkage to care and management of uncomplicated cases, while facilitating referral as per their mandate. The CCs' outreach function needs to be appropriately resourced to fulfill the enormous tasks of health promotion and community-based NCD screening especially targeting underserved populations such as older men. Low cost, low technology tools could be considered, such as the “string test for obesity” which measures the waist to height ratio and is an alternative to BMI (the string method could also be used for self-monitoring of central obesity while patients pursue lifestyle changes as part of their treatment). 9. Harness existing health work force/structures to address NCD burden – Bangladesh should consider task shifting within its health cadres in order to accommodate the changing needs for long-term monitoring and treatments, and involve paramedics and SACMOs where possible. MCH services could provide nutritional education and support to pregnant women beyond the usual nutritional advice relevant to pregnancy itself. 10. Expand human capacity – Bangladesh should adopt good practices from elsewhere on harnessing community support for the care continuum, the mobilisation of volunteers and lay workers, and the use of peer approaches. 11. Pilot an integrated Chronic Care Passport – Akin to BADAS’ diabetic book, such a patient-held passport is recommended in order to facilitate medication refill at community clinics and support self-monitoring of patients. Such a passport would also help referrals and continuity of care and likely improve the quality of care. 5.5 Surveillance, monitoring and evaluation There are various systems collecting data on NCD burden. The DHIS2, which has recently been strengthened in terms of both coverage of reporting hospitals and capacity to use ICD-10 codes, is a key system. However, our analysis of the DHIS2 hypertension and diabetes data suggest that many hospital admissions may not be captured through this system; the incidence of hospitalisation for T2DM and its complications was only 3.8/100,000 population, and for primary hypertension and hypertensive disease, it was only 19/100,000 population. Furthermore, the DHIS2 does not report disease-specific out-patient visits. Data reflecting disease episodes of hypertension and diabetes that don’t lead to hospital admission are therefore lacking. The national Sample Vital Registration System of the Bangladesh Bureau of Statistics is another pillar providing disease-specific estimates of causes of death. The 2011 BDHS remains a useful data source, including for this study. It collected anthropometry, blood pressure and blood glucose data, and it did so in older people too (including data of people in their fifties, sixties, and older ages). The icddr,b Matlab, which has been conducting health and demographic surveillance since 1966, has provided a wealth of longitudinal data and increasingly on NCD issues. Benefit- cost analyses have been conducted for hypertension and diabetes interventions, which provide powerful advocacy for investment. While there is also a wealth of data collection for research purposes, many studies are very small scale and it is not possible to draw robust conclusions from them at a population level. The forthcoming STEPS will be a useful source for monitoring NCD trends, since the same standardised questions will be asked as in the 2010 STEPS.148 42 CONCLUSIONS AND RECOMMENDATIONS Evaluations like the COBRA-BPS pilot study are central to finding effective care models and understanding the resource needs. RECOMMENDATIONS 12. Build in NCD surveillance - Bangladesh should incorporate NCDs into existing disease surveillance systems. This will enable the Government to better assess trends and its progress towards SDG Target 3.4 (reducing the premature NCD mortality by one-third by 2030). 13. Strengthen NCD M&E for better decision-making – Bangladesh should roll out the Unique Patient Identifier as a prerequisite for a powerful M&E system, and set NCD service delivery targets to measure progress. The country should document the experiences with the new MOH NCDC model and share the learning. 43 This page is for collation purposes. BANGLADESH BURDEN OF DISEASE – SUMMARY VIEW Figure A 1.1 Disability-Adjusted Life Years by cause, Bangladesh, 2016 Source: Institute for Health Metrics and Evaluation. Global Burden of Disease, 2016. 45 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH | ANNEX 1 Figure A 1.2 Distribution of population reporting chronic illness in preceding 12 months by type of illness, HIES2016 Source: HIES 2016. Figure A 1.3 Prevalence of reported diabetes by sub-population, 2016 HIES Source: HIES 2016. 46 ANNEX 1 | HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH Figure A 1.4 Death attributable to risk factors, Bangladesh, 2016 Source: Institute for Health Metrics and Evaluation. Global Burden of Disease, 2016. 47 BENCHMARKING BANGLADESH WITH COMPARISON COUNTRIES DEATH, DISABILITY AND LIFE EXPECTANCY Table A 2.1 Benchmarking Bangladesh with countries with similar GNI per capita (ATLAS method; <20% difference in GNI per capita) AGE- AGE- DALYS STANDARDISED STANDARDISED YLL (DISABILITY- LIFE GNI PER DEATH RATE RATE ADJUSTED EXPECTANCY AT CAPITA (PER 100,000) (PER 100,000) LIFE YEARS) BIRTH 2016 2016 2016 2016 2016 COUNTRY $ RATE RANK RATE RANK RATE RANK YEARS RANK Kyrgyz Republic 1,100 602 5 19,829 2 29,302 2 71.24 3 Tajikistan 1,110 481 1 20,115 3 28,652 3 71.76 2 Mauritania 1,120 492 2 22,807 5 32,254 4 70.16 4 Cambodia 1,140 618 6 22,752 4 32,318 5 68.75 5 Cameroon 1,200 911 11 48,709 11 58,107 11 60.14 9 Lesotho 1,210 1,525 12 67,150 12 78,427 12 50.31 12 Zambia 1,300 904 10 44,772 9 53,887 9 58.62 11 Bangladesh 1,330 524 3 17,177 1 27,552 1 72.62 1 Ghana 1,380 670 7 31,106 8 40,268 8 66.01 8 Kenya 1,380 579 4 26,821 6 35,730 6 66.84 7 Pakistan 1,510 671 8 26,909 7 36,985 7 67.61 6 Cote d'Ivoire 1520 903 9 47,580 10 57,404 10 59.87 10 Sources: World Development Indicators (GNI per capita, Atlas method (current US$), 2016); https://data.worldbank.org; Institute for Health Metrics and Evaluation. Global Burden of Disease, 2015. https://vizhub.healthdata.org/gbd-compare/ Accessed 8 Nov, 2017. Note: YLL = Years of Life Lost. Table A 2.2 Benchmarking Bangladesh with countries in the South Asia Region AGE-STANDARDISED AGE-STANDARDISED DALYS (DISABILITY- LIFE DEATH RATE (PER YLL RATE (PER ADJUSTED EXPECTANCY AT 100,000) 100,000) LIFE YEARS) BIRTH 2016 2016 2016 2016 COUNTRY RATE RANK RATE RANK RATE RANK YEARS RANK Afghanistan 910 8 44,926 8 55,566 8 58.0 8 Bangladesh 524 3 17,177 4 27,552 4 72.6 4 Bhutan 486 2 16,316 3 27,313 3 73.8 3 India 744 7 23,808 6 35,435 6 68.6 6 Maldives 311 1 7,380 1 16,754 1 77.1 2 Nepal 604 5 20,625 5 30,683 5 70.9 5 Pakistan 671 6 26,909 7 36,985 7 67.6 7 Sri Lanka 592 4 13,656 2 24,134 2 77.4 1 Sources: https://data.worldbank.org; Institute for Health Metrics and Evaluation. Global Burden of Disease, 2015. https://vizhub.healthdata.org/gbd-compare/ Accessed 6 April, 2018. 48 ANNEX 2 | HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH PREMATURE DEATH DUE TO NCDS Table A 2.3 Benchmarking Bangladesh with countries with similar GNI per capita (ATLAS method; <20% difference in GNI per capita) CHRONIC 2016 ISCHEMIC CEREBRO- CHRONIC OBSTRUCTIVE GNI DIABETES HEART VASCULAR KIDNEY PULMONARY PER ALL NCDS MELLITUS DISEASE DISEASE DISEASE DISEASE CAPITA 2016 2016 2016 2016 2016 2016 COUNTRY US$ % Rank % Rank % Rank % Rank % Rank % Rank Kyrgyz Rep. 1,100 57.0 12 0.6 1 16.0 12 8.3 11 1.5 10 2.2 11 Tajikistan 1,110 42.2 9 1.4 8 10.9 10 5.5 9 1.3 8 1.7 9 Mauritania 1,120 32.0 7 1.0 6 4.3 7 2.8 6 1.1 7 0.6 3 Cambodia 1,140 45.3 10 1.6 9 5.7 8 7.2 10 1.5 9 1.3 8 Cameroon 1,200 22.3 4 0.8 4 1.9 2 1.9 2 0.8 4 0.6 3 Lesotho 1,210 18.3 2 2.2 12 2.2 3 2.4 4 0.6 3 1.2 7 Zambia 1,300 18.9 3 0.8 4 2.4 4 1.7 1 0.5 2 0.5 1 Bangladesh 1,330 51.9 11 1.9 11 10.3 9 9.8 12 1.9 12 4.2 12 Ghana 1,380 29.5 6 1.3 7 3.5 6 3.6 7 1.0 6 0.6 3 Kenya 1,380 16.4 1 0.6 1 1.7 1 1.9 2 0.3 1 0.5 1 Pakistan 1,510 42.1 8 1.7 10 11.1 11 4.9 8 1.8 11 2.0 10 Cote d'Ivoire 1,520 24.3 5 0.9 5 3.2 5 2.7 5 0.9 5 0.6 3 Sources: World Development Indicators (GNI per capita, Atlas method (current US$), 2016); Institute for Health Metrics and Evaluation. Global Burden of Disease, 2015. https://vizhub.healthdata.org/gbd-compare/ Accessed 8 Nov, 2017. Table A 2.4 Benchmarking Bangladesh with countries in the South Asia Region CHRONIC ISCHEMIC CEREBRO- CHRONIC OBSTRUCTIVE DIABETES HEART VASCULAR KIDNEY PULMONARY ALL NCDS MELLITUS DISEASE DISEASE DISEASE DISEASE 2016 2016 2016 2016 2016 2016 COUNTRY % Rank % Rank % Rank % Rank % Rank % Rank Afghanistan 36.1 1 2.0 5 9.1 1 4.2 1 1.7 1 1.1 1 Bangladesh 51.9 6 1.9 2 10.3 2 9.8 8 1.9 3 4.2 5 Bhutan 50.4 5 2.5 6 10.9 3 4.5 2 2.3 6 4.1 3 India 47.5 4 1.9 2 12.5 6 4.9 3 2.0 5 5.1 8 Maldives 73.2 8 2.9 8 20.2 8 5.9 6 5.6 8 5.0 7 Nepal 45.4 3 1.9 2 10.9 3 5.0 5 1.9 3 4.9 6 Pakistan 42.1 2 1.7 1 11.1 5 4.9 3 1.8 2 2.0 2 Sri Lanka 70.6 7 6.8 7 16.6 7 6.5 7 3.3 7 2.9 3 Sources: World Development Indicators (GNI per capita, Atlas method (current US$), 2016); Institute for Health Metrics and Evaluation. Global Burden of Disease, 2015. https://vizhub.healthdata.org/gbd-compare/ Accessed 6 April, 2018. 49 HYPERTENSION AND DIABETES BURDEN BY DIVISION Figure A 3.1 Estimated total burden of hypertension in adults aged 18 years and above in Bangladesh Divisions Sources: Data Sources: Population estimates from Health Bulletin 2016 and UN Population data on age structure; Hypertension prevalence from 2011 BDHS in 35+, adjusted for adults 18+ using 2016 meta analysis of 24 hypertension prevalence studies by Fatema et al. 2016 PLoS ONE 11 (8): e0160180. Dhaka Division data from 2011 BDHS were used for the newly demarcated Dhaka and Mymensingh Divisions. Figure A 3.2 Estimated total burden of diabetes in adults aged 18 years and above in Bangladesh Divisions Sources: Population estimates from Health Bulletin 2016 and UN Population data on age structure; Diabetes prevalence from 2011 BDHS in 35+, adjusted for adults 18+ using 2016 meta analysis of 24 hypertension prevalence studies by Fatema et al. 2016 PLoS ONE 11 (8): e0160180. Dhaka Division data from 2011 BDHS were used for the newly demarcated Dhaka and Mymensingh Divisions. 50 HOSPITAL ADMISSIONS FOR HYPERTENSION AND TYPE-2 DIABETES DHIS2 DATA COVERING THE PERIOD FROM 1 NOVEMBER 2016‒31 OCTOBER 2017. Figure A 4.1 Reported hospital admissions for hypertension by Bangladesh Division Note: ICD-10 codes included are primary hypertension (I10) and hypertensive disease (I11‒I13) 51 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH | ANNEX 4 Figure A 4.2 Reported hospital admissions for diabetes by Bangladesh Division Note: ICD-10 codes included are type-2 diabetes (E10) and diabetes with complications (E100-108) 52 BARRIERS IN THE CONTINUUM OF CARE ON PATIENT AND PROVIDER SIDES Table A 5.1 Hypertension care: Overview of patient and provider barriers to the continuum of cascades BARRIERS TO… PATIENT SIDE PROVIDER SIDE Screening/ ▪ Poor understanding of hypertension ▪ Weak integration of hypertension care at diagnosis control measures primary level ▪ Health seeking and disease ▪ Low capacity and focus to address hypertension prevention behaviours at PHC levels ▪ Awareness of future treatment costs ▪ Poor referral mechanisms ▪ Lack of risk perception ▪ Poor information flow between level of diagnosis ▪ Lack of symptoms and PHC level ▪ Scepticism about treatment ▪ Non-compliance with screening guidelines at PHC level due to insufficient training and lack of equipment Treatment ▪ Cost of drugs ▪ Poor patient support for linkage to treatment initiation ▪ Poor knowledge on disease ▪ Consultations too short for proper patient ▪ Non-adherence to lifestyle changes engagement ▪ Uncertainty with drugs availability Treatment ▪ Non-adherence to regular check-ups ▪ Recall systems missing monitoring ▪ Improper blood pressure measurement Treatment ▪ Long-term cost of drugs (low income) ▪ Low capacity to deliver chronic care at PHC level adherence/ ▪ Poor knowledge on hypertension ▪ Continuity of drug availability at PHC level disease control impacts, complications and CVDs ▪ Lack of routine system to track BP control ▪ Memory or psychiatric problems ▪ Inadequate drug doses or combinations ▪ Polypharmacy with complicated ▪ Physician inertia with failure to change or dosing schedules increase dose regimens when not reaching goal Sources: Das et al. 2017. Health literacy in a community with low levels of education: findings from Chakaria, a rural area of Bangladesh. BMC Public Health (2017) 17:203; National Guidelines for Management of Hypertension in Bangladesh, 2013. NCDC/National Heart Foundation/WHO; Kasonde et al. (2017). Evaluating medicine prices, availability and affordability in Bangladesh using WHO and Health Action International methodology; In-country discussions with key informants. 53 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH | ANNEX 5 Table A 5.2 Diabetes care: Overview of patient and provider barriers to the continuum of care BARRIERS TO… HEALTH CARE CLIENT SIDE HEALTH CARE PROVIDER SIDE Screening/ ▪ Poor understanding of diabetes control ▪ Weak integration of diabetes care at diagnosis measures public sector primary level ▪ Insufficient knowledge about risk factors ▪ Lack of statistics at PHC level ▪ Religious beliefs of disease etiology ▪ Non-compliance with screening ▪ Health seeking and disease prevention guidelines at PHC level due to insufficient behaviours training and lack of equipment ▪ Cost of diagnosis (e.g. test strips) ▪ Awareness of future treatment costs ▪ Lack of risk perception Treatment ▪ Cost of oral anti-diabetes drugs ▪ Poor patient support for linkage to initiation ▪ Cost of insulin analogs treatment ▪ Poor knowledge on disease ▪ Consultations too short for proper ▪ Non-adherence to lifestyle changes patient engagement ▪ Uncertainty with drugs availability Treatment ▪ Travel and clinic waiting times affecting ▪ Overburdened specialist facilities able to monitoring regular blood sugar checks provide comprehensive check-ups with ▪ Cost of travel and cost of testing cardiovascular, renal and eyesight ▪ Low satisfaction with diabetes care provided examinations ▪ Need to pay private laboratory for routine ▪ Lack of availability of basic HbA1C testing government/NGO medical services ▪ Need to purchase self-monitoring equipment ▪ Low availability of HbA1C test in public such as glucometer sector (equipment and supplies) Treatment ▪ Insufficient knowledge about disease control ▪ Insufficient insulin titration adherence/ and prevention of complications ▪ Continuity of access to oral drugs disease control ▪ Long-term cost of drugs ▪ Lack of routine system to track ABC ▪ Sub-optimal support to patient from family control ▪ Poor control of obesity for better glycaemic ▪ Lack of diagnostic capacity at PHC level control and prevention of complications, to detect vascular flow problems linked to wrong perception of own weight ▪ Misconceptions about disease ▪ Non-compliance to dietary and physical activity advice ▪ Concerns over medication side effects ▪ Beliefs and prejudices regarding the intake of food ▪ Fear of hypoglycemia ▪ Episodes of hypoglycemia Sources: Das et al. 2017. Health literacy in a community with low levels of education: findings from Chakaria,a rural area of Bangladesh. BMC Public Health (2017) 17:203; Siddique et al 2017; . Fatema et al. 2017; Islam FMA et al 2014. Knowledge, attitudes and practice of diabetes in rural Bangladesh: the Bangladesh population based diabetes and eye study (BPDES). PLoS One. 2014;9(10):e110368; Islam SMS et al. 2017. Patients’ perspective of disease and medication adherence for type 2 diabetes in an urban area in Bangladesh: a qualitative study. BMC Res Notes (2017) 10:131; Munni US et al. 2017. Nutritional beliefs and practices among diabetic pregnant mothers in a tertiary care hospital in Bangladesh. Diabetes & Metabolic Syndrome: Clinical Research & Reviews 11, 287–290; Emral R et al 2017 - Self- reported hypoglycemia in insulin-treated patients with diabetes; Lewis and Newell 2014. 54 INTERNATIONAL COMPARISON OF DIABETES DIAGNOSIS RATES Figure A 6.1 Diagnosis rates of diabetes across countries Source: IDF, 2017. Note: Only includes countries/ territories with >50,000 adult diabetes cases 55 BANGLADESH’S NCD POLICY CONTEXT IN COMPARISON Figure A 7.1 Bangladesh NCD policy context compared to countries with similar GNI per capita Note: ATLAS method; <20% difference in GNI per capita. Figure A 7.2 Bangladesh NCD score card results in international comparison Source: NCD scorecard on http://www.ncdglobalscorecard.org/ 56 HEALTH SECTOR CHARACTERISTICS Figure A 8.1 Bangladesh total health spending, 2014 Source: http://www.healthdata.org/bangladesh. Figure A 8.2 Allocations and level of expenditure of HPNSDP in FY 2015‒16 Source: Health Bulletin 2016 57 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH | ANNEX 8 Figure A 8.3 Share of government funding across health implementation plans, 2017‒22 Source: Programme Implementation Plan, 4 Health, Population and Nutrition Sector Programme 2017‒22. Table A 8.1 Facility distribution in rural areas of Bangladesh LEVEL FACILITIES NUMBER OF FACILITIES BEDS Upazila PHC upazila hospitals 482 19,508 Community centers 13,336 Union PHC hospitals 51 830 Ambulatory care facilities 3,083 Secondary and tertiary level Hospitals 131 29,426 Source: Health Bulletin 2016 58 COSTS OF NCD INTERVENTIONS Figure A 9.1 Unit cost distribution for hypertension management, diabetes mellitus, stroke and acute ischemic heart disease Source: Brouwer et al. BMC Public Health (2015) 15:1183. Note: Costs per patient, based on a systematic review of LIC and LMIC data. Figure A 9.2 Estimated cost of interventions to reduce NCD risk factors in three countries Source: Beaglehole 2011. Note: Annual cost per person in US$ 59 LIST OF MEDICINES FOR HYPERTENSION AND DIABETES IN GOVERNMENT NCD CONTROL PLAN Table A 10.1 Drug list for Diabetes Mellitus SL #. PREPARATION NAME OF ITEMS 1 Tablet Glibenclamide 5 mg 2 Tablet Glicalzide 80 mg/30mg 3 Tablet Gilipizide 5 mg 4 Tablet Glimepiride 1 mg 5 Tablet Metformin HCL 850 mg 6 Tablet Metformin 500 mg 7 Tablet Linagliptin 5 mg 8 Tablet Regaglinide Source: Operational Plan NCD Control 2017‒22 (Plan appendix IX). Table A 10.2 List of medicines for Hypertension SL #. PREPARATION NAME OF ITEMS SL #. PREPARATION NAME OF ITEMS 1 Tablet Amlodipine 5 mg 12 Tablet Frusemide 40 mg and Torasemide 2.5 mg 2 Tablet Propranolol 10 mg/ 13 Tablet Spirolactone 25 mg/ Spirolactone 40 mg 50 mg + Frusemide 20 mg 3 Tablet Atenolol 50 mg 14 Tablet Thiazide 50 mg + Amiloride 5 mg 4 Tablet Carvidolol 6025 mg/12.5 mg 15 Tablet Thiazide 25 mg + Triamtrene 50 mg 5 Injection Frusemide 20 mg/2 ml 16 Injection Manitorl 20% (injectable) 6 Tablet Metoprolol 50 mg 17 Tablet Acetazolamide 250 mg 7 Tablet Captopril 25 mg 18 Tablet Atenolol 50 mg + Amoldipine 5 mg 8 Tablet Emalapril maleate 19 Tablet Benzapril 10 mg + 5 mg/10 mg Amoldipine 5 mg 9 Tablet Losartan potassium 20 Tablet Indapamide 1.25 mg + 25 mg/ 50 mg Perindopril 4 mg 10 Tablet Valsartan 40 mg/80 mg/ 21 Tablet Losartan 50 mg + 160 mg Hydrochlorothiazide 12.5 mg 11 Tablet Thiazide 25 mg/50 mg 22 Tablet Valsartan 80 mg + Hydrochlorothiazide 12.5 mg 60 THE NEW NCDC CARE MODEL AND THE CONTINUUM OF CARE Figure A 11.1 Infographic: New model of NCD care in A) Urban and B) Rural Bangladesh A. Urban settings B. Rural settings 61 This page is for collation purposes CITATION AND NOTES 1 WHO 2017 NCD Progress Monitor 2 Biswas T et al. (2017) Bangladesh policy on prevention and control of noon-communicable diseases: a policy analysis. BMC Public Health 17:582. 3 Koehlmoos T et al. (2016). Controlling the Burden of Non-Communicable Diseases in Bangladesh: Benefit-cost Analysis of Prevention Policies and Interventions: Bangladesh priorities. 4 Huque R, 2016. Costs and Benefits of addressing Diabetes and Smokeless Tobacco Consumption via Community Clinics. Bangladesh Priorities, Copenhagen Consensus Center, 2016. License: Creative Commons Attribution CC BY 4.0. 5 Islam SMS et al. 2017 6 Fatema K et al (2016) Prevalence of Risk Factors for Cardiovascular Diseases in Bangladesh: A Systematic Review and Meta-Analysis. PLoS ONE 11 (8): e0160180. doi:10.1371/journal.pone.0160180 7 Systolic blood pressure 160 mmHg or above, diastolic blood pressure 100 mmHg or above 8 DGHS/National Heart Foundation/WHO 2013. National Guidelines for Management of hypertension in Bangladesh. WHO. ISBN 978-984-33-6855-3 9 IDF website, 2017 10 Fatema K et al (2016) Prevalence of Risk Factors for Cardiovascular Diseases in Bangladesh: A Systematic Review and Meta-Analysis. PLoS ONE 11 (8): e0160180. doi:10.1371/journal.pone.0160180 11 Mahatab H, Zaman MM, Ullah AKMJ (2013). National Guidelines for the Management of Diabetes Mellitus in Bangladesh at Primary Care. DGHS / BIRDEM / WHO Dhaka Bangladesh. 12 Vas A, Devi ES, Vidyasagar S, Acharya R, Rau NR, George A, et al. Effectiveness of self-management programmes in diabetes management: A systematic review. International journal of nursing practice. 2017;23(5). 13 World Development Indicators, indicator ‘cause of death, by NCD (% of total)’ 14 UNGA Resolution A/RES/66/2. Political Declaration of the High-level Meeting of the General Assembly on the Prevention and Control of Non-communicable Diseases. New York (NY): United Nations; 2012 (http://daccessdds-ny.un.org/doc/UNDOC/GEN/N11/458/94/PDF/N1145894.pdf?OpenElement 15 https://sustainabledevelopment.un.org/sdg3 16 WHO 2014. Global Status Report on Non-communicable Diseases. World Health Organization, Switzerland, 2014. 17 WHO 2014. Global Status Report on Non-communicable Diseases. World Health Organization, Switzerland, 2014. 18 WHO 2014. Assessing national capacity for the prevention and control of non-communicable diseases: report of the 2013 global survey. Geneva: WHO; 2014 19 WHO 2014. Global Status Report on Non-communicable Diseases. World Health Organization, Switzerland, 2014. 20 WHO 2011 From Burden to “Best Buys”: Reducing the Economic Impact of NCDs 21 Mulligan et al. 2006. Economic evaluations of NCD interventions in developing countries: a critical review of the evidence base. Cost Effectiveness and Resource Allocation 2006, 4:7 22 World Bank data, 2016 estimation 23 GBD Results Tool, http://ghdx.healthdata.org/gbd-results-tool 24 Household Income and Expenditure Survey 2016: http://www.bbs.gov.bd/ 25 World Development Indicators, World Bank data. 26 E.g., Infant mortality rate reducing from 134/100,000 live births in 1980 to 64 in 2000 and 28 in year 2016, and Maternal mortality ratio falling from 351/100,000 live births in 2007 to 210 in year 2011, https://data.worldbank.org, accessed 11 Nov 2017 27 https://www.healthdata.org/sites/default/files/files/country_profiles/GBD/ ihme_gbd_country_report_bangladesh.pdf 63 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH | CITATION AND NOTES 28 The Tobacco Atlas, https://tobaccoatlas.org/country/bangladesh/, accessed 9 April 2018 29 Eriksen M et al, 2012 M, The Tobacco Atlas. Fourth Ed. Atlanta, Georgia: World Lung Foundation. 30 Hatsukami D et al, 2014. Smokeless Tobacco and Public Health: A Global Perspective. National Cancer Institute and the Centers for Disease Control and Prevention. 31 Huque R, 2016. Costs and Benefits of addressing Diabetes and Smokeless Tobacco Consumption via Community Clinics. Bangladesh Priorities, Copenhagen Consensus Center, 2016. License: Creative Commons Attribution CC BY 4.0. 32 WHO SEARO 2007. 33 The risk of NCDs starts during gestation and infancy, and accumulates from early childhood (e.g. Aboderin et al. 2001, Life Course Perspectives on Coronary Heart Disease, Stroke and Diabetes: Key Issues and Implications for Policy and Research. Geneva, Switzerland: WHO). Several studies have shown that malnourished women tend to deliver low-birth-weight babies who are more likely to have stunted growth, cognitive impairment, poor school performance and reduced productivity, which all increase the probability of developing hypertension and CVD (e.g. Barker et al. 1989, Weight in Infancy and Death from Ischaemic Heart Disease. The Lancet 334: 577–80). 34 World Bank data, https://data.worldbank.org/, accessed 11 Nov 2017 35 See compilation of studies in Table II “Prevalence of hypertension in Bangladesh” by Islam AKMM et al. 2016. Cardiovascular disease in Bangladesh: A review. Bangladesh Heart Journal; 31(2) : 80-99 36 Fatema K et al (2016) Prevalence of Risk Factors for Cardiovascular Diseases in Bangladesh: A Systematic Review and Meta-Analysis. PLoS ONE 11 (8): e0160180. doi:10.1371/journal.pone.0160180 37 An individual is classified as having hypertension if s/he has blood pressure levels >=140 mmHg SBP or >=90 mmHg DBP, or s/he is currently taking antihypertensive medication to lower their blood pressure (Table 15.3.1., 2011 DHS) 38 BP 120-139 mmHg SBP or 80-89 mmHg DBP and not reporting taking anti-hypertension drugs. 39 Chowdhury et al. 2016.Hypertension among adults in Bangladesh: Evidence from a national cross-sectional survey. BMC Cardiovascular Disorders (2016) 16:22 40 Asia Arsenic Network (2016). Risk reduction of non-communicable diseases in Jessore District. Final report. February 2016. 41 Ali N et al. 2017. Hypertension prevalence and influence of basal metabolic rate on blood pressure among adult students in Bangladesh. BMC Public Health (2018) 18:58 42 Stage 2 = BP 160+ mmHG SBP or 100+ mmHg DBP 43 WHO 2013. National Guidelines for Management of hypertension in Bangladesh. ISBN 978-984-33-6855-3 44 Huque R, 2016 Costs and Benefits of addressing Diabetes and Smokeless Tobacco Consumption via Community Clinics. Bangladesh Priorities, Copenhagen Consensus Center, 2016. License: Creative Commons Attribution CC BY 4.0. 45 IDF 2017 Database website 46 An individual is classified as having hypertension if s/he has blood pressure levels >=140 mmHg SBP or >=90 mmHg DBP, or s/he is currently taking antihypertensive medication to lower their blood pressure (Table 15.3.1., 2011 DHS) 47 BP 120-139 mmHg SBP or 80-89 mmHg DBP and not reporting taking anti-hypertension drugs. 48 Fottrell E et al. 2016. The effect of community groups and mobile phone messages on the prevention and control of diabetes in rural Bangladesh: study protocol for a three-arm cluster randomised controlled trial. Trials (2016) 17:600 49 Alam DS et al. 2016. Overweight and abdominal obesity as determinants of undiagnosed diabetes and pre- diabetes in Bangladesh. BMC Obesity (2016) 3:19 50 Rahman MS et al. 2015. Awareness, treatment, and control of diabetes in Bangladesh: A Nationwide Population-Based Study. PLoS ONE 10(2): e0118365. doi:10.1371/journal.pone.0118365 51 STEPS survey 2010. http://www.who.int/chp/steps/2010_STEPS_Report_Bangladesh.pdf 52 Islam SMS et al. (2017). Healthcare use and expenditure for diabetes in Bangladesh. BMJ Global Health 2017;2:e000033. doi:10.1136/bmjgh-2016-000033 53 A note on completeness of capturing health facilities within diabetes care network 54 Islam 2009 in WB NCD report). 55 Lewis CP & Newell JN (2014). Patients’ perspectives of care for type 2 diabetes in Bangladesh –a qualitative study. BMC Public Health, 14:737 56 Fatema K et al. (2017). Knowledge attitude and practice regarding diabetes mellitus among Nondiabetic and diabetic study participants in Bangladesh. BMC Public Health, 17:364 57 Farzana S, Mumu SJ, Ferdous A, Liaquat A, Sharmin H, Ahmed KR (2012). Knowledge, attitude and practice of type 2 diabetic patients regarding obesity: study in a tertiary care hospital in Bangladesh. J Public Health Africa; 3:1. 64 CITATION AND NOTES | HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH 58 Islam FMA, Chakrabarti R, Dirani M, Islam MT, Ormsby G, Wahab M, Critchley, C, Finger RP (2014). Knowledge, attitudes and practice of diabetes in rural Bangladesh: the Bangladesh population based diabetes and eye study (BPDES). PLoS One; 9(10):e110368. 59 Siddique MKB et al. (2017). Diabetes knowledge and utilization of healthcare services among patients with type 2 diabetes mellitus in Dhaka, Bangladesh. BMC Health Services Research (2017) 17:586 60 Siddique MKB et al. (2017). Diabetes knowledge and utilization of healthcare services among patients with type 2 diabetes mellitus in Dhaka, Bangladesh. BMC Health Services Research (2017) 17:586 61 Siddique MKB et al. (2017). Diabetes knowledge and utilization of healthcare services among patients with type 2 diabetes mellitus in Dhaka, Bangladesh. BMC Health Services Research (2017) 17:586 62 Fatema K et al. (2017). Knowledge attitude and practice regarding diabetes mellitus among Nondiabetic and diabetic study participants in Bangladesh. BMC Public Health, 17:364 63 Islam SMS et al. (2017). Patients’ perspective of disease and medication adherence for type 2 diabetes in an urban area in Bangladesh: a qualitative study. BMC Res Notes (2017) 10:131 64 Lewis CP & Newell JN (2014). Patients’ perspectives of care for type 2 diabetes in Bangladesh –a qualitative study. BMC Public Health, 14:737 65 Banu B et al. (2014). Prescribing Behavior of Diabetes Treating Physicians in Selected Health Care Facilities of the Diabetic Association of Bangladesh. Indian Journal of Public Health, Volume 58, Issue 3, 180-185 66 Lewis CP & Newell JN (2014). Patients’ perspectives of care for type 2 diabetes in Bangladesh –a qualitative study. BMC Public Health, 14:737 67 Mahatab H, Zaman MM, Ullah AKMJ (2013). National Guidelines for the Management of Diabetes Mellitus in Bangladesh at Primary Care. WHO Dhaka Bangladesh. 68 Biswas T et al. (2017) Bangladesh policy on prevention and control of noon-communicable diseases: a policy analysis. BMC Public Health 17:582. 69 WHO 2017. Non-communicable Diseases Progress Monitor, 2017. Geneva: World Health Organization; 2017. Licence: CC BY-NC-SA 3.0 IGO. 70 NCD scorecard assessment, http://www.ncdglobalscorecard.org/ 71 Financing Global Health Database 2016, http://www.healthdata.org/bangladesh 72 MOHFW (2016). Health Bulletin 2016 73 Program Implementation Plan, 4 Health, Population and Nutrition Sector Programme 2017-2022 74 Global financing database, from IHME Bangladesh country profile http://www.healthdata.org/bangladesh 75 Financing Global Health Database 2016, http://www.healthdata.org/bangladesh 76 Holloway, 2010 77 WHO (2006). The world health report 2006: working together for health. Geneva: World Health Organization. 78 Shams El Arifeen, Aliki Christou, Laura Reichenbach, Ferdous Arfi na Osman, Kishwar Azad, Khaled Shamsul Islam*, Faruque Ahmed,Henry B Perry, David H Peters; Community based approaches and parterships: innovations in health -service Delivery in Bangladesh, Lancet 2013 79 Urban Health Strategy: http://uphcp.gov.bd/cmsfiles/files/National%20Urban%20Health%20Strategy-2014(1).pdf. See also: http://uphcp.gov.bd/Vision_Mission 80 Rubinstein A et al. (2010). Estimation of the Burden of Cardiovascular Disease Attributable to Modifiable Risk Factors and Cost-Effectiveness Analysis of Preventative Interventions to Reduce This Burden in Argentina. BMC Public Health 10 (1): 627; Wang YC et al. (2001). Effectiveness and Cost-Effectiveness of Blood Pressure Screening in Adolescents in the United States. Journal of Pediatrics 158 (2): 257–64.e1–7. 81 Gaziano TA et al. (2005). Cost-Effectiveness Analysis of Hypertension Guidelines in South Africa: Absolute Risk versus Blood Pressure Level. Circulation 112 (23): 3569–76.; Lim SS et al. (2007). Prevention of Cardiovascular Disease in High-Risk Individuals in Low-Income and Middle-Income Countries: Health Effects and Costs.” The Lancet 370 (9604): 2054–62.; Rubinstein A et al. (2010). Estimation of the Burden of Cardiovascular Disease Attributable to Modifiable Risk Factors and Cost-Effectiveness Analysis of Preventative Interventions to Reduce This Burden in Argentina. BMC Public Health 10 (1): 627. 82 DCCT (Diabetes Control and Complications Trial) Research Group (1993). The Effect of Intensive Treatment of Diabetes on the Development and Progression of Long-Term Complications in Insulin-Dependent Diabetes Mellitus. New England Journal of Medicine 329 (14): 977–86.; UKPDS Group (1998). Intensive Blood-Glucose Control with Sulphonylureas or Insulin Compared with Conventional Treatment and Risk of Complications in Patients with Type 2 Diabetes (UKPDS 33). The Lancet 352 (9131): 837–53. 83 DPP Research Group (2012). The 10-Year Cost-Effectiveness of Lifestyle Intervention or Metformin for Diabetes Prevention: An Intent-to-Treat Analysis of the DPP/DPPOS.” Diabetes Care 35 (4): 723–30. 84 Echouffo-Tcheugui JB et al. (2011). Screening for Type 2 Diabetes and Dysglycemia. Epidemiologic Reviews 33 (1): 63–87; Gillies CL et al. (2008). Different Strategies for Screening and Prevention of Type 2 Diabetes in Adults: Cost-Effectiveness Analysis. BMJ 336 (7654): 1180–85; Hoerger TJ et al. (2004). Screening for Type 2 65 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH | CITATION AND NOTES Diabetes Mellitus: A Cost-Effectiveness Analysis. Annals of Internal Medicine 140 (9): 689–99; Khunti, K et al.(2012). A Comparison of Cost per Case Detected of Screening Strategies for Type 2 Diabetes and Impaired Glucose Regulation: Modelling Study.” Diabetes Research and Clinical Practice 97 (3): 505–13. 85 ADA (American Diabetes Association). 2016. Standards of Medical Care in Diabetes—2016. Diabetes Care 39 (Suppl 1); IDF (2013). Clinical Practice Guidelines. Brussels: IDF. http://www.idf.org/guidelines; NICE (2013). Preventing Type 2 Diabetes: Risk Identification and Interventions for Individuals at High Risk (PH38). Public Health Guidance 2013. http://www.nice.org.uk/guidance /index.jsp?action=byID&o=13791. 86 WHO 2011. Scaling up action against non-communicable diseases: how much will it cost? ISBN 978 92 4 150231 3 87 Huque R (2016). Costs and Benefits of addressing Diabetes and Smokeless Tobacco Consumption via Community Clinics. Bangladesh Priorities, Copenhagen Consensus Center, 2016. License: Creative Commons Attribution CC BY 4.0. 88 Siddique MKB et al. (2017). Diabetes knowledge and utilization of healthcare services among patients with type 2 diabetes mellitus in Dhaka, Bangladesh. BMC Health Services Research (2017) 17:586 89 Islam MZ (2010). Burden of Diabetes Mellitus: Experience from Urban and Rural Communities of Bangladesh. Thesis 90 Rahman MZ et al. (2013). Economic burden of type 2 diabetes mellitus among the patients attending combined military hospital, Dhaka. JAFMC Bangladesh. Vol 9, No 1 (June) 2013 91 Koehlmoos T et al. (2016). Controlling the Burden of Non-Communicable Diseases in Bangladesh: Benefit-cost Analysis of Prevention Policies and Interventions: Bangladesh priorities. 92 IDF 2017 database 93 Afroz A et al. (2016). Healthcare cost of type 2 diabetes mellitus in Bangladesh: a hospital-based study. Int J Diabetes Dev Ctries (April–June 2016) 36(2):235–241 94 Lewis CP & Newell JN (2014). Patients’ perspectives of care for type 2 diabetes in Bangladesh –a qualitative study. BMC Public Health, 14:737. 95 Islam SMS et al. (2017). Healthcare use and expenditure for diabetes in Bangladesh. BMJ Global Health 2017;2:e000033. doi:10.1136/bmjgh-2016-000033 96 Humaira S (2012). Cost-effectiveness analysis of medical intervention in patients with early detection of diabetic retinopathy in a tertiary care hospital in Bangladesh. The Experiment, December .2012 Vol.5 (3),2 93-303 97 Unnikrishnan RI et al. (2007). Prevalence and Risk Factors of Diabetic Nephropathy in an Urban South Indian Population: The Chennai Urban Rural Epidemiology Study (CURES 45). Diabetes Care 30 (8): 2019–24. 98 World Bank (2011). Chronic Emergency: Why NCDs Matter.” Health, Nutrition, and Population Discussion Paper. 2011. Washington DC: World Bank. 99 WHO (2011). From Burden to “Best Buys” - Reducing the Economic Impact of Non-Communicable Diseases in Low- and Middle-Income Countries 100 WHO (2015). WHO global coordination mechanism on the prevention and control on NCDs, working group on how to realize governments commitment to provide financing for NCDs, Geneva 2015, http://www.who.int/nmh/ncd-coordination-mechanism/Policybrief5.2docx.pdf 101 Zaman MM et al (2016). Non-communicable Disease Prevention and Control Situation in a Primary Health Care Setting of Bangladesh: Design and Baseline Findings of an Intervention. Chronic Dis Int; 3(1): 1021. 102 National Institute of Population Research and Training (NIPORT), Associates for Community and Population Research (ACPR), and ICF International. 2016. Bangladesh Health Facility Survey 2014. Dhaka, Bangladesh: NIPORT, ACPR, and ICF International. 103 Zaman MM et al (2016). Non-communicable Disease Prevention and Control Situation in a Primary Health Care Setting of Bangladesh: Design and Baseline Findings of an Intervention. Chronic Dis Int; 3(1): 1021. 104 Zaman MM et al (2016). Non-communicable Disease Prevention and Control Situation in a Primary Health Care Setting of Bangladesh: Design and Baseline Findings of an Intervention. Chronic Dis Int; 3(1): 1021. 105 WHO, 2003. A public health approach to antiretroviral treatment: overcoming constraints. 106. WHO, 2014. Report on the Regional meeting on strengthening the integration and management of non- communicable diseases in primary health care. Cairo, Egypt 8–10 September 2014. 107. Mallow JA, Theeke LA, Barnes ER, Whetsel T, Mallow BK. Using mHealth Tools to Improve Rural Diabetes Care Guided by the Chronic Care Model. J Rural Nurs Health Care, 2015; 14(1): 43–65. doi:10.14574/ojrnhc.v14i1.276]. 108. Döbler A, Herbeck Belnap B, Pollmann H, Farin E, Raspe H, Mittag O. Telephone-delivered lifestyle support with action planning and motivational interviewing techniques to improve rehabilitation outcomes. Rehabil Psychol. 2018 May;63(2):170-181. doi: 10.1037/rep0000224]. 109. Capozza K, Woolsey S, Georgsson M, Black J, Bello N, Lence C, Oostema S, North C. Going Mobile With Diabetes Support: A Randomized Study of a Text Message–Based Personalized Behavioral Intervention for Type 2 Diabetes Self-Care. Diabetes Spectrum May 2015, 28 (2) 83-91; DOI: 10.2337/diaspect.28.2.83]. 66 CITATION AND NOTES | HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH 110. Fottrell E, Ahmed N, Morrison J, Kuddus A, Shaha SK, King C, Jennings H, Akter K, Nahar T, Haghparast-Bidgoli H, Khan A,Costello A, Azad K. Community groups or mobile phone messaging to prevent and control type 2 diabetes and intermediate hyperglycaemia in Bangladesh (DMagic): a cluster-randomised controlled trial. Lancet Diabetes Endocrinol. 2019 Mar; 7(3): 200–212. doi: 10.1016/S2213-8587(19)30001-4]. 111 WHO, 2003. A public health approach to antiretroviral treatment: overcoming constraints. 112 Leotsakos A, Zheng H, Croteau R, Loeb JM, Sherman H, Hoffman C, et al. Standardization in patient safety: the WHO High 5s project. International Journal for Quality in Health Care. 2014;26(2):109-16. doi: 10.1093/intqhc/mzu010 113 WHO, 2010. Package of Essential Noncommunicable (PEN) Disease Interventions for Primary Health Care in Low-Resource Settings. 114 Key informant interviews and NCDC Operational Plan 2017-2022 115 Kane J, Landes M, Carroll C, Nolen A, Sodhi S. A systematic review of primary care models for non- communicable disease interventions in Sub-Saharan Africa. BMC family practice. 2017;18(1):46. Epub 2017/03/24. doi: 10.1186/s12875-017-0613-5. 116 The second pilot was a NCD Prevention Project piloted by the Japan International Cooperation Agency that included a NCD screening model, a health guidance model for lifestyle education, a health promotion model and a social marketing model that disseminates information on the importance of NCD prevention and invite people to attend NCD services. The third pilot was the National Initiative to Reinforce and Organize General Diabetes Care in Sri Lanka (NIROGI Lanka) of the Sri Lanka Medical Association that included recruiting and training diabetes nurse educators, training technicians to provide support for diabetic foot, working with a shoe manufacturer to develop diabetes-specific protective footwear, and community empowerment and family participation through health-promotion activities carried out by trained volunteers in work places, schools and community groups. 117 Joshi R, Alim M, Kengne AP, Jan S, Maulik PK, et al. (2014) Task Shifting for Non-Communicable Disease Management in Low and Middle Income Countries – A Systematic Review. PLoS ONE 9(8): e103754. doi:10.1371/journal.pone.0103754 118 Lewin S, Munabi-Babigumira S, Glenton C, Daniels K, Bosch-Capblanch X, et al. (2010) The effect of lay health workers on mother and child health and infectious diseases. Cochrane Database of Systematic Reviews: No.: CD004015. DOI:004010.001002/14651858 119 World Bank, 2014. Evaluation of Interventions to Increase the proportion of PLHIV who are diagnosed, initiated on, adhering to and being retained in HIV treatment in South Africa. Formative Qualitative Research: Phase 1 Report 120 Joshi R, Alim M, Kengne AP, Jan S, Maulik PK, Peiris D, et al. Task shifting for non-communicable disease management in low and middle income countries--a systematic review. PloS one. 2014;9(8):e103754. Epub 2014/08/15. doi: 10.1371/journal.pone.0103754. 121 Some D, Edwards JK, Reid T, Van den Bergh R, Kosgei RJ, Wilkinson E, et al. Task Shifting the Management of Non-Communicable Diseases to Nurses in Kibera, Kenya: Does It Work? PloS one. 2016;11(1):e0145634. Epub 2016/01/27. doi: 10.1371/journal.pone.0145634 122 Venables E, Edwards JK, Baert S, Etienne W, Khabala K, Bygrave H. "They just come, pick and go." The Acceptability of Integrated Medication Adherence Clubs for HIV and Non Communicable Disease (NCD) Patients in Kibera, Kenya. PloS one. 2016;11(10):e0164634. Epub 2016/10/21. doi: 10.1371/journal.pone.0164634. 123 Tapela et al, 2017. Treatment of non-communicable disease in rural resource-constrained settings: a comprehensive, integrated, nurse-led care model at public facilities in Rwanda. The Lancet Global Health, Volume 3, S36. 124 Rattanavipapong et al, 2016. One Step Back, Two Steps Forward: An Economic Evaluation of the PEN Program in Indonesia. Health Systems & Reform, 2(1):84–98. 125 WHO, 2003. A public health approach to antiretroviral treatment: overcoming constraints. 126 White, 1996. Depoliticising development. Development in Practice 1996;6(1). 127 Fisher EB, Boothroyd RI, Coufal MM, Baumann LC, Mbanya JC, Rotheram-Borus MJ, et al. Peer support for self- management of diabetes improved outcomes in international settings. Health affairs (Project Hope). 2012;31(1):130-9. Epub 2012/01/11. doi: 10.1377/hlthaff.2011.0914 128 Bloom, G., Kanjilal, B., Lucas, H., & Peters, D. H. (2012). Transforming health markets in Asia and Africa: Improving quality and access for the poor. Taylor and Francis. DOI: 10.4324/9780203102060 129 Taniguchi D, LoGerfo J, van Pelt M, Mielcarek B, Huster K, Haider M, et al. Evaluation of a multi-faceted diabetes care program including community-based peer educators in Takeo province, Cambodia, 2007-2013. PloS one. 2017;12(9):e0181582. Epub 2017/09/26. doi: 10.1371/journal.pone.0181582. 130 Taniguchi D, LoGerfo J, van Pelt M, Mielcarek B, Huster K, Haider M, et al. Evaluation of a multi-faceted diabetes care program including community-based peer educators in Takeo province, Cambodia, 2007-2013. PloS one. 2017;12(9):e0181582. Epub 2017/09/26. doi: 10.1371/journal.pone.0181582. 67 HYPERTENSION AND TYPE-2 DIABETES IN BANGLADESH | CITATION AND NOTES 131 Ama de-Graft Aikens and Charles Agyemang, 2015. Book. Chronic Non-Communicable Diseases in low- and middle-income countries. 132 Grady PA, Gough LL. Self-management: a comprehensive approach to management of chronic conditions. American journal of public health. 2014;104(8):e25-31. Epub 2014/06/13. doi: 10.2105/ajph.2014.302041 133 Grady PA, Gough LL. Self-management: a comprehensive approach to management of chronic conditions. American journal of public health. 2014;104(8):e25-31. Epub 2014/06/13. doi: 10.2105/ajph.2014.302041 134 WHO, 2008. Caring for people with chronic conditions: A health system perspective 135 Islam SM, Niessen LW, Seissler J, Ferrari U, Biswas T, Islam A, et al. Diabetes knowledge and glycemic control among patients with type 2 diabetes in Bangladesh. SpringerPlus. 2015;4:284. Epub 2015/06/24. doi: 10.1186/s40064-015-1103-7. 136 WHO, 2002. Innovative Care for Chronic Conditions. Building Blocks for Action. 137 WHO, 2008. Caring for people with chronic conditions: A health system perspective 138 PAHO. Factsheet: Passport Care, 2012. http://www.paho.org/hq/index.php?option=com_docman&task=doc_details&gid=17514&Itemid=1639&lan g=en 139 Vas A, Devi ES, Vidyasagar S, Acharya R, Rau NR, George A, et al. Effectiveness of self-management programmes in diabetes management: A systematic review. International journal of nursing practice. 2017;23(5). Epub 2017/08/02. doi: 10.1111/ijn.12571. 140 Related to the ICDM and also implemented in South Africa is the “Integrated Clinical Services Management (ICSM)” model. It is defined as a health system strengthening model that builds on the strengths of the HIV programme to deliver integrated care to patients with chronic and/or acute diseases or who came for preventative services by taking a patient-centric view that encompasses the full value chain of continuum of care and support. 141 Wagner E, Austin B, Davis C, Hindmarsh M, Schaefer J, Bonomi A. Improving chronic illness care: translating evidence into action. Health Affairs 2001;20(6):64–78. 142 World Health Organisation. Innovative care for chronic conditions: building blocks for action. Global Report. Non-communicable Diseases and Mental Health. Geneva: World Health Organisation, Report No.: ISBN 92 4 159 017 3; 2002. 143 Mahomed OH and Asmall S (2015). Development and implementation of an integrated chronic disease model in South Africa: lessons in the management of change through improving the quality of clinical practice, Int J Integr Care; Oct–Dec. 144 NDOH (2014). Integrated chronic disease management: A step-by-step manual to guide implementation. http://www.kznhealth.gov.za/family/Integrated-chronic-disease-management-manual.pdf 145 Folb N (2017). Non-communicable diseases in public sector primary care clinics in South Africa: multimorbidity, control, treatment, socioeconomic associations, and evaluation of educational outreach with a clinical management tool. Thesis, March 2017. 146 Mahomed, O.H. & Asmall, S., 2017, ‘Professional nurses’ perceptions and experiences with the implementation of an integrated chronic care model at primary healthcare clinics in South Africa’, Curationis 40(1), a1708. 147 www.differentiatedcare.org 148 http://www.searo.who.int/bangladesh/stepssurvey/en/ 68