ict-enabled monitoring and evaluation methods of

ICT-ENABLED MONITORING AND EVALUATION METHODS OF HEALTH COVERAGE FOR SUB-SAHARAN HEALTH FACILITIES Universal Health Coverage (UHC) is at the center of the Sustainable Development Agenda 2030. During the period 2010–2016, the PhD researcher made an evaluation of the indicators allowing quantification of the impact of health coverage schemes on patients, applied in 6 subSaharan countries: Rwanda, Burundi, DRC, Tanzania, Mali and Senegal. After an analysis of potential health coverage indicators, the most relevant ones were calculated on the basis of patient health records collected via OpenClinic GA, a health management information system (HMIS) used in several sub-Saharan health facilities (HF). Results of this research show that DRC and Malian health facilities (HF) have a level of patient coverage lower than those in Rwanda and Burundi, where the patient is covered by more solidarity-based health insurance

ICT-ENABLED MONITORING AND EVALUATION METHODS OF HEALTH COVERAGE FOR SUB-SAHARAN HEALTH FACILITIES

Gustave Karara

ICT-ENABLED MONITORING AND EVALUATION METHODS OF HEALTH COVERAGE FOR SUB-SAHARAN HEALTH FACILITIES A field study in Rwanda, Burundi, DRC, Tanzania, Mali and Senegal Thesis submitted in fulfilment of the requirements for the degree of Doctor in Medical Sciences (PhD)

Gustave Karara Academic year 2018-2019 Promoters: Prof. Dr. Ronald Buyl, Vrije Universiteit Brussel Prof. Dr. Marc Nyssen, Vrije Universiteit Brussel

schemes. This research also managed to estimate the average costs of health insurance expenditure per patient and pathology related financial burdens in the studied sub-Saharan HF. More efforts are needed to achieve a good financial protection of patients in sub-Saharan HF. Only Rwandan HF had reached the indicators recommended by WHO thanks to the high reimbursement rates of CBHI (Mutuelles de santé). The research contributed to establish relevant parameters for assessing the level of health coverage in sub-Saharan HF using ICT-HMIS patient routine data. Additional broader studies involving more HF are needed in order to draw further conclusions on the role of health insurance schemes in health coverage in sub-Saharan countries. The same methods could also be applied to all health facilities in developing countries through implementation of an adequate ICT infrastructure for health information management.

9

789057

Gustave Karara

W W W.V U B P R E S S.B E

DEPARTMENT OF PUBLIC HEALTH UNIT OF BIOSTATISTICS AND MEDICAL INFORMATICS

188244

I S B N 9 7 8 9 0 5 71 8 8 2 4 4

Gustave_Karara_omslag.indd 1

3/10/18 11:35

FACULTY OF MEDICINE AND PHARMACY DEPARTMENT OF PUBLIC HEALTH, UNIT OF BIOSTATISTICS AND MEDICAL INFORMATICS

ICT-enabled monitoring and evaluation methods of health coverage for sub-Saharan health facilities A field study in Rwanda, Burundi, DRC, Tanzania, Mali and Senegal Thesis submitted in fulfilment of the requirements for the degree of Doctor in Medical Sciences (PhD)

Gustave Karara

October 2018

Gustave_Karara_def.indd 1

3/10/18 08:10

© Gustave Karara 2018 Uitgeverij VUBPRESS Brussels University Press VUBPRESS is an imprint of ASP nv (Academic & Scientific Publishers nv) Keizerslaan 34 B-1000 Brussels Tel. +32 (0)2 289 26 50 Fax +32 (0)2 289 26 59 E-mail: [email protected] www.aspeditions.be ISBN 978 90 5718 824 4 NUR 870 Legal deposit D/2018/11.161/110 All rights reserved. No parts of this book may be reproduced or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the author.


3/10/18 08:10

ICT-enabled monitoring and evaluation methods of health coverage for sub-Saharan health facilities 2018, Gustave Karara All rights reserved. No parts of this work may be reproduced in any form or by any means graphic, electronic, or mechanical, including photocopying, recording, taping, or information storage and retrieval systems without the written permission of the publisher. Products that are referred to in this document may be either trademarks and/or registered trademarks of the respective owners. The publisher and the author make no claim to these trademarks. While every precaution has been taken in the preparation of this document, the publisher and the author assume no responsibility for errors or omissions, or for damages resulting from the use of information contained in this document or from the use of programs and source code that may accompany it. In no event shall the publisher and the author be liable for any loss of profit or any other commercial damage caused or alleged to have been caused directly or indirectly by this document. Printed: October 2018 in Belgium

Author Gustave Karara Promoters Prof.Dr. Ronald Buyl Prof.Em.Dr.Ir. Marc Nyssen Consultant Dr. Frank Verbek e

Special thanks to: All the people who contributed to the research processes that have led to this document. To my colleagues in Belgium and in Africa. Many thank s to Frank Verbek e for his support and orientation throughout this PhD study. Special thank s to Marc Nyssen and Ronald Buyl for their patience and their very fruitful advice for my research. I am most grateful to my wife, Julienne Umuhoza and to my k ids Cyusa, Byusa and Gaju for having so bravely survived my long absences at home in the past 8 years. I also thank my parents, sisters and brothers for their encouragement and compassion.

Production VUB Press


3/10/18 12:46

JURY

Chairman: Prof. Kurt Barbe, Vrije Universiteit Brussel Promotors: Prof. Ronald Buyl, Vrije Universiteit Brussel Prof. Em. Marc Nyssen, Vrije Universiteit Brussel Members: Prof. Koen Putman, Vrije Universiteit Brussel Prof. Frederik Questier, Vrije Universiteit Brussel Dr. Célia Boyer, Health On the Net Foundation, Geneva Prof. Bruno Meessen, Instituut voor Tropische Geneeskunde Antwerpen


3/10/18 12:46

Contents

5

Table of Contents Foreword

0

Part I Preface

12

Part II Introduction

15

1 Universal ................................................................................................................................... Health Coverage 16 ................................................................................................................................... 18 2 UHC indicators monitoring Health services ......................................................................................................................................................... coverage 18 Health financial protection ......................................................................................................................................................... 19

3 UHC ................................................................................................................................... monitoring challenges 21 Household-based ......................................................................................................................................................... surveys 21 Health facility data sources ......................................................................................................................................................... 22

4 Facility-based ................................................................................................................................... HMIS in sub-Saharan health facilities 22 Facility-based ......................................................................................................................................................... HMIS w eaknesses 22 Health coverage m onitoring challenges ......................................................................................................................................................... 24

5 ICT................................................................................................................................... innovations in health facilities 24

Part III Research hypothesis, limitations and study plan

27

Part IV Materials and methods

30

1 Activity ................................................................................................................................... 1: Analysis of UHC situation in sub-Saharan countries 31 ................................................................................................................................... 32 2 Activity 2: Define a set of health service metrics for health coverage monitoring ................................................................................................................................... 32 3 Activity 3: Identify RHSC for health facilities ................................................................................................................................... 33 4 Activity 4: Integrate RHSC in the ICT-HMIS tool ................................................................................................................................... 34 5 Activity 5: Implement of the ICT-HMIS in health facilities ................................................................................................................................... 34 6 Activity 6: Evaluate the health coverage in health facilities Health coverage ......................................................................................................................................................... indicators in sub-Saharan health facilities 35 Com parison on health coverage indicators in sub-Saharan health facilities ......................................................................................................................................................... 35

Part V Results

37

1 Activity ................................................................................................................................... 1: Analysis of UHC situation in sub-Saharan countries 37 Health expenditure ......................................................................................................................................................... and financial risk protection 37 Health services coverage indicators ......................................................................................................................................................... 39 Situation of health insurance schem es ......................................................................................................................................................... 40 Rw anda .................................................................................................................................................. 41 Burundi .................................................................................................................................................. 42 DRC .................................................................................................................................................. 42 Tanzania .................................................................................................................................................. 43 Mali .................................................................................................................................................. 44 Senegal .................................................................................................................................................. 44 In summary .................................................................................................................................................. 45

2 Activity ................................................................................................................................... 2: Define a set of health service metrics for health coverage monitoring 46

2018, Gustave Karara


3/10/18 08:10

6

ICT-enabled monitoring and evaluation methods of health facilities for sub-Saharan health facilities

......................................................................................................................................................... 46 Health coverage indicators for health facilities Health insurance coverage .................................................................................................................................................. 47 Financial protection of patients .................................................................................................................................................. 47 Health coverage m etrics in health facilities ......................................................................................................................................................... 48 Health facilities utilization .................................................................................................................................................. 48 Health facility identification ........................................................................................................................................... 48 Patient identification ........................................................................................................................................... 49 Admission,........................................................................................................................................... Discharge and Transfer (ADT) 49 Useful monitoring metrics ........................................................................................................................................... 50 Health services coverage .................................................................................................................................................. 52 Health insurance schemes ........................................................................................................................................... 52 Health insurance plans ........................................................................................................................................... 53 Useful monitoring metrics ........................................................................................................................................... 53 Health services burden .................................................................................................................................................. 54 Health service description ........................................................................................................................................... 55 Health services consumed ........................................................................................................................................... 55 Useful monitoring metrics ........................................................................................................................................... 56 Morbidity and mortality in health facilities .................................................................................................................................................. 57 Patient morbidity rate ........................................................................................................................................... 57 Patient mortality rate ........................................................................................................................................... 58 Useful monitoring metrics ........................................................................................................................................... 58 Health facility management .................................................................................................................................................. 59 Health services revenue ........................................................................................................................................... 59 Health services payment ........................................................................................................................................... 60 Useful monitoring metrics ........................................................................................................................................... 60 In summary .................................................................................................................................................. 61

3 Activity ................................................................................................................................... 3: Identify RHSC for health facilities 62 International ......................................................................................................................................................... codifications in health 63 Diseases classification .................................................................................................................................................. 63 International Classification of Diseases (ICD) ........................................................................................................................................... 63 International Classification of Primary care (ICPC) ........................................................................................................................................... 64 Diagnostic ........................................................................................................................................... and Statistical Manual of Mental Disorders (DSM) 65 Logical Observation Identifier Names and Codes (LOINC) .................................................................................................................................................. 66 Anatomical, Therapeutic and Chemical Classification System (ATC) .................................................................................................................................................. 66 Medical procedures classification .................................................................................................................................................. 67 RHSC for health facilities ......................................................................................................................................................... 68 KHIRI Pathology Grouping Set (KPGS) .................................................................................................................................................. 68 LOINC codification .................................................................................................................................................. 69 ATC codification .................................................................................................................................................. 70 Generic care delivery classification (GCD) .................................................................................................................................................. 70

4 Activity ................................................................................................................................... 4: Integration of RHSC in the ICT-HMIS tool 72 Health facility ......................................................................................................................................................... adm inistration 73 Health facility identification .................................................................................................................................................. 73 Health facility departments configuration .................................................................................................................................................. 74 Patient adm inistration ......................................................................................................................................................... 76 Patient identification .................................................................................................................................................. 76 Patient demographic data .................................................................................................................................................. 77 Patient health insurer configuration .................................................................................................................................................. 78 Adm ission, Discharge and Transfer ......................................................................................................................................................... 80 Outpatient.................................................................................................................................................. registration 80 Inpatient registration .................................................................................................................................................. 81 Transfer .................................................................................................................................................. 82 Discharge.................................................................................................................................................. 82 Financial m anagem ent ......................................................................................................................................................... 83



3/10/18 08:10

Contents

7

.................................................................................................................................................. 83 Health services Patient invoice .................................................................................................................................................. 84 Payment and cash transfers .................................................................................................................................................. 86 Insurer invoice .................................................................................................................................................. 87 Medical record ......................................................................................................................................................... 89 Record identification .................................................................................................................................................. 89 History of .................................................................................................................................................. health events 89 Generic SOAP structure .................................................................................................................................................. 90 Health registration forms .................................................................................................................................................. 91 Medical prescription management .................................................................................................................................................. 93 Diagnosis .................................................................................................................................................. classifications 94 Warnings .................................................................................................................................................. 95 Vaccination management .................................................................................................................................................. 96 Pharm acy......................................................................................................................................................... m anagem ent 96 Laboratory data m anagem ent ......................................................................................................................................................... 97 Laboratory.................................................................................................................................................. analysis 97 Order entry .................................................................................................................................................. 99 Results entry .................................................................................................................................................. 99 Medical im aging data m anagem ent ......................................................................................................................................................... 100 Order entry .................................................................................................................................................. 100 Results entry .................................................................................................................................................. 101 Global health barom eter system ......................................................................................................................................................... 102

5 Activity ................................................................................................................................... 5: Implementation of the ICT-HMIS in health facilities 104 Rw andan......................................................................................................................................................... Health Facilities 106 Reference hospitals .................................................................................................................................................. 107 University........................................................................................................................................... Teaching Hospital of Kigali (CHUK) 107 Neuro-Psychiatric Hospital - Caraes Ndera (NPH-CN) ........................................................................................................................................... 110 District hospitals .................................................................................................................................................. 111 Nyamata District Hospital (NYDH) ........................................................................................................................................... 111 Rw amagana District Hospital (RWDH) ........................................................................................................................................... 113 Gihundw e........................................................................................................................................... District Hospital (GIDH) 115 Butaro District Hospital (BUDH) ........................................................................................................................................... 116 Private health facilities .................................................................................................................................................. 117 Croix du Sud hospital (CDS) ........................................................................................................................................... 118 La Médicale Polyclinic (LMED) ........................................................................................................................................... 119 Biomedical........................................................................................................................................... Center (BMC) and KHIBD Dental Center (DC) 121 Burundian Health Facilities ......................................................................................................................................................... 123 Reference hospitals .................................................................................................................................................. 123 Military Hospital of Kamenge (HMK) ........................................................................................................................................... 124 Prince Louis Rw agasore Clinic (CPLR) ........................................................................................................................................... 125 University........................................................................................................................................... Teaching Hospital of Kamenge (CHURK) 127 Prince Regent Charles Hospital (HPRC) ........................................................................................................................................... 128 District hospitals .................................................................................................................................................. 129 Ngozi Regional Hospital (NGORH) ........................................................................................................................................... 130 Muramvya........................................................................................................................................... District Hospital (MUDH) 131 Kirundo District Hospital (KIDH) ........................................................................................................................................... 132 Private hospital .................................................................................................................................................. 134 Centre Medico-Chirurgical de Kinindo (CMCK) ........................................................................................................................................... 134 DRC Health Facilities ......................................................................................................................................................... 136 University.................................................................................................................................................. Teaching Hospital of Lubumbashi (CUL) 136 Provincial.................................................................................................................................................. Reference Hospital of Bukavu (HPGRB) 138 Malian Health Facilities ......................................................................................................................................................... 139 University Teaching Hospital-African Institute of Tropical Ophthalmology (IOTA) .................................................................................................................................................. 140



3/10/18 08:10

8


.................................................................................................................................................. 141 Nianankoro Fomba Hospital of Segou (HNFS) Reference Health Center Commune II (CSREF2) .................................................................................................................................................. 143 Reference Health Center Commune III (CSREF3) .................................................................................................................................................. 144 Reference Health Center Commune IV (CSREF4) .................................................................................................................................................. 146 Senegalese Health facilities ......................................................................................................................................................... 147 Foundiougne (CSFOU) and Passy (CSPAS) health Centers .................................................................................................................................................. 148 Tanzanian health facility ......................................................................................................................................................... 149 CCBRT hospital .................................................................................................................................................. 149

6 Activity ................................................................................................................................... 6: Evaluation of the health coverage in health facilities 151 Health coverage ......................................................................................................................................................... in sub-Saharan health facilities 151 Rw andan.................................................................................................................................................. health facilities 152 Patient health insurance coverage (PHIC) ........................................................................................................................................... 152 Patient health services coverage (PHSC) ........................................................................................................................................... 153 Patient health services payment rate (PHSP) ........................................................................................................................................... 155 Patient out-of-pocket payment (POOP) ........................................................................................................................................... 156 Burundian health facilities .................................................................................................................................................. 158 Patient health insurance coverage (PHIC) ........................................................................................................................................... 158 Patient health services coverage (PHSC) ........................................................................................................................................... 159 Patient health services payment rate (PHSP) ........................................................................................................................................... 161 Patient out-of-pocket payment (POOP) ........................................................................................................................................... 162 DRC health facilities .................................................................................................................................................. 163 Patient health insurance coverage (PHIC) ........................................................................................................................................... 163 Patient health services coverage (PHSC) ........................................................................................................................................... 164 Patient health services payment rate (PHSP) ........................................................................................................................................... 165 Patient out-of-pocket payment (POOP) ........................................................................................................................................... 165 Malian health facilities .................................................................................................................................................. 165 Patient health insurance coverage (PHIC) ........................................................................................................................................... 166 Patient health services coverage (PHSC) ........................................................................................................................................... 166 Patient health services payment rate (PHSP) ........................................................................................................................................... 167 Patient out-of-pocket payment (POOP) ........................................................................................................................................... 167 Senegalese health facilities .................................................................................................................................................. 168 Comparison in third reference hospitals .................................................................................................................................................. 169 Health insurance expenditure in Rw andan and Burundian health facilities ......................................................................................................................................................... 170 Rw andan.................................................................................................................................................. health facilities 170 Burundian health facilities .................................................................................................................................................. 173 Case Mix......................................................................................................................................................... evaluation in three hospitals of Rw anda, Burundi and DRC 175 Main KPGS-pathology groups monitored in the three hospitals .................................................................................................................................................. 175 Morbidity .................................................................................................................................................. and Mortality per pathology group in CHUK 177 Morbidity ........................................................................................................................................... per pathology group in CHUK 177 Mortality per pathology group in CHUK ........................................................................................................................................... 179 Financial .................................................................................................................................................. burden per pathology group in the three hospitals 182 Health services coverage per pathology group in the three hospitals .................................................................................................................................................. 186 Measurem ent of health coverage im pact in health facilities ......................................................................................................................................................... 188 Patient demographic data .................................................................................................................................................. 188 Health facility utilization metrics .................................................................................................................................................. 191 Average number of outpatient encounters (ANOE) ........................................................................................................................................... 191 Average rate of re-admission (ARRA) ........................................................................................................................................... 192 Average length of stay (ALOS) ........................................................................................................................................... 193 Health facility revenue metrics .................................................................................................................................................. 194 Health facility reimbursement metrics .................................................................................................................................................. 197

Part VI Discussion

200

1 Health ................................................................................................................................... coverage indicators for health facilities 200



3/10/18 08:10

Contents

9

2 Health ................................................................................................................................... services coverage metrics 202 ................................................................................................................................... 203 3 Health services classification issues ................................................................................................................................... 204 4 Health services coverage in health facilities Rw andan......................................................................................................................................................... health facilities 205 Burundian health facilities ......................................................................................................................................................... 205 DRC health facilities ......................................................................................................................................................... 206 Malian health facilities ......................................................................................................................................................... 207 Case Mix......................................................................................................................................................... in three reference hospitals 207

5 Impact ................................................................................................................................... of OpenClinic GA implementation 210 ................................................................................................................................... 210 6 Lessons learned from OpenClinic GA projects

Part VII Conclusion

213

Part VIII Further research

215

Part IX Publications

218

Part X Bibliography

221

Part XI Annexes

231

1 Annex ................................................................................................................................... 0: List of tables and figures 231 ................................................................................................................................... 234 2 Annex 1: List of used acronyms ................................................................................................................................... 237 3 Annex 2: List of sub-Saharan health facilities using OpenClinic GA ................................................................................................................................... 238 4 Annex 3: Mapping between KPGS codes and ICD-10/ICPC-2 ................................................................................................................................... 240 5 Annex 4: LOINC codes used in district hospital ................................................................................................................................... 241 6 Annex 5: List of top 5 health insurance plans in health facilities ................................................................................................................................... 243 7 Annex 6: List of health services Annex 6.1: ......................................................................................................................................................... List of 20 health services at CHUK 243 Annex 6.2: List of health services at CSFOU ......................................................................................................................................................... 243

8 Annex ................................................................................................................................... 7: CAP Survey questionnaire in OpenClinic GA utilization 244 ................................................................................................................................... 247 9 Annex 8: Used health expenditures definitions ................................................................................................................................... 248 10 Annex 9: Used statistics and data OpenClinic ......................................................................................................................................................... GA statistics data 248 Insurer and invoicing statistics .................................................................................................................................................. 248 Financial .................................................................................................................................................. metrics 250 Department statistics .................................................................................................................................................. 251 Global hospital report .................................................................................................................................................. 252 Global distribution of pathologies .................................................................................................................................................. 255 GHB Statistics data ......................................................................................................................................................... 257 Table sizes .................................................................................................................................................. 258 Financial .................................................................................................................................................. data 259 Diagnostics .................................................................................................................................................. 260 SQL queries on databases ......................................................................................................................................................... 260 Insured and uninsured patients .................................................................................................................................................. 260 Insured and uninsured patient encounters .................................................................................................................................................. 262 Insured and uninsured new out/in-patients .................................................................................................................................................. 264 Lenght of.................................................................................................................................................. stay among insured and uninsured patients 266



3/10/18 08:10

10


268 Length of.................................................................................................................................................. stay per insurer among insured and uninsured patients Financial .................................................................................................................................................. burden and coverage per pathology group 271 Num ber of patients and encounters ......................................................................................................................................................... 273 Number of patients .................................................................................................................................................. 273 Number of outpatients .................................................................................................................................................. 277 Number of outpatient encounters .................................................................................................................................................. 279 Number of inpatients .................................................................................................................................................. 280 Number of inpatient encounters .................................................................................................................................................. 282

11 Annex ................................................................................................................................... 10: OpenClinic GA system 283 OpenIT-MIA ......................................................................................................................................................... 284 Open source softw are .................................................................................................................................................. 285 Web-based user interface .................................................................................................................................................. 286 GEHR m odel ......................................................................................................................................................... 287 GEHR architecture .................................................................................................................................................. 287 Application domain .................................................................................................................................................. 288 OpenClinic GA standard features ......................................................................................................................................................... 288 OpenClinic GA features .................................................................................................................................................. 288 OpenClinic GA modules .................................................................................................................................................. 289 OpenClinic GA im plem entation process ......................................................................................................................................................... 290 Openclinic GA technical requirements .................................................................................................................................................. 290 OpenClinic GA setup and configuration .................................................................................................................................................. 291 Users training and follow -up .................................................................................................................................................. 292 OpenClinic GA security setup ......................................................................................................................................................... 293 Access control .................................................................................................................................................. 293 Access monitoring .................................................................................................................................................. 294 Information archiving .................................................................................................................................................. 294 OpenClinic GA rem otely m onitoring setup ......................................................................................................................................................... 295 E-mailing .................................................................................................................................................. group and OpenTicket 295 OpenVPN.................................................................................................................................................. server 296

Part XII Executive summary

298

Index

0



3/10/18 08:10

ICT-enabled m onitoring and evaluation m ethods of health coverage for sub-Saharan health facilities

Part


I

3/10/18 08:10

12

1

ICT-enabled monitoring and evaluation methods of health coverage for sub-Saharan health facilities

Preface After my studies in medicine and public health, I immediately started to acquire professional experiences in Belgium as: Administrative and financial manager in a hospital Deputy Investigator at the Institute of Public Health ICT trainer in a social non-profit organization Information and Communication Technology (ICT)! This area had interested me since I began teaching it to African health professionals who came to study a master in public health at my university. The computer tool that we found easily accessible in Europe was a "rare commodity" for African people. I had the chance to own my first computer at the beginning of my medical studies. A few years later, I began to improve my ICT knowledge and applied it to the health sector. Although ICT is not a panacea for the health sector, it has gradually become an indispensable tool to improve the quality of health care for both developed and developing countries. Homecoming In 2001, I returned for the first time in my country of birth: Rwanda. As a consultant to the Ministry of Health (MoH) through the MIDA-GL (Migration for Development in Africa-Great Lakes) program. The country I had known during the genocide and the war in 1994 had completely changed. The population was back, the health facilities were working, new buildings were coming out of the ground, the country was reborn. I was involved in the strategy development of community based health insurance schemes (Mutuelles de santé). A few years later, I started supporting the public hospitals in the city of Kigali to develop a health information system (HIS) for the management of patient records. It was then necessary to propose ICT-HIS solutions that could support in the management of the patient records. It was during these missions in Rwanda I met the senior developer of the OpenClinic GA software, at the CHUK (University Teaching Hospital of Kigali) where the solution has been implemented for the first time in sub-Saharan health facility. OpenClinic GA project The OpenClinic GA project was initiated by a Belgian physician. I quickly joined the project at its beginnings. The project aimed to improve health information management by using ICT methods and implementing an integrated hospital information management system, OpenClinic GA software. This ICT-HIS solution is the result of a combined expertise of computer engineers, specialists in public health, statisticians and physicians of the VUB (Vrije Universiteit Brussel) with a high-level training in ICT. The pilot project began at the CHUK. Very soon, several Rwandan health facilities, both public and private, joined the project. Later, several health facilities in sub-Saharan region interested by the results of the project, adopted OpenClinic GA software. My research For several years, I had been involved in installing, customizing, configuring and training activities of OpenClinic GA software in several health facilities in Rwanda, Burundi, DRC, Mali, Tanzania, Senegal, Congo-Brazzaville and Gabon. From these activities, humanity has grown up: patience and perseverance became the watchword. And as the coordinator of different projects in these countries, the acquired expertise has pushed me to improve more and more the quality in the implementation of projects. These implementation activities have been combined with research activities in the field of health informatics. Three important objectives were formulated for the research project: Improve the quality and efficiency of the management of health facilities Structure and facilitate the encoding of patient data in the system Facilitate the extraction and analysis of health indicators for reporting. The patient data information collected by health facilities for management has been secondarily used



3/10/18 08:10

Preface

13

for monitoring of universal health coverage, in my research. Thus, the extracted health indicators are useful to: The managers of the health facilities for the improvement and planning, The public authorities to guide health and social policy, The partners in the health development and universities to guide their support and research. OpenClinic GA projects have proved to be a great opportunity to combine the experiences gained in public health and 7-years of study in health informatics from 2010 till 2017. The project was nominated for the Digital for Development (D4D) award by the Belgian Cooperation in 2016. Lessons learned Conducting ICT projects in Africa was not only an academic experience, it was also a life lessons experience. We often think African people have one culture, it is one people. The research work all over the continent has shown that there are different ways of thinking and reacting depending on the context of each country, each health facility. The Openclinic project has not been implemented using a standard method in all sub-Saharan health facilities, the implementation was not a copy / paste! First of all, it was necessary to understand the daily life of the healthcare professionals who faced several challenges: lacks of medicines, medical equipment, water, electricity in their structures, perpetual armed conflicts in certain regions, etc. I had to take time to know them better, beyond emotionality. I had to have patience to find the best method to push them to adopt the new patient management system chosen by the management committee. I learned to be persistent especially when it was necessary to accompany users whose level of knowledge in ICT was different. Better understanding the local problems has enabled me to carry out the successful research: from nothing to operational ICT system, running autonomously and producing results, leading to improve patient treatment, hospital management and producing metrics for research. The main purpose of the OpenClinic project was not to provide a miracle solution for better patient records management, but first and foremost to get healthcare staff understanding that ICT could solve a part of management problems encountered and appropriating the new ICT system which will gradually become their daily tool of work. Then, I had to share with them the results of their work, self-evaluate, and produce reports automatically. The results should soon appear and especially be shared among colleagues. The sharing of information that was at first inconceivable gradually became a reality. Finally, it was necessary to avoid limiting ourselves to the figures, because behind every statistic there was a health service provided to the patient who was remaining at the center of our preoccupation.



3/10/18 08:10


Part


II

3/10/18 08:10

Introduction

2

15

Introduction Since the 58th World Health Assembly (WHA) resolution in May 2005, Universal Health Coverage (UHC) has become a major point of attention, gradually integrated into health policies of countries. A growing focus on the goal of universal coverage in health systems appeared. Indeed, the WHA (2005) has urged all World Health Organization (WHO)'s member states to aim for affordable universal coverage and access for all citizens on the basis of equity and solidarity. UHC means that all people receive essential health services they need in quality and without being exposed to financial hardship. The WHA resolution WHA58.33 (WHA, 2005) focused on "Sustainable health financing, universal coverage and social health insurance". This resolution was part of a global vision in the effort to reduce social inequities and the fight against poverty advocated by the United Nations Millennium Development Goals (MDGs). This vision was reinforced by the United Nations Sustainable Development Summit in September 2015 where world leaders adopted the 2030 Agenda for Sustainable Development. A set of 17 Sustainable Development Goals (SDGs) was defined (UNDP, 2015). The aim of SDGs is end poverty, fight inequality and injustice, and tackle climate change by 2030. The third Goal of SDGs concerns the "Good health and well-being" (http://www.un.org/ sustainabledevelopment/health/). It includes a bold commitment to achieve universal health coverage by 2030. Moving towards UHC is a dynamic, continuous and specific process for each country. UHC has a direct impact on a population’s health and better health contributes to the human development. Access to health services enables people to be more productive and active contributors to their families and communities (WHO, 2012-2013). Many African countries still struggle to deliver quality and affordable health services and too many households across the continent are forced to borrow money or sell assets to pay for health care (Matshidiso, 2015) (Susann and Iandry, 2015). For coverage of several basic health services (including services for HIV, tuberculosis, malaria, non-communicable diseases, maternal and child health) and improved sanitation (hygiene), sub-Saharan Africa lags well behind the rest of the world. The region accounts for approximately 25% of the world’s disease burden, yet it has just 3% of its doctors (Matshidiso, 2015). Some sub-Saharan Africa countries make remarkable efforts to move towards UHC. In Ghana, a taxfunded national health insurance system, known as the National Health Insurance Scheme, covers 95% of diseases that affect Ghanaians, enabling financial protection and expanding coverage (Matshidiso, 2015). By implementing ambitious reforms that started in 2000, with the goal of universal health coverage, Rwanda currently sustains one of the most elaborate health insurance schemes, the Community Based Health Insurance (CBHI), which covers over 90% of the population (MoH Rwanda, 2010). These systems are examples of the types of reform that can help African countries minimize catastrophic out-of-pocket health care costs that exacerbate poverty. Discussions are still ongoing on how the progress towards UHC in countries and globally can be measured. Some indicators related to health services and financial protection have been defined. There is agreement on the range of indicators currently used to measure the health MDGs, some are tracking service coverage (e.g. deliveries attended in health facilities) and some are showing the impact of this (e.g. maternity mortality rates). There is also general agreement about the indicators that can be used to measure financial risk protection (usually defined as paying more than 40% of their disposable income on health in any time period) by calculating the proportion of the population pushed into poverty by out-of-pocket health payments and/or the proportion suffering severe financial consequences (WHO and WB, 2014) (WHO, 2015). There are still challenges to track UHC progress in the countries. The main challenges concern (WHO and WB, 2015): Reliable sources and data on the health services coverage and financial protection indicators



3/10/18 08:10

16


Sufficient information to monitor effective health coverage levels Reality measurement of inequalities in health coverage. The major information sources of UHC data remain the household surveys and the health facilities. Major challenge concerns the standardization and harmonization of data collection-related to health coverage through health facility reporting systems. The systematic usage of national or international standards for codification and classification of health services information remains critical to enable recorded data in health facilities. Moreover, such standards are extremely useful and sometimes even available for health facilities information management as structured information, for example: Patient identification Health insurance data Health services (medical procedures, medicines, lab and imaging exams ...) nomenclature Reimbursement tariffs Reasons for encounter codification Diagnoses codification Etc. Consistent use of standard coding would allow comparison of similar health facilities, in one country or in several countries, on their situation or evolution related to health coverage. This peer comparison could also give a sight on the progress made towards UHC across countries. Several studies conclude that it is difficult to effectively monitor UHC without contribution of Information and communication technology (ICT). Indeed, ICT innovations in health (or eHealth) are key enablers for achieving and measuring UHC (Susann et al, 2015; Sahay et al., 2014; Verbeke, 2012). This thesis is situatued in the general framework of UHC and it is aimed to contribute to establish evaluation parameters for health facilities, relevant to this general framework. Since 2010, we attempted to establish methods of health coverage evaluation and monitoring by using patient information from health facilities. These methods have subsequently been integrated into a health management information system (HMIS) currently implemented in several sub-Saharan health facilities. Our field study comprised Rwanda, Burundi, The Democratic Republic of Congo (DRC), Mali, Senegal and Tanzania. The aim of our research is to develop a methodology, based on recording health care information, making use of health service nomenclatures and using ICT-tools to improve monitoring of health coverage in sub-Saharan health facilities. Results of this large study is my contribution in the research on UHC specifically in low-income countries. In the following section, we explore the concepts of UHC, challenges posed by data monitoring and the opportunities offered by ICT-tools to improve health coverage monitoring in sub-Saharan facilities.

2.1

Universal Health Coverage In literature, different terms exist to specify Universal Health Coverage labeled by the acronym UHC: "universal health care", "universal health-care coverage", "universal coverage" or simply "health coverage". In the global monitoring of health coverage, the concept "universal health coverage" is preferred to be used. The word "universal" is relating to all or the whole population with the need to consider equal access to the health services, including those currently left behind. Similarly, the term "health" is considered in all its physical, psychological and social dimensions including beliefs, values, and expressed needs of various sub populations, and consider how actions beyond the health sector can be implemented (O'Connell et al., 2014). For the term "coverage", its results are considered, moving from measurement of access to assessment of utilization, appropriateness, and quality.



3/10/18 08:10

Introduction

17

At the level of health facility, we preferred to use the term of "health coverage" and focus only to the patients encountering health facilities. As we only have information on the health facility users, universality cannot be documented. Totality of health care expenses should cover more elements than what we can obtain via the health facilities. To obtain the global view on the health care expenses of an individual or a family, we would need an electronic health care platform, covering the whole health care system. Currently such platforms are no available in the countries under consideration (and even in Europe, few countries could provide such a platform). Moreover, privacy considerations will make data collection very problematic in health facilities. The First Global Monitoring Report on UHC for the WHO and WB (2015) gave a complete definition of UHC: "all people receive the health services they need, including health initiatives designed to promote better health (such as anti-tobacco policies), to prevent illness (such as vaccinations), and to obtain treatment, rehabilitation, and palliative care of sufficient quality to be effective while at the same time ensuring that the use of these services does not expose the beneficiary to financial hardship" This definition precises the scope of health services to be covered and the population with ultimate goal of achieving "Health for all". UHC integrates equity in access, through financial risk protection and implicitly associated with equity in financing. The World Health Reports (WHO, 2010) and Busse et al.(2007) presented the concept of UHC in three dimensions drawn in what has become known as the coverage cube (Figure 1): The health services that are needed, The number of people that need them, The costs to whoever must pay users as patients and third party funders especially health insurances or governments.

Figure 1. Health coverage cube (WHO, 2010 and Busse et al., 2007).

The total volume of the large box is the cost of all services for everyone at a particular point in time. The volume of the smaller blue box shows the health services and costs that are covered from prepaid and pooled funds. The goal of UHC is for people to obtain the services they need at a cost that is affordable to themselves and to the nation as a whole. The three dimensions we consider are the breadth, depth and height of coverage, as spelled out in the 2010 World Health Reports. Breadth (alternatively called width) is short for population coverage; depth for health service coverage, referring to the range of services covered; and height for financial coverage, referring to the proportion of costs covered. All countries struggle to fill the total volume of the cube, including developed countries, which may for example, be fighting to maintain their levels of coverage in the face of rising health services costs (OECD, 2013). That is the reason why moving towards UHC is a perpetual fight for governments, a



3/10/18 08:10

18


dynamic process that must be sensitive to the constantly changing demographic, epidemiological and technological trends. All governments should therefore decide what health services are needed, and how to make sure they are universally available, affordable, efficient, and of good quality.

2.2

UHC indicators monitoring The UHC monitoring framework defined by WHO and the WB Group (2013) highlighted two major components to assess: health services coverage and financial protection coverage for all people.

2.2.1

Health services coverage The health services include prevention, promotion, treatment, rehabilitation and palliative care, and these services must be sufficient to meet health needs, both in quantity and in quality (WHA, 2005). In most countries, the move towards UHC gradually expands access starting from a narrow set of essential health services that are accessible to public and private sector wage earners. However, this approach has often increased health inequities since some groups are more likely to access these services because they are poor people or those working in informal sectors (Carin et al., 2008). Clarity is needed about how much of health services, UHC policies address, whether or not they include other sectors, and, correspondingly, what degree of health inequities UHC can plausibly act upon. To address inequities, experiences from countries that have adopted a broader definition of health indicate that UHC policies might require, at a minimum, establishment of a comprehensive social health platform that provides a continuum of care across an individual's lifespan for all diseases (O'Connell et al., 2014). The health services coverage indicator proposed by Ng et al. (2014), Martinez et al. (2011) and Lozano et al. (2007) in monitoring UHC is the effective coverage. Effective coverage is defined as the fraction of potential health gain that is actually delivered to the population through the health system, given its capacity. This indicator generally offers a direct and flexible means to measure health system performance at different levels. It is a sensitive indicator that links three important aspects: Coverage of health services: refers to the individuals (population) in need of a particular service (Need) Use of health services: refers to the use of services (Use) Access to such services: refers to the actual health benefit experienced from the service (Quality) Effective coverage is a good parameter to evaluate health programs performance, and also provides data of where and to whom the system should address national efforts and resources to achieve the purposes and goals set (Martinez et al., 2011). Boerma et al. (2014) cited a dozen global health services coverage indicators selected by WHO that could be used to track UHC progress. They are classified according to the type of intervention: promotion (prevention) and treatment. Behind some indicators determined by MDGs, those health service coverage indicators include also the control of some non-communicable diseases (NCDs) as diabetes and hypertension as well as neglected tropical diseases (NTDs) such as infectious diseases (malaria, tuberculosis, HIV, etc). Table 1 presents those monitoring health services coverage indicators, their areas and sources of data usually used.



3/10/18 08:10

Introduction

19

Table 1. Health service coverage indicators and prim ary data sources (Boerma et al., 2014)

These dozens of health services coverage indicators could be used to evaluate UHC progress. But countries should select the indicators that are most relevant to their own situation. Basically, the major information sources of UHC data are the household surveys and the health facilities. Household surveys provide accurate population statistics on coverage of services and financial protection, disaggregated by socioeconomic, status, gender and other relevant variables. Health facility data are source for several intervention indicators including coverage of health services, morbidity and mortality rates of diseases, length of stay, health data quality and other information related to health insurance such as patient insurance status, coverage levels, etc.

2.2.2

Health financial protection The need for health financial protection is determined by the proportion of costs that individuals must themselves cover by making direct and immediate out-of-pocket (OOP) cash payments. Under UHC, there would be no out-of-pocket payments that exceed a given level of affordability: 20%-40% of the total of health service expenditures and set at zero for the poorest and most disadvantaged people (WHO, 2010 and 2013). The health financial risk protection for people passes through the health financing process established generally by governments. To structure and analyze the various important issues in the health financing process, the conceptual framework of Busse et al. (2007) is used. It integrates the functions represented in Figure 2 and makes the interactions of those functions explicit in order to facilitate a global understanding of health financing protection. The generic character of the framework, which attempts to integrate characteristics of health financing schemes in the selected high-income countries, can also be applied to low- and middle-income countries. It differentiates the health financing process into the



3/10/18 08:10

20


functions of collecting revenue, pooling funds, and purchasing services. Decisions on coverage and benefit entitlements (depth, breadth and height) have to be made.

Figure 2. Functions of Health Financing System s (Busse et al., 2007)

The health financing protection is heavily based on public revenues (VAT, other taxes or resources) or mandatory contributions for health insurance (often referred to as social health insurance contributions). Before these functions can take place, decisions on health coverage and benefit rights (who is covered? which services are covered? what proportion of the cost is covered?) have to be made (Kutzin, 2012). However, more often, to achieve better health coverage, evaluations and changes in health financing protection are perpetually made and affect the different functions represented. The way these mechanisms are implemented varies considerably from country to country. Even European governments inject general revenues into their health protection systems to ensure coverage for those unable to contribute. For developing countries, the structure of their economy, with a large share of the population outside salaried employment (informal sector), makes it difficult to enforce either income taxes or payroll taxes on most citizens. Thus, other health financial protection arrangements, especially community-based, are gradually institutionalized in those countries with low and middle incomes. The Community-based Health Insurance (CBHI) become health and financial protection systems which cover more and more people in developing countries, especially in sub-Saharan Africa region, to ensure the path towards the UHC. These health insurance schemes are generally initiated by opinion leaders in community, associations of farmers or professional collectivities or by partners in development. The WHO and WB (2015) defined financial protection monitoring in two components: Catastrophic health expenditures and Impoverishing health expenditures. For financial protection indicators focused on the level of out-of-pocket (OOP) payment to monitor the (lack) financial protection. Six concepts extend to which health expenditures strain households’ finances are considered. Table 2 shows those indicators.



3/10/18 08:10

Introduction

21

Table 2. Financial protection indicators (WHO and WB, 2015)

Given these monitoring indicators of UHC, there is a desire for standardization of these measures so that they are comparable across countries and over time. The challenge is to know the household's income and its health expenditure, particularly in countries where the informal sector is predominant. In the next chapter, we present the challenges in gathering UHC indicators at the globally and country levels and at the end we discuss the problems encountered by health facilities in developing countries for data management in the health information system.

2.3

UHC monitoring challenges Monitoring health coverage indicators in the framework UHC still remains a challenge at the global level, country level and local level. As seen, the primary information of UHC comes out from household surveys and health facilities data. In the health facilities, health management information systems (HMIS) enable data gathering in many countries.

2.3.1

Household-based surveys For most indicators, effective household-based surveys are preferred as the better data source even for health services and financial protection coverage. Typically, the two main indicators used for financial protection coverage (catastrophic and impoverishing expenditure) depending on household expenditure data, are obtained through household surveys. Unfortunately, household surveys are far from perfect and have been criticized (WHO and WB, 2015) specifically for: The lack of standardization across countries regarding the recall period used (which hinders comparability) The fact that many household health surveys do not include medical technical measurement such as biomedical markers, which would provide much richer information on effectiveness of coverage.



3/10/18 08:10

22

2.3.2


Health facility data sources Health facility data are often the primary source for district level data on health variables that are used for national level planning, monitoring and evaluation. They are usually reported in annual health statistics reports. These data cover a wide range of topics, including causes of death, acute and chronic diseases, health interventions and health services management and are essential for patient, health facility and health management. In health facilities-based data sources, data collection is a routine activity and health-care facilities tend to collect large amounts of health care information as an administrative work behind the health care activities. The management of health facility information is produced through the administrative records for managing staff, patients, assets, medicines procurement and disbursement, as well as finances. Although most countries have functioning facility-based health management information systems (HMIS), these data continue to have a number of weaknesses. The WHO, Health Metrics Network (2011), Aqil et al. (2009) and Vital Wave Consulting (2009) showed that in many countries, facility-based HMIS remain incomplete due to: Inaccurate and delayed reporting, Lack of accessible databases and poor data quality control procedures, Poorly trained staff that receive little management support, Health staff struggling with systems overburdened by excessive data collection and thus spending hours completing multiple monthly reporting forms, or due to creation of multiple health programsbased reporting systems that further increase the data collection and reporting burden on health workers. For the purpose of evaluating patient health expenses and coverage, the health facility data are an important component but they are limited to the registration in hospitals and other health facilities. Health services provided by other health structures (medication provided by external pharmacies to the patients, contribution of patients to the health insurance organization, etc.) are not taken into account in the recorded data of patients in the health facilities. Therefore, we do not consider them into our research. However, via the health facility data sources, we obtain reliable information that can be considered as a lower limit that the patient spends on the health care. In many African countries, the health costs in the hospitals are the main health care expenses of the patients. The WHO aims to take these data of claims into consideration in the “13th General programme of Work 2019 – 2023” adopted by the 71st WHA (WHO, 2018).

2.4

Facility-based HMIS in sub-Saharan health facilities Health Management Information Systems (HMIS) aim to describe the health situation and trends, and assess health system performance, by using a wide range of data sources including health facility data, administrative returns, household surveys, civil registration, national health accounts (NHA) and health research especially designed to assist in the management and planning of health programs and health care (WHO, Regional Office for the Western Pacific, 2004). HMIS is an instrument which, could be used to monitor health coverage indicators but encounters many challenges in sub-Saharan health facilities.

2.4.1

Facility-based HMIS weaknesses Data collected through HMIS in health facilities continue to have a number of weaknesses (incomplete, inaccurate, poor quality). In sub-Saharan health facilities, the reasons of these failing have been exposed by different authors: Verbeke (2012), Agil et al. (2009) and WHO, Health Metrics Network (2011). Poorly document health information due to:



3/10/18 08:10

Introduction

23

Incorrect and incomplete information Poor patient identification (no complete name, year of birth missing or incomplete, etc.), Absence of unique patient identification allowing to retrieve the patient record, Weakness of data concordance (errors in gender attribution to the patient, service of encounter). A computerized system with data checking at higher levels of the system improves data quality. Poorly adapted clinical ‘international’ standards & guidelines due to the insufficient number of professional competencies and the frequent unavailability of diagnostic, therapeutic instruments or medicines. The development of well adapted care pathways for sub-Saharan countries will have to take into account many of these aspects. Lack of skilled personnel mainly in remote areas caused by the brain drain, seeking for better salaries and economic conditions elsewhere. Plans improving motivation of health workers in the public sector and their deployment in remote areas should be reinforced. Workload of health workers caused by the overload of health indicators to collect needed by the Ministry for health, health programs and the NGOs working in the health sector. This staff spends many hours to collect excessive data, to compile the data manually, to fill in administrative documents, daily, monthly or annual reports. These administrative activities make health workers weary and they reproduce the same period data for another period while adding some random modifications to their reported data. Real-time based information recording in a ICT-HMIS could significantly reduce the overhead of health workers in health facilities and would obtain result in automatic gathering of statistical data and health indicators. Lack of accessible health information: Most sub-Saharan health facilities do not have access to a recent books, guidelines or scientific publications in relevant medical and public health domains. Another problem, in the majority of health facilities, the report documents of medical departments are kept in the departments. As departments are not interlinked electronically, sharing of information becomes almost impossible in practice. Another issue is that those data are kept on the (personal) computers of the health workers, without any security provision nor back-up. The application of modern ICT based knowledge management tools might solve a number of problems related to these communication and security issues. Lack of data structured and standardization: In many sub-Saharan health facilities, the usage of national or international standards for coding, classification and nomenclature of health servicesrelated health information remains exceptional. This standardization is essential to allow data exchange between health structures as well as in the indicators comparison. Nonetheless, such standards are even indispensable for the structured information collected within the HMIS framework. Currently there are standards that should be used for health intervention, procedures and services classification. Such as: Reason for encounter, diagnoses and causes of death: ICD-10 (International Classification of Diseases, 10th edition) (WHO, 2016), ICPC-2 (International Classification for Primary Care, 2nd edition) (WHO-WONCA, 2016) Diseases, functioning and disability: ICF (International Classification of Functioning, Disability and Health) (CDC, 2015) Medical procedures and clinical terms: Systematized Nomenclature of Medicine, Clinical Terms (SNOMED-CT, 2018) Pharmaceutical products: ATC/DDD (Anatomical Therapeutic Chemical/Dose per Day for Drug) (WHO and NIPH, 2016) or RxNorm produced by U.S.National Library of Medecine (NLM, 2014)



3/10/18 08:10

24


Health Interventions and services: ICHI (International Classification of Health Interventions) (WHOFIC, 2012), ACHI (Australian Classification of Health Interventions) (METEOR, 2016) and Belgian nomenclature of INAMI/RIZIV (Institut National d'Assurance Maladie-Invalidité) (2018). Laboratory analyses: Logical Observation Identifiers Names and Codes (LOINC, 2017)

2.4.2

Health coverage monitoring challenges The WHO, Health Metrics Network (2011) highlighted gaps in data quality from facility-HMIS in some areas such as: Service coverage Maternity care: Antenatal visits are underutilized and there is a poor quality of maternity register data Immunization coverage: Countries have either separated or integrated reporting of vaccination and data quality is variable Tuberculosis notification and treatment success: they function well in most countries, through a separate tuberculosis control program reporting system HIV control: Monitoring systems are problematic but improving, but statistics on adherence and outcomes are often poor Service utilization Hospital admission/discharge rates: Poor reporting by large government hospitals; private sector is often not included Outpatient visit rate per person per year: Completeness of reporting often a problem Service management Length of stay and bed occupancy rates for hospitals: Commonly reported but not by diagnosis and data quality Drug availability: Reporting of stock-outs of tracer medicines increasingly used but data quality is variable Morbidity and mortality Causes of death in hospitals (ranked): Commonly used, but poor quality because of poor use of ICD (International Classification of Diseases) Prevalence of malaria: Sentinel clinics may provide reliable trend data, but only few countries have reliable national trend data Maternal deaths: Institutional mortality rate often poorly reported, except if special systems are in place to ensure completeness HIV prevalence among pregnant women: Sentinel surveillance in antenatal clinics, increasing the use of PMTCT (Prevention of Mother-To-Child Transmission) data with more variable quality The problems of data collection and reporting by health facilities and to a lesser extent from districts affect not only the ability to use data to manage services at the local level, but also the quality of statistics for core indicators. The ICT revolution brings opportunities to sub-Saharan countries in their efforts to strengthen the HMIS.

2.5

ICT innovations in health facilities The health systems strengthening required in order to reduce waste of resources, maximize coverage, and provide better quality health care at a lower cost can be addressed with appropriate ICT. Health facilities rapidly and accurately started managing information and reporting routine health information using the open-source, web-based District Health Information Software (DHIS) reporting



3/10/18 08:10

Introduction

25

system. Sahay et al. (2014) and Simba (2004) showed that: ICT solutions can empower patients and communities to engage at all levels of the health system ICT solutions have the potential to reduce healthcare costs to families, improve equitable access to quality services, efficiently link health systems with social protection programs, and increase accountability and sustainability in health service delivery. This assertion was also proven by Umulisa (2010) in a study conducted in the university hospital of Kigali. Optimizing existing ICT infrastructure and making strategic investments in eHealth solutions may accelerate health coverage monitoring. Computerization allows transmission of disaggregated data to the national level. This makes data validation an easy exercise at each level. The introduction and adoption of ICT system in health facilities by entering routine data widen the scope of analyses reducing bulkiness of data reported and enable data to reach its destination much faster to the users. Computerization of the routine data facilitates detection of errors if the health information system is configured to alert the operator on values that are unlikely. There are actually significant opportunities, particularly in low resource environments, for timely and innovative use of cheaper and faster ICT solutions to deliver health coverage monitoring in the right context and at the right time in health facilities. Signs of encouragement are currently visible and some indicators can prove the reality that subSaharan countries move towards ICT development in health facilities: Decreasing costs of ICT equipment and IT-solutions: The costs of computers and other IT equipment are progressively decreasing in the sub-Saharan countries. Some countries like Rwanda and Nigeria became pioneers of a strong political commitment to improve utilization of ICT in several domains (health, education, finance, etc.). Rwanda does not apply taxes to computers and IT-solutions (software, anti-virus), that situation allows running computers and software solution in several institutions including health facilities. Acceleration of internet accessibility: The accelerated connection of sub-Saharan countries to much cheaper optical fiber based Internet is very promising for the propagation of ICT-solutions in many public and private institutions. In Rwanda the policy to connect all public administrations including referral and district hospitals on optical fiber is running up. 3G and 4G wireless data solutions have become available in a growing number of regions. Speed internet connections using 4G mobile wireless data solutions are implementing by several internet providers in Rwanda, Senegal and Tanzania. Exploitation of alternative energy sources, such as solar power, which are abundantly available on the African continent need to be exploited. Many successful set-up of small autonomous solar power systems have demonstrated that such solutions can be very beneficial in remote areas where electricity supply is insufficiently reliable or even completely out of reach (De Bruijn et al., 2010). Solar-power allowed health facilities to run computers in DRC and Senegal in remote areas where there is no electricity (Nyssen et al., 2011). Development of human capacity in ICT: ICT-related human resource capacity building in subSaharan countries remains for the health sector an urgent issue of concern for the sustainability of ICT programs. Reinforcement of local educational and training capacities, adapted to regional health care needs is for most developing countries a promising topic. The example of health informatics training in Rwanda based on a detailed Central African region health workers needs assessment in ICT and IMIA (International Medical informatics Association) knowledge base will allow gradually to fulfill the gaps in health information systems in the region (Wright et al., 2015). The growing interest of donors and governments to introduce Open Source IT solutions for hospital management in several sub-Saharan health facilities: Several OpenClinic GA implementations are recorded in more than twenty health facilities both public and private in more than 8 countries in sub-Saharan Africa.



3/10/18 08:10


Part


III

3/10/18 08:10

Research hypothesis, limitations and study plan

3

27

Research hypothesis, limitations and study plan Our research starts from the hypothesis that "Health coverage can be assessed and monitored in sub-Saharan health facilities using ICT methods based on structured patient record data available for clinical practice". The study attempts to evaluate to what extent health services coverage for patients who use health facilities in sub-Saharan countries: Get the health services they need, and Have financial risk protection in receiving those health services. The purpose of our research is to define protocols consisting of practical application of methods and techniques using ICT-tools to improve monitoring of health services coverage and thereby contributing to the improvement of patient health coverage. The results can help in measuring the impact of countries UHC health policy in health facilities and in benchmarking health facilities on their levels of health coverage. The study leads to two groups of research questions: 1. Questions about improving health coverage including financial protection: What are health services that patients consumed? How are inequities in health coverage in health facilities? What is the sustainability in health service delivered by health facilities? 2. Questions about measurement. Which data and indicators are needed to: Monitor health services coverage? Monitor financial protection? One task for this research is to help define a set of common indicators to evaluate progress towards health coverage across sub-Saharan health facilities. Another task is to develop methods to track patient expenses, to measure the extent of financial protection as well as the access to health services and their evolution over time. Therefore, we had to: Document and analyze the sub-Saharan regional health policy situation regarding UHC, health facilities and health insurances. A review of health policy of a set sub-Saharan countries (Rwanda, Burundi, DRC, Mali, Senegal and Tanzania) focused on UHC was done: what is the actual health services coverage in these countries and how do their health financing systems currently function. Analyze different health insurance schemes and -levels implemented by a set of health insurance organizations in sub-Saharan countries. Develop a minimum set of health service metrics that enable health coverage monitoring and evaluation in sub-Saharan Africa health facilities. Develop a reference health services classification (RHSC) for sub-Saharan health facilities to facilitate health services standardization and peer comparison. Integrate health service codes and classifications towards a health information management system (HMIS) that was put in the public domain. Install the modified HMIS including the RHSC middleware in 28 participating health facilities in six sub-Saharan countries: Rwanda (10), Burundi (8), DRC (2), Mali (5), Senegal (2) and Tanzania (1). Evaluate and monitor the evolution of patient health services coverage in these health facilities. Evaluate the evolution of health services reimbursement and patient out-of-pocket (POOP) payment in the health facilities. By this pilot study, it will be able to track the health coverage from patient electronic data recorded for clinical practice. Indeed, electronic health information recorded by the health facility for the patient management will be secondary used for health coverage monitoring.



3/10/18 08:10

28


My ambition was to make use of readily available data in the electronic health records to contribute to the evaluation of patient health coverage metrics. These parameters could be considered as additional indicators, complementary to the traditional methods of evaluating health coverage through household surveys. The limitation of this study is that we do not disposed information of the whole health care system in the countries which are part of the study; moreover we could only compare similar health facilities presenting similar case mixes, regarding the coverage of health care services. Since we did not capture incomes data of individual patient, our study could only draw conclusions on relative health coverages, comparing to the full costs of the treatments and the average incomes in the countries. In summary, we discern five limiting factors: We consider the expenses of patients in one health care facility We cannot take into account other expenses (pharmacy, health insurances fees, etc.) We do not consider the hospital costs financed by other sources, thereby reducing the patient’s costs We are not aware of the patient’s revenues, we know the GDP per capita but this is only an estimation, useful for comparisons Representativeness of the health care facilities: our studies are focusing on second and third level hospitals. Health centers and other care structures are not included in our main comparison study.



3/10/18 08:10


Part


IV

3/10/18 08:10

30

4


Materials and methods Our research project began in 2010. The document presents the results for the period 2010-2016. The field study covered six sub-Saharan countries: Rwanda, Burundi, Tanzania, RD Congo, Mali and Senegal. During the 6 years' research period, 28 health facilities in these countries have implemented an ICT-HMIS, called OpenClinic GA software through our ICT for Development (ICT4D) project in the Vrije Universiteit Brussel (VUB). OPENCLINIC GA (OPEN system for Comprehensive heaLth facility INformation management in low Income Countries General Availability) is an Open Source hospital management information system covering management of administrative, financial, clinical, lab, x-ray, pharmacy, meals distribution and other data. The software also covers extensive statistical and reporting capabilities. Through patient information entered in OpenClinic GA, we focused on data related to the health coverage indicators monitoring in health facilities and the most essential ones concerning the: Patient health insurance coverage Information related to the patient's Identification Information related to the patient's health Insurance Patient health services coverage Information related to patient's health services consumed Information related to insurance health services covered Patient financial protection Information related to patient's health services payment Information related to insurance health services reimbursements The second ICT-tool is a data warehouse, the Global Health Barometer (GHB), installed in our project server at the university. The GHB enables fully automatic health & performance indicator extraction from local OpenClinic GA databases and merging of the resulting data into a single central database. The system allows extracting of information related to health facility activities, patient health indicators and information related to the patient health insurances. From this webbased dashboard, we get information concerning: Key figures of OpenClinic GA utilization by health facility Users configured Insurer plans used Patient health records Delivered health services Patient invoices Insurer invoices Health facility revenues per month or year period, provided by: Patients revenue (out and in-patient) Insurers revenue (out and in-patient) Patient health indicators Number of hospitalization and visits (encounters) Number of diagnoses and reasons for encounters Bed occupation rates Methods developed for the evaluation of the study hypothesis has been subdivided in 6 different activities summarized in this diagram:



3/10/18 08:10

Materials and methods

31

Figure 3. Methods for health coverage m onitoring in sub-Saharan health facilities

1. Analyze of the financial health policy situation regarding UHC in sub-Saharan countries, especially for the health expenditure, financial risk protection and some health services coverage indicators. 2. Define a minimum set of health service metrics that enable health coverage monitoring and evaluation in sub-Saharan Africa health facilities 3. Identify reference of health services classifications (RHSC) for sub-Saharan health facilities to facilitate health services standardization and peer comparison. 4. Make necessary modifications to health service codes and classifications towards the HMISOpenClinic GA that was put in the public domain. 5. Implement the modified OpenClinic GA including the RHSC middleware in 28 participating subSaharan health facilities. 6. Evaluate the measurements selected for health coverage monitoring. Some measurements are enumerated: Patient health insurance coverage (PHIC) Patient health services coverage (PHSC) Patient health services payment rate (PHSP) Patient out-of-pocket (OOP) payment (POOP) Health insurance expenditure per patient (HIEX) Burden for patient per pathology Coverage per pathology

4.1

Activity 1: Analysis of UHC situation in sub-Saharan countries As the first step, an analysis of UHC situation in sub-Saharan countries was to be performed especially for six countries: Rwanda, Burundi, RDC, Tanzania, Mali and Senegal. Intended results: Documentation and analysis of the sub-Saharan regional health insurance policy situation



3/10/18 08:10

32


regarding UHC A review of health insurance schemes and health insurance organizations in a national picture of health expenditure and financial risk protection for the set of sub-Saharan countries (Rwanda, Burundi, DRC, Mali, Senegal and Tanzania): What is the actual population coverage in these countries? How do their health financing systems currently function? What is the level of health financing by the insurance organizations? What is the out-of-pocket payment? Definition of health services coverage indicators that could be used to monitor UHC progress and a comparison situation making use of those indicators between the six sub-Saharan countries: What is the situation of health coverage in these countries? Methodology Studying WHO and World Bank statistics and reports related to the health protection financing in sub-Saharan countries Studying national and regional health financing policy Documenting health insurance schemes existing in the six sub-Saharan countries Studying WHO's intervention coverage indicators that could be used to monitor UHC progress and their measurement in the six sub-Saharan countries.

4.2

Activity 2: Define a set of health service metrics for health coverage monitoring A minimum set of health service metrics must be developed in order to monitor and evaluate the health coverage in sub-Saharan health facilities. Intended results: A definition of a set of health services metrics that enable the measurement of health coverage in sub-Saharan health facilities. How to define each metric (numerator, denominator)? Health service utilization Health service coverage Health service burden Information related to health facility management Meaning and relevance: description of what each metric measures and why it is relevant in health coverage monitoring and evaluation A list of required structured information elements (data input) for every metric. Health information management issues can deal with different categories of information: Administrative information management Financial information management Clinical information management Pharmaceutical information management Technical (laboratory, medical imaging...) information management Methodology: Review and analysis of health services standard measurements (metrics) that could be applied for health coverage monitoring Define health information data required to establish every health service metric.

4.3

Activity 3: Identify RHSC for health facilities A Reference of Health Services Classifications (RHSC) for sub-Saharan health facilities must be identified to facilitate health services standardization and peer-comparison. Intended results:



3/10/18 08:10


33

A set of health service category codes that show which health coverage is essential. This RHSC could concern different groups of medical procedures, health services and conditions: Care deliveries Medical procedures Reason for encounter Diagnoses Pharmaceutical products Laboratory exams Medical imaging exams Identification of the RHSC for health facilities. The RHSC would use international or national standards for health services or health conditions Methodology: Review of the international codifications related to the health services and health conditions in health facilities Inventory of all health care deliveries dispensed in sub-Saharan health facilities Create the generic care delivery (GCD) classification code and assign for each health care delivery one GCD classification code Create reference of health services classifications (RHSC) and assign classification to health procedures and services

4.4

Activity 4: Integrate RHSC in the ICT-HMIS tool The standard codes of health services classifications (developed in activity 3) must be configured in an integrated health management information system-IT tool that can help the management of subSaharan health facilities. The used software HMIS-OpenClinic GA was put in the public domain. OpenClinic GA should facilitate the implementation and usage of the metrics defined in Activity 2 for health coverage monitoring in those health facilities. Intended results: A number of OpenClinic GA software modules (data entry tools) must allow to configure all information related to the patient health services cost and health coverage. These modules must at least integrate the following data: Health facility and medical department information Patient identification Health insurer information Out and in-patient encounter information Reason for encounter and diagnoses information Health care deliveries information Laboratory analyses Drugs and consumables Invoicing information A number of software modules (monitoring tools) must allow to automatically calculate the metrics identified in Activity 2. This calculation is based on electronic data coming from the existing OpenClinic GA in the health facility (where relevant) completed with indicators centralized by using the Global Health Barometer (GHB) data warehouse. Methodology: Configure the standard RHSC defined for health coverage monitoring in OpenClinic GA modules Create in statistics module links to generate automatically measurement indicators for health coverage in health facilities



3/10/18 08:10

34

4.5


Activity 5: Implement of the ICT-HMIS in health facilities The RHSC codes developed in Activity 3 and installed in OpenClinic GA in Activity 4 allows health coverage monitoring and evaluation using the health services metrics developed in Activity 2. In this phase, we implemented a six-years OpenClinic GA project (2010-2016), involving 28 participating health facilities in Rwanda (10), Burundi (8), DRC (2), Mali (5), Senegal (2) and Tanzania (1). The OpenClinic GA software was installed in the participating sub-Saharan health facilities with the purpose of managing the health information and automatically calculating the metrics defined in Activity 2. The installation and configuration of health services codes must be performed in the OpenClinic GA installed in the participating health facilities. The criteria of selection of these health facilities are: Minimum ICT infrastructural requirements for a proper implementation of OpenClinic GA must be available: OpenClinic server and user computers Local network allowing user computers connection to the OpenClinic server Internet network that allows the external connection to the OpenClinic server Proper packaging of the OpenClinic GA software for the relevant ICT environments (operating systems and hardware) must be installed and configured: OpenClinic software installation including the implementation test phase OpenClinic software configuration in order to meet specific local requirements Availability of OpenClinic software on a public server through internet access End users, data managers and ICT staff must be trained in order to enable them to properly use and monitor the system: Training all staff (administrative, administrative, technical, etc.) Existence of management project team (advanced users that are able to provide support to endusers) OpenClinic GA software should be operational during the period of the study OpenClinic software is used for health facility management OpenClinic software is maintained so that it remains functional Intended results: Automatically extraction of health coverage monitoring indicators should be easily reported OpenClinic GA system must be used systematically used by administrative and cashiers staffs for patient identification, ADT management and patient invoicing OpenClinic GA system should be used in the second phase of implementation by medical and paramedical staff to manage medical records, laboratory, pharmacy and imaging services. Methodology: Installation of OpenClinic GA on the local server and specific configurations Training and follow-up local staffs Quality control of data entries

4.6

Activity 6: Evaluate the health coverage in health facilities The main purpose of the study is to evaluate and monitor the health coverage in sub-Saharan health facilities using measurements and indicators automatically generated by OpenClinic GA software. Therefore, in Activity 6, we analyzed the results obtained from health service metrics developed in Activity 2 and implemented in Activity 5.



3/10/18 08:10


4.6.1

35

Health coverage indicators in sub-Saharan health facilities We have to determine health coverage indicators for monitoring that can be applied to data of OpenClinic GA available in the participating health facilities. Intended results Results in each health facility studied are presented using health coverage indicators defined as: - Patient coverage - Patient payment - Health services costs reimbursement Comparison between health facilities is made in each country. Those indicators should be compared: - Which health insurance scheme is more used by patient? - How many patients are insured or not? - What is the proportion of OOP on the total costs of health services? - What is the amount of Patient OOP? Strategies to improve the metric score: which strategies can be followed by health insurers and/or health facilities in order to achieve a ‘better’ result in health coverage? Methodology Extract defined health coverage metrics from OpenClinic GA for each participating health facility by country using SQL queries on the data base Consult defined statistics in OpenClinic GA and health indicators in the GHB Document the data quality of the resulting relevant information elements for every health facility.

4.6.2

Comparison on health coverage indicators in sub-Saharan health facilities A Comparison has to be made between the measurements of health coverage indicators obtained from different health facilities of in same country or in different countries and in defined period. Intended results: Peer comparison for same categories of health facilities are obtained Case mix evaluation is made concerning - Morbidity and mortality - Burden per pathology - Coverage per pathology The impact of health coverage in health facilities is measured and peer compared by used indicators related to: - Utilization - Revenue - Reimbursement Methodology: Extraction of health coverage indicators from OpenClinic GA for health facilities Statistical analysis based on indicators peer-comparison by using OpenClinic GA statistics module and GHB health indicators



3/10/18 08:10


Part


V

3/10/18 08:11

Results

5

37

Results Sub-Saharan African countries are moving towards UHC. In some countries including Zambia, prepayment of health services is enhanced due to an increase in government funding supported by external donors. Other countries use an alternative path. For example, Ghana has passed a health insurance law and Lesotho has explored the feasibility of a social health insurance reform. Kenya’s National Hospital Insurance Fund has been examining ways of extending coverage to the informal sector (Carrin et al., 2008). These examples show the different pathways that follow the sub-Saharan countries to achieve extending health coverage to more of the population, especially by integration the informal sector.

5.1

Activity 1: Analysis of UHC situation in sub-Saharan countries In this activity, we analyzed the health insurance policy situation in six sub-Saharan countries (Rwanda, Burundi, DRC, Mali, Senegal and Tanzania). Some health services coverage indicators that could be used to monitor global UHC progress were reviewed and compared between the six sub-Saharan countries. Demographic and financial figures were obtained from the World health statistics reports (WHO, 2014; 2015) and 2014 Word Bank report. Our first analysis focused on the health expenditure and financial risk protection indicators, then on the indicators concerning health services coverage. Finally, we present the situation of health insurance schemes in the six studied countries.

5.1.1

Health expenditure and financial risk protection In our six studied countries, we describe the situation of UHC. Some definitions regarding health expenditures definitions are founded in Annex 8. The following table expresses some figures on health expenditures in Rwanda, Tanzania, Burundi, Senegal, Mali and DRC. Table 3. Health expenditure in six sub-Saharan countries (WHO-WB statistics, 2014)

According to data of GDP per capita, Burundi and DRC show values that are below the poverty line defined by the WHO and the WB of USD1.25 per day per capita (international poverty line for lowincome countries). The figure 4 summarizes the repartition of health expenditure per capita in countries of our study.



3/10/18 08:11

38


Figure 4. Per capita expenditure on health at average exchange rate in 2012

For a more accurate comparison, we used the USD and the purchasing power parity (PPP) exchange rates. The total health expenditure per capita is higher in Rwanda (USD70/Intl$158) compared to other countries in our study. For this country, the public expenditure related to health represents 58.6% as of total health expenditure. Tanzania follows Rwanda with a total health expenditure of USD42/Intl$117. The DRC (USD15/Intl$25) has low expenses in public health per capita. Private expenditures in Tanzania and Mali exceed government expenditures. They represent 61.9% of health expenditures for each country. For other countries, the public participation in health financing is more important than the private one. The figure 5 compares the out-of-pocket (OOP) expenditure with the other private health expenditures.

Figure 5. Per capita private expenditure on health at average exchange rate in 2012 (USD)

In 2012, the OOP expenditure on health was USD25.90 per capita in Mali. This amount represents 99.6% of the total private expenditure on health. For Senegal, the OOP expenditure (USD17.03 represented a high proportion (77.3%) in private health expenditure. The OOP health expenditure per



3/10/18 08:11

Results

39

capita was around USD13.00 in Rwanda and Tanzania representing respectively 45.6% and 51.9% of total private health expenditure. The OOP health expenditure in DRC (USD4.68) and Burundi (USD4.39) were the lowest in studied countries. In Senegal and Mali, the OOP is more involved in the health services financing than other private health expenditure (insurance, charitable donations, and direct service payments by private corporations). Rwanda and Tanzania have managed to establish some pooling of private health expenditure. In Burundi also the total expenditure is split equally between out of pocket and other private expenditure. In the three other countries, Senegal, DRC and Mali, the expenditure on health are not equally shared with in Mali, where the private pooling has failed despite years of support to the mutuelles de santé(CBHI).

5.1.2

Health services coverage indicators WHO proposed dozens of intervention coverage indicators that could be used to monitor UHC progress. Countries should select those indicators that are most relevant to their own situation (Boema et al., 2014) (WHO, 2015). Figure 6 provides an illustrative application of the use of 12 intervention indicators with data from six sub-Saharan countries (Rwanda, Tanzania, Burundi, Senegal, Mali and DRC). The means are separately computed for promotion/prevention (red color) and for treatment indicators (blue color). The treatment mean was computed as unweighted average of five selected indicators: Percentage of skilled birth attendance (Maternal and newborn care) Percentage of population with advanced HIV infection using ARV therapy Percentage of male persons with hypertension who are successfully treated (Hypertension treatment) Treatment-success rate for new tuberculosis cases (TB case detection) Percentage of persons with diabetes who are receiving successful treatment (Diabetes control) The promotion and prevention mean was computed as unweighted average of 6 interventions areas including indicators: Family planning coverage with modern methods satisfied among women 15-49 years (FP need satisfied) Antenatal care coverage at least 4 visit (Antenatal visit) Full immunization (e.g. DTP3) coverage among infants (1-year-old) (DTP3 immunization coverage) Prevalence of no tobacco smoking in the past 30 days among male adults (>=15 years) (Non-use of tobacco) Percentage of population using improved drinking water sources (Improved water source) Percentage of population using improved sanitation facilities (Adequate sanitation facilities)



3/10/18 08:11

40


Figure 6. Coverage of selected indicators (dots) w ith m ean of all health services coverage (red and blue bar) in 6 sub-Saharan countries

In general, prevention coverage is considerably higher than for treatment. Rwanda's mean for prevention is 66% (the average is reduced due to weak antenatal visit coverage, 35%), followed by Senegal (60%), Burundi (59%), Tanzania (51%), DRC (48%) and Mali (47%). Rwanda also has the highest mean for treatment coverage (51%), followed by DRC, Burundi and Mali (45%), at the end Tanzania and Senegal (42%). Two indicators, one for prevention (DTP3 immunization coverage) and the other for treatment (TB case detection) are higher covered in the six countries (between 72 and 98%). The values for effective coverage of Diabetes control were particularly low in all countries. The similar patterns are observed in some developing countries such Egypte (15%), Bangladesh (20%) and Chile (30%) (Aguilera et al., 2014) (Boerma et al., 2014) and this problem of low coverage worsens as household income is low (Hosseinpoor et al., 2014).The Family Planning coverage is also lower in Mali (10%), Senegal and DRC ( 30% patient encounters) and the PIH scheme was almost non-existent (0.94) were observed between out-and in-patient health insurance coverage scheme distributions. The correlation coefficient was very strong in reference hospitals (r>0.99) and a little less in GIDH (r=0.96) and CDS (r=0.95) where high differences were noted in use of SHI scheme (-11.7%) and PHI (-10.9%) respectively from out- to in-patient encounters. In general, there were no significant differences in the use of health insurance schemes in studied Rwandan facilities among out- and in-patient encounters.

5.6.1.1.2 Patient health services coverage (PHSC)

The following graphic shows the general situation of Patient health services coverage rate (PHSC) with individual POOP79%) for health services consumed and inpatients are better insured in reference and private hospitals compared to the outpatients. 5.6.1.1.3 Patient health services payment rate (PHSP)

The patient health service payment rate (PHSP) in Rwandan health facilities is represented by the graph 50. This metric expresses the proportion amount paid by the patient to cover his health services consumed. The PHSP for health services must not exceed 25% of total health services consumption in the framework of health coverage.

Figure 50. PHSP in 10 Rw andan health facilities

The general PHSP in the 4 participating private Rwandan health facilities exceeded the threshold of 25% during the period of study. It was higher in specialized medico-technical private clinics, 32.6% in BMC (n=76842) and 29.9% in KHIBD-DC (n=23912), than private hospitals: 29.1% in LMED (n=270459) and 25.9% in CDS (n=270845). In the public health facilities, the PHSP was below the threshold of 25%. It still reached 20% in CHUK (n=348904) and in Rwamagana hospital (n=124545). It was lower (14.4%) in the NPH-CN (n=50453). In this neuro-psychiatric hospital, the participation of the health insurance in the expenses of health services consumed exceeded 85%. The following table shows the PHSP for inpatient admissions and outpatient consultations in 8 studied hospitals in Rwanda. Out-pa ti ents In-pa ti ents Di fference



3/10/18 08:17

156


22.6% (n=279 859) 13.4% NPH-CN (n=45 800) 16.2% BUDH (n=27 270) 17.5% NYDH Di s tri ct (n=101 458) hos pi ta l s 18.8% GIDH (n=84 100) 22.6% RWDH (n=82 188) 25.9% CDS Pri va te (n=244 186) hos pi ta l s 28.8% LMED (n=270 054) 24.0% Total (n=1 134 915)

Reference hos pi ta l s

CHUK

17.6% (n=47 171) 15.8% (n=10 754) 14.8% (8 830) 16.4% (n=41 322) 19.4% (n=45 795) 21.1% (n=52 699) 27.0% (n=33 510) 36.3% (n=4 351) 20.5% (n=244 844)

-5.0% 2.5% -1.4% -1.1% 0.6% -1.5% 1.1% 7.5% -3.5%

Table 20. PHSP of out- and in-patients in 8 Rw andan hospitals

Averages of PHSP for outpatients and inpatients remained below the 25%-threshold of the total amount of health services accounted. In-patients are less covered (-3.5%) than outpatients. In private hospitals, outpatients and inpatients were poorly covered because their payments exceed 25%threshold of health services consumed. An effort is still needed to reach the patient financial protection in private health facilities for e.g. by financing these structures for some activities and then reduce the share of patients in the costs of health services. Health insurance coverage should be more important in these facilities. 5.6.1.1.4 Patient out-of-pocket payment (POOP)

The Patient out-of-pocket payment (POOP) expresses the average amount paid by the patient for health services. The following graph shows the situation in Rwandan health facilities.

Figure 51. POOP in 10 Rw andan health facilities

Globally, the POOP was higher in private hospitals than public hospitals. It was exceeded in average USD20.00 at Croix du Sud hospital and it was USD17.00 at CHUK hospital. The POOP was around



3/10/18 08:17

Results

157

USD6.00 in district hospitals. The POOP was higher in private facilities and national referral hospitals because the tariff of health services applied is also high. For e.g. a Medical specialist consultation for insured public patients costs USD4.38 in a district hospital, USD8.54 in private hospitals and USD8.70 in national referral hospitals. Average POOP for uninsured patients were USD6.57, USD13.14 and USD14.60 respectively in district, private and referral hospitals. In general, basic health services (particularly medical procedures) whose prices are set by the MoH, cost two times more in referral and private hospitals than in district hospitals. Although the tariff of health services in NPH-CN and CHUK are almost the same, the POOP at NPHCN was lower because the PHSP was also lower in that hospital (14.3%) than in the CHUK (20.7%) as seen previously. Rwandan national health policy longer supports mental health care with subsidies of free care for all disadvantaged persons (8.4% of all patients). And mental health care is still more reimbursed by the health insurances than general health services provided in health facilities. The following table shows the POOP for inpatients and outpatients in 8 studied hospitals in Rwanda. Table 21. POOP of Out- and in-patients in 8 Rw andan hospitals

Out-pa ti ents USD12.50 CHUK Reference (n=279 859) hos pi ta l s USD7.23 NPH-CN (n=45 800) USD3.64 BUDH (n=27 270) USD2.90 NYDH Di s tri ct (n=101 458) hos pi ta l s USD3.08 GIDH (n=84 100) USD3.28 RWDH (n=82 188) USD16.94 CDS Pri va te (n=244 186) hos pi ta l s USD9.52 LMED (n=270 054) USD10.10 Total (n=1 134 915)

In-pa ti ents Di fference USD44.30 +254.3% (n=47 171) USD34.04 +371.1% (10 754) USD12.97 +256.2% (n=8 830) USD13.47 +364.8% (n=41 322) USD9.59 +211.4% (n=45 795) USD14.95 +355.5% (n=52 699) USD73.93 +336.4% (n=33 510) USD24.90 +161.6% (n=4 351) USD28.36 +180.9% (n=244 844)

In global, the POOP average of inpatients was higher (+181%) than the POOP of outpatients. The largest difference of POOP was observed at NPH-CN (+371%) and the smallest at LMED (+162%). POOPs for in-and out-patients remain high in private hospitals and in reference hospitals because the health service costs are higher in these health facilities than in district hospitals. Compared to the OOP per capita in Rwanda (USD13.22) (WHO-WB statistics, 2014), out- and in-POOP in these hospitals includes the value of national OOP. The average POOP in participating health facilities reflects the national average OOP in household despite the non-representativeness of health facilities. To be treated in Rwandan hospitals remains accessible thanks to the important coverage of health insurance. Greater involvement of health insurance in private hospitals is recommended to reach the health coverage in health facilities.



3/10/18 08:17

158 5.6.1.2


Burundian health facilities These indicators are analysed: Patient health insurance coverage (PHIC) Patient health services coverage (PHSC) Patient health services payment rate (PHSP) Patient out-of-pocket payment (POOP) The results are published in the manuscript "OpenClinic GA Open Source Hospital Information System Enabled Universal Health Coverage Monitoring and Evaluation in Burundian Hospitals" (Karara et al., 2017).

5.6.1.2.1 Patient health insurance coverage (PHIC)

The following graphic represents the situation of patient health insurance coverage (PHIC) schemes used for outpatient encounters in the participating hospitals of Burundi. References number of out and in-patient encounters that used health insurance schemes during the period of our study are found in activity 5.

Figure 52. PHIC for outpatients in 8 Burundian hospitals

In general, free health services (FREE) and social health insurance (SHI) were the mostly frequent used insurance schemes by outpatients. FREE scheme was more frequently used in 18.9% to 32.7% outpatient encounters in district hospitals. This situation could be explained by the important number of pregnancy women and children under five that encounters these second level reference hospitals for follow up visits. The SHI scheme was more frequently (34.3%-69.1%) used in reference hospitals and especially in HMK (69.1%, n=1599814). Only a fraction of Burundian citizens (formal sector, civil servants, army and police staff and their relatives) are covered by this scheme. The PATIENT scheme use was highest (64.8%) in CMCK (n=126863), a private hospital. In this hospital, private health insurance (PHI) schemes were also mostly used (28.4%) by outpatient encounters. The Community based health insurance (CBHI) scheme was mainly encountered (7.5%-16.1%) in district hospitals. This scheme was almost non-existent in other hospitals. Inpatient health insurance coverage schemes followed almost similar distribution as showed in the following graph.



3/10/18 08:17

Results

159

Figure 53. PHIC for inpatients in 8 Burundian hospitals

Correlations between 0.28 and 0.98 were observed between out-and in-patient health insurance coverage schemes distribution. Strong correlations were observed at CMCK (r=0.98) and at HMK (r=0.91) due to the stability of health insurance schemes encountered in these hospitals. A weak correlation was in KIDH (r=0.28) where high differences were noted in use of FREE scheme (+36.7%) and SHI scheme (-21.8%) between in- and out-patient encounters. We observed a significant overall increase in the use of FREE scheme for inpatient encounters in all hospitals and use of CBHI scheme in district hospitals at the expense of SHI scheme. There was a difference in the use of health insurance schemes in Burundian facilities between outand in-patient encounters where the FREE scheme took more consideration in admission encounters. Apparently, Free healthcare policy is mostly applied in hospitalization at second and third reference level compared to outpatient visits. 5.6.1.2.2 Patient health services coverage (PHSC)

The following graphic shows the general situation of Patient health services coverage rate (PHSC) in participating Burundian hospital during the study period. Number of patients and encounters analyzed during the period of our study can be found in the activity 5.



3/10/18 08:17

160


Figure 54. PHSC in 8 Burundian hospitals

The PHSC exceeded 40% in public hospitals and was below 30% (90% USD13 19% 4.9% NPH-CN 10 754 53% 93% USD34 16% CHURK 25 939 59% 70% USD99 33% 34.6% Burundi USD286 23%-30% USD4 21% HMK 48 487 66% 69% USD132 57% 46.0% CUL 2 928 26% 7% USD150 96% 33.9% DRC USD442 71%) or in public (PHSC>81%) health facilities. In the Neuro-psychiatric hospital of Ndera (NPH-CN) and in BUDH, the 92%-PHSC threshold was exceeded whether in hospitalization or in out-consultation. The specialized nature of these public hospitals in mental health and oncology prompts patients to have a good health care coverage to limit the financial risks associated with expensive and long treatments (Carrin et al., 2008). Poor and uncovered patients are often covered by the MoH or charity and NGO associations. The high level of insured patients seen in private health facilities is explained by health services covered by public insurance schemes (RSSB / MMI) and private insurances (PHI). Patients covered by public or private insurance are more likely to be treated in private hospitals where they hope to receive better quality of care (Basumail et al., 2012). The proportion of patients’ out-of-pocket payments compared to total health services consumed expenses (PHSP) was around 24% for out-patients and 20% for in-patients in the 8 participating Rwandan hospitals. The PHSP was globally below the 25%-threshold of accounted health services, but remained high in privates hospitals (26%-36%) due to the high costs of health services applied in these hospitals. We noted the proportion of the amounts paid by patients has generally decreased during the study period thanks to the increase in patient health service coverage (PHSC). We developed these findings in the study made in six Rwandan hospitals during the period 2011-2014 (Karara et al., 2015). The average amounts actually paid by patients (POOP) for health services uncovered varied between USD10 for outpatients and USD28 for inpatients. These amounts tend to increase at CHUK and private hospitals during the study period. We saw also an increase of health insurance expenditures (HIEX) due to the increasing of health service costs brought by new medical technologies in imaging and laboratory. Regarding the financial protection, the World Bank and WHO statistics (2014) reported for Rwanda an OOP of USD13.22 representing 18.9% of total health expenditure. Comparing these national indicators and the results in the 8 participating hospitals, the range of the POOP in these hospitals (USD10-USD28) included the national average, but the proportion of the amounts paid by the patients related to the total amount of health services consumed (20%-24%) was high because these health facilities represented only the hospitals of second and third reference level. These results show that, with an exhaustiveness and a possibility to track the patient across health facilities encountered, we could well measure the impact of UHC policy in Rwandan health facilities from data recorded in the ICT-HMIS.

6.4.2

Burundian health facilities In Burundian health facilities, the results showed that patient health services coverage (PHSC) was globally 71% for inpatients and 61% for outpatients. This indicator was higher in public hospitals, when it is compared to private hospitals due to the important intervention of health coverage plans oriented towards Free health service and Social health insurance schemes. We noted the



3/10/18 08:22

206


intervention of CBHI plans in district hospitals especially for outpatient encounters. The CBHI scheme was predominantly based on the use of the CAM insurance plan. The results of health coverage in the participating hospitals were higher than in the reality at national level where the population health coverage was estimated between 23%-30% in 2014. The group of patients in our population seemed apparently better insured than the general population. This is the consequence of the adverse selection and the moral hazard observed in certain health insurance contexts (Parmar et al., 2012; Noterman et al., 1995). In the case of Burundi, the adverse selection is influenced by the voluntary enrollment in CBHI plans (e.g. CAM), thus more people with greater health care needs to adhere in this insurance scheme. The moral hazard is influenced by the political choice to insure systematically and freely pregnant women and children under five years that can clearly lead to the over-consumption of health services. CBHI plans and government are thus required to find the necessary funds to pay the invoices of health services consumed by the insured users. The delay or non-reimbursement of health facilities results in shortages of medicines and other medical equipment, delays in staff salaries ... and in the decreasing of the quality of care. The OOP expenditure in Burundi was 21% of total health expenditure in 2012 (WB and WHO statistics, 2014-2015). Bearing in mind that the hospitals studied were at the second and third referral level, the patient health service payment (PHSP) remained globally above the 25%-threshold both for inpatients (37%) and for outpatients (47%) due to health services not covered by certain health insurance plans at that level. This situation has also been observed in Rwandan hospitals for some plans of health insurance schemes in private hospitals (Karara et al., 2015). The POOP was also higher for inpatients (USD42) and outpatients (USD5) than the national average (USD4.39) as could be expected. The in-POOP represented 35% of the average income per capita and it was high in private hospitals (CMCK, HMK) due to the high costs of hospitalization health services. These high costs in hospitalization affect the accessibility of health services in these hospitals. The policy of Free health care combined with the improvement of other types of health insurance schemes such as CBHI including everyone, a mandatory public insurance system and more involvement of the population in the health financing (through insurance contributions, co-payments, taxes ...) are ways to deal with the consequences of selective policy to provide health coverage to a part of the population and move towards sharing health risks.

6.4.3

DRC health facilities The PHIC in the two university hospitals of DRC varied between 25% and 40%. The PHSC was in average 16% for inpatients and 17% for outpatients. National level indicators showed health insurance coverage at 2% and the OOP as proportion of total health expenditure at 31%. We can suspect that patients treated in the two hospitals have an easier access to these healthcare than the majority of the population. In reality, treated patient are public and private institution staff in the region, the staff and students of the two universities through private insurances and the university insurance. In the two hospitals, participation of patients is more required to cover the costs of health service consumed. Health services in those hospitals were more accessible to insured patients and with the capacity to pay USD150 and USD27 per year for hospitalization and consultation respectively. The POOP remained high in the two DRC hospitals compared to peer hospitals in other countries, indeed the intervention of health insurance was essentially oriented to private health insurance plans covered less patients and with capacity to pay contribution. Other health insurance schemes need to be improved in the UHC law to be adopted in DRC. The process of the development of those health insurance mechanisms will require a combination of new ICT tools for better management (Karara et al., 2014).



3/10/18 08:23

Discussion

6.4.4

207

Malian health facilities Despite the growing concern of the Malian authorities to move towards UHC, the public participation in the social health insurance was still low and the private health insurance shows a weak participation. On the national level, the population covered by a SHI is only 1.9%, and respectively 2.7% for CBHI and 1.7% for PHI schemes. Globally, the population health insurance coverage was less than 6% and therefore more than 94% did not have health insurance coverage. This situation is reflected on the level of health facilities studied where between 59% and 93% of patient encounters were not insured. The most applied health insurance scheme, as registered by the participating health facilities was the SHI represented by the AMO of Mali. Patients covered ranged from 2% for outpatients to 20% for inpatients in the studied health facilities. This PHSC indicator was higher in the two third reference hospitals, showing that insured patients more encountered the CHU-IOTA and HNFS than the health centers. The same situation was observed in Rwandan and Burundian hospitals (Karara et al., 2015), and in many developing countries, where social and private insured patients prefer the third-level reference and private hospitals more specialized and equipped than second- or first-level reference health facilities (Basumail et al., 2012). Regarding the financial risk protection, the World Bank and WHO statistics (2014-2015) reported an OOP of 25.90USD representing 61.7% of total expenditure on health in Mali. The OOP expenditure as proportion of total expenditure on health was above the 25%-threshold as it was also for patients studied in the 4 participating health facilities (82%). Health services were either less covered by AMO or in totality paid by the high number of uninsured patients. This PHSP was of course a little higher at the level of the patients encountered health facilities than in the general population due to the non-representativeness of the participating facilities. On other hand, the range of POOP in these facilities (USD6-USD56) included the OOP at the national level and in-POOP represents 8% of annual average income per capita. Health services are financially affordable for average patient, partially thanks to the funding to these health facilities (government, NGOs, donors, etc.). Furthermore, the improvement of the health insurance policy may increase the level of coverage and reduce the patient's share on health care expenses.

6.4.5

Case Mix in three reference hospitals Case mix evaluation results of comparison of morbidity, mortality and, financial burden and coverage for 5 pathology groups resulting to KPGS codes mapped from ICD-10 and ICPC-2 classifications in 3 reference hospitals (CHUK, CHURK and HPGRB). Those pathologies are monitored in the framework of UHC in countries to assess the progress in treatment intervention: 15B: Other complications of the pregnancy, childbirth and the puerperium. 04B: Diabetes Mellitus 09C: Hypertensive diseases 01B: Tuberculosis 01M: HIV Disease For the morbidity at CHUK, Other complications of pregnancy (15B) remained more represented than other studies pathology groups during the study period. The number of new cases increased by 9.6% between 2010 and 2014, and then decreased by 20.8% till 1679 cases in 2016. At national level, the data from DHIS2 show also an increasing cases of obstetrical problems in hospitals between 2010 (13902) and 2014 (22361) (Ministry of Health, Rwanda, 2014). Results for a study conducted at CHUK in 11 pathology groups (Verbeke et al., 2013) showed that Pregnancy related problems were more significantly represented among uninsured (43%) than insured (29%). The Tuberculosis morbidity rate has remained around 0.6% over the period of study, but the number



3/10/18 08:23

208


of new cases decreased by 65.4%: 86 cases have been registered in 2016. According to the World Data Atlas, the national trend has also shown a decrease in incidence of tuberculosis in the population, from 88 to 56 per 100000 inhabitants. This decreasing over the time is due to the goals of Tuberculosis control policy by improving cases management through availability of Tuberculosis drugs and close followup of patients under treatment (Ministry of Health, Rwanda, 2014). The morbidity of Diabetes decreased by 79.5% between 2010 and 2016. According to WHO data, the prevalence of Diabetes at the national level shows an upward trend since decades to reach 2.8% in 2016. Despite this increase at the national level, the number of cases treated at the CHUK (which is a national reference center) has greatly decreased following to the increasing of the centers of Diabetes screening and treatment (Rwanda Diabetes Association, www.rwandadiabetes.com). Concerning hospital morbidity rate of Hypertension, it increased from 2.3% in 2010 to 4.0% in 2016. The Rwanda Annual Health Statistics Booklet (2009-2014) also showed that the proportion of people in high blood pressure increased from 1.9% in 2009 to 6.4% in 2014 in Rwandan health facilities (Ministry of Health, Rwanda, 2014). A population-based national estimate of hypertension prevalence study conducted in 2016 shows a prevalence of 15.4% in a sample of 7116 participants (Nahimana et al., 2017). Hypertension was one of the top 10 pathologies diagnosed in hospitals of Rwanda in 2014. The number of new cases decreased by 36.9% at CHUK. As for Diabetes, centers for screening and treatment of hypertension and other cardiovascular diseases have increased with the acquisition of equipment of screening more and more efficient. The HIV disease morbidity rate declined by 86.7% between 2010 (4.5%) and 2016 (0.6%) at CHUK. Although HIV prevalence at the national level has declined since 2000, from 5.2% to 3.1% (World Data Atlas), this situation alone does not explain the sudden decrease in HIV disease-related morbidity observed in CHUK, considered as the reference center for HIV screening and treatment. The number of both public and private health facilities providing Antiretroviral Therapy (ARV) almost doubled between 2010 and 2015 from 295 to 540. This situation may be the main reason for the decreasing of HIV disease cases at CHUK over the 7 study-years. At the CHUK, the global mortality decreased from 6.3% in 2010 to 3.4% in 2016. The number of death cases and the mortality rate were high for patients with HIV Disease compared to others studied pathology groups. We observed an increase of Inpatient mortality rate (IMTR) from 1.6% to 4.6% for Other complications of pregnancy between 2015 and 2016. This increasing of mortality could be explained by the high level of morbidity during the same period. The mortality rate of the 5 pathology groups remained relatively lower than the global mortality rate of all pathologies. However, in the study conducted between 2009 and 2013 integrating 11 pathology groups (Verbeke et al., 2013), showed an IMTR of 7.11% for insured inpatients and 8.80% for uninsured inpatients. The IMTR showed a decreasing trend during the study period. The hospital death rate (ILTR) was also high for HIV Disease. Apart for Other complications of pregnancy where the death rate is low ( 90%) and Burundi (23%-30%) than in Mali (

ict-enabled monitoring and evaluation methods of

ict-enabled monitoring and evaluation methods of

Suggest Documents

Evaluation of Different Monitoring Methods of Laser ...

ACritical Evaluation of Methods of Monitoring Diabetic ... - Diabetes Care

Field Evaluation of Three Methods for Monitoring Populations of ...

Monitoring and Evaluation - InterAction

Monitoring Forsmark – evaluation and

Evaluation of monitoring methods available for safeguards ... - Julkari

Monitoring and Evaluation - InterAction

An evaluation of rapid methods for monitoring vegetation ... - IWMM

Evaluation of Surveillance Methods for Monitoring House ... - UCR CS

Evaluation of methods for monitoring drug resistance in chronic ...

IST EVALUATION AND MONITORING

COLLABORATIVE MONITORING AND EVALUATION

Evaluation of Surveillance Methods for Monitoring House ... - Cs.ucr.edu

Monitoring and evaluation - TeachingEnglish

IST EVALUATION AND MONITORING

Monitoring 'monitoring' and evaluating 'evaluation': an ...

Monitoring & Evaluation

Monitoring & Evaluation

Monitoring and Evaluation of Sexual and ...

Monitoring and evaluation - GIZ [PDF]

Agricultural Monitoring and Evaluation - OECD.org

Performance Monitoring and Evaluation in

Visual Traffic Monitoring and Evaluation

Monitoring and Evaluation Perspectives - ScholarlyCommons