Deployment at community level of artemether-lumefantrine and rapid diagnostic tests

Raya valley, Tigray, Ethiopia
World Health Organization, 2009

CONTENTS

Acknowledgements

Preface

Executive summary

Introduction

Study area
Geography and demography
Potential effect on vector control of malaria epidemic during the study
Health service infrastructure in Tigray

Methods
Study design
Data collection

Results and discussion
Profile of community health workers and training outcomes
Community treatment services
Serial surveys of seasonal prevalence of malaria and of knowledge, attitude and practices
Performance of rapid diagnostic tests
Adherence of patients to the 6-dose artemether-lumefantrine regimen
Pharmacovigilance
Gealth impact of use of rapid diagnostic tests and artemether-lumefantrine
Mortality surveys
Challenges to assessing impact
Satisfaction of beneficiaries of the project
Cost of diagnosis clinically and with the rapid diagnostic test at community level

Lessons learnt
Study design
Project personnel
Database management
Logistics
Project management
Capacity-building
Effect of low incidence of malaria in Phase II of the project

Project budget
Project budget input and expenditure
Financing and supplies
Operational costs

Monitoring and evaluation

Conclusions

Recommendations

ACKNOWLEDGEMENTS

The Tigray Health Bureau, the World Health Organization (WHO) through its Global Malaria Programme, the Italian Ministry of Labour, Health and Social Policies – Health Sector, Novartis Farma SPA (Italy), the Department of Preventive Medicine in Migration, Tourism and Tropical Dermatology Centre of Mekelle are grateful to all the i9nstitutions and individuals that contributed their time, energy and funding to make this project successful.

Special thanks are due to Dr Tedros Adhanom, Minister of Health of Ethiopia, Dr Gebre Ab Barnabas, Head of the Tigray Health Bureau, Mr Hailemarian Lemma, Mr Alem Desta and Mr Asefaw Getachew of the Malaria Control Department of the Tigray Health Bureau, and Dr Andrea Bosman and Dr Wilson Were of the WHO Global malaria Programme, for their dedicated efforts throughout the project. Monitoring and follow-up of the project would have been impossible without the involvement of Dr Gianfranco Costanzo of the Italian Ministry of Labour, Health and Social Policies – Health Sector department of innovation, Ms Angela Bianchi (Novartis Farma SPA, Italy), Dr Anne-Claire Marrast (Novartis Pharma AG, Basel) and Professor Aldo Morrone and Dr Luigi Toma (Department of Preventive Medicine in Migration, Tourism and Tropical Dermatology of the San Gallicano Hospital in Rome), Dr Nathan Mulure (Novartis Pharma, Nairobi, Kenia), Mr Roberto Ferrara (Novartis Farma SPA, Italy), Mr Ambachew Medhin (WHO Ethiopia) and Dr Giacomo Stefanoni (intern, Department of Preventive Medicine in Migration, Tourism and Tropical dermatology, San Gallicano, Rome). These individuals also provided administrative and technical guidance, comments to strengthen the project and continuous monitoring of activities through regular teleconferences.

Grateful thanks are also due to Mr Hailemariam Lemma, Mr Alem Desta, Mr Goitom Mehari and Mr Abrha Kahsay for their personal commitment in preparing operational manuals for the different project tasks, training participants and regularly tracking project activities, including data management, and to Mr Berhane Hailesilassie for designing, conducting and summarizing the focus group discussions. Professor Peter Byass and Dr Edward Fottrell (Umea University, Sweden) were responsible for the design, implementation and analysis of the mortality survey; and Ms Appia Augustina Appiah-Danquah (University of Ghana Medical School, Accra) contributed to the mid-term evaluation of the pharma-covigilance activities. The contributions of the Italian Ministry of Labour, Health and Social Policies – health Sector and of Novartis Farma SPA (Italy), which provided financial support to this project are gratefully acknowledged.

Above all, however, the key elements in the success of this project were the community health workers, who generously volunteered their and their families’ valuable time, resources and efforts to best serve people affected by malaria in remote rural areas. Last but not least, this project would not have been possible without the full involvement and continuous support of the communities, the health authorities and the health workers in both districts throughout the project.

PREFACE

Since 2004, WHO has promoted strategies for improving access to treatment through home management of malaria, in order to ensure prompt diagnosis and effective treatment for people living in areas with limited access to general health services. Such situations are common in remote rural areas, where the burden of malaria is highest. Experience in many countries during the 1990s showed that this strategy, implemented with chloroquine or sulfadoxine-pyrimethamine provided by community health workers, community drug distributors, accredited medicine sellers and even mothers as coordinators, can affect mortality and severe morbidity due to malaria and considerably reduce the burden on health facilities.

The introduction of artemisinin-based combination therapies, which cost at least 20 times more than conventional antimalarial agents, has reduced home-based management of malaria in recent years , as these highly effective medicines are administered mainly by the general health services. A few countries in Asia (e.g. Cambodia and Viet Nam) have used these therapies in home-based management of malaria, but no African countries have done so.

In order to evaluate the feasibility and effect of implementing this strategy in Africa, a community-based observational study was designed for application in Tigray, northern Ethiopia. In line with international best practice, an important component of this project was use of rapid diagnostic tests to confirm a diagnosis of malaria before treatment with artemisinin-based combination therapies as part of home-based management of malaria.

This region was chosen because a large-scale, community based malaria control programme in 1994-2002 involving treatment in the community by over 700 health workers had been completed successfully. Both projects were rooted in a strong spirit of community involvement and participation, which pervades all aspects of the life of the people of Tigray.

This report describes the approaches and the results obtained, including the evidence base and a detailed description of the project, to provide guidance for extension of this experience to other regions of Ethiopia and other countries of Africa.

The communities in the project area have benefited; they appreciated the service provided by the community health workers and the availability of free diagnosis and effective antimalarial treatment near their homes. The challenge now will be to sustain this approach to malaria control, which has proven to be feasible and effective, with multiple sources of funding, thus helping the people to best take care of themselves.

EXECUTIVE SUMMARY

In July 2004, following a survey at 11 sentinel sites in Ethiopia in 2003, which showed high failure rates with sulfadoxine-pyrimethamine (up to 71.8% on day 28 of follow-up), the Ethiopian Ministry of Health issued new guidelines for the treatment of malaria. In the new policy, sulfadoxine-pyrimethamine was replaced by artemether-lumefantrine as first-line therapy. This new drug is more expansive than sulfadoxine-pyrimethamine; thus, so that it would be used effectively, the policy stated that a diagnosis of malaria must be confirmed by parasitological examination, either by microscopy or by rapid diagnostic testing. This posed a challenge in Tigray in northern Ethiopia, where management of malaria close to patients’ homes in remote rural areas was provided by well-established community-based treatment by volunteer health workers. Additional concern was raised about widespread, possibly irrational drug use, which can lead to the development of resistance to artemether-lumefantrine.

Because of the high cost of the new drug, the risk for greater drug pressure, potential adverse drug reactions in communities with no adequate safety monitoring system and exposure of uninfected patients, it was decided to conduct operational research to evaluate the role of the existing community health worker service in providing access to artemisinin-based therapies to widely dispersed, poor, predominately rural populations with poor access to health-care services. A project was therefore designed to improve the skills of community health workers, teaching them to use simple rapid diagnostic tests, to administer artemether-lumefantrine safely and to report regularly to staff in the general health services. If the performance of community workers with rapid diagnostic tests was as good as that in peripheral health institutions, the approach would ensure early diagnosis and treatment of malaria in outlying rural areas, decreasing the burden of malaria and promoting community participation in malaria control.

To validate the feasibility of this approach, a pilot operational research project was launched in May 2005 in Tigray, involving provision of artemether-lumefantrine after a simple rapid diagnostic test (ParaCheck-Pf cassette) at community level by volunteer health workers. Two approximately homogeneous districts were selected: Alamata, the intervention district, and Raya Azebo, the control district.

The intervention was initiated at the end of April 2005. during the first year (phase I), 33 community health workers provided prompt treatment with artemether-lumefantrine on the basis of clinical diagnosis alone. During the second year of the project (Phase II), 50% of the health workers were equipped with and trained to use rapid diagnostic tests, while the remainder continued to make diagnosis solely on the basis of clinical evidence. In the control district (Raya Azebo), many health workers were gradually demobilized from the service, in line with the new national treatment policy.

The project was multifaceted. It included improving the knowledge and skills of community health workers in the diagnosis and case management of malaria, providing them with a rapid diagnostic test and artemether-lumefantrine and building the capacity of malaria managers through short- and long-term post-graduate training courses. In addition, a number of surveys were undertaken to evaluate the prevalence of malaria infection, adherence to treatment regimens, performance of the rapid diagnostic test, morbidity and mortality from malaria and pharmacovigilance. “Focus group” discussions were used to measure community satisfaction and perceptions.

In the intervention district, the community health workers treated approximately 58% of al suspected and confirmed cases of malaria, representing, during the 26 months of the project, 76654 patients. This resulted in a lower patient load at health institutions in the intervention district (54774) than in the control district (101535), reducing the burden on health services in the intervention district.

During Phase II of the project, the health workers who were equipped with rapid diagnostic tests screened 5123 patients and identified 526 cases of P. falciparum malaria (10.3% positivity). The positive cases were treated with artemether-lumefantrine, and negative cases were given chloroquine for possible P. vivax infection, as per the new treatment policy. During the same period, the remaining community health workers in the intervention district treated 10475 patients for malaria on the basis of clinical diagnosis alone. If the positivity rate among patients treated solely on the basis of clinical diagnosis with the rapid diagnostic test (10.3%), up to 9397 patients might have been inappropriately treated with artemether-lumefantrine.

A comparison of diagnosis with the rapid diagnostic test and by expert microscopy showed that the test performed very well, with 96.4% sensitivity, 76% specificity, a positive predictive value of 59.5% and a negative predictive value of 98.2%. the theoretical saving made by equipping community health workers with rapid diagnostic tests was calculated to be US$ 1.41 per patient examined (based on the public sector price of Coartem [20 mg artemether-129 mg lumefantrine tablets, Novartis Pharma AG] in 2001-2006). Moreover, blood safety procedures and the diagnostic performance of community health workers using Paracheck-Pf were good and quantitatively similar to those of personnel in health institutions.

Patients treated by the community health workers had similar adherence to the treatment regimen as those treated at health facilities, although the figures were low. During an initial survey, adherence to the six-dose regimen by patients treated by community health workers was poorer than those treated in health facilities (19.4% versus 8.4% definitely not adherent).

The project was initiated coincidentally at the time of a malaria epidemic, which made it possible to evaluate the effectiveness of the invention. Although the two districts had similar eco-epidemiological conditions at the start of the epidemic and an overall reduction in malaria transmission during the post-epidemic year (second year of the project), the intensity of transmission and the severity of the epidemic were lower in the intervention district. During the first year of the project, the crude parasite rate was threefold lower in the intervention than in the control district. The adjusted incidence rate ratio for mortality from malaria at the end of the two years, assessed in a household mortality survey, was 0.60 (p=0.013; 95% confidence interval: 0.4-0.9), indicating that 40% fewer deaths due to malaria occurred in the intervention district than in the control district.

In conclusion, distribution of artemether-lumefantrine by community health workers using simple rapid diagnostic tests at community level for a widely dispersed, poor, primarily rural, hard-to-reach population is feasible if the health workers are appropriately trained, equipped with simple tests and supported by frequent supervision. The findings of this study indicate that the community health worker service should be upgraded by giving them artemisinin-based combination therapies, such as artemether-lumefantrine, and rapid diagnostic tests. Such interventions in areas of low-to-moderate malaria transmission can reduce morbidity, mortality and transmission of malaria as well as the case loads of the general health services. The use of rapid diagnostic tests can therefore generate cost savings, and programmes in which both artemisinin-based combination therapies and rapid diagnostic tests are used are less expensive than those based only on clinical approaches for the management of malaria.