Tuberculosis control in Bangladesh: success of the DOTS strategy

Transcription

1 INT J TUBERC LUNG DIS 2(12): IUATLD Tuberculosis control in Bangladesh: success of the DOTS strategy J. A. Kumaresan,* A. K. Md. Ahsan Ali, L. M. Parkkali *Global Tuberculosis Programme, World Health Organization (WHO), Geneva, Switzerland, Mycobacterial Disease Control, Ministry of Health and Family Welfare, WHO Representative Office, Bangladesh SUMMARY SETTING: Tuberculosis (TB) has been a major publichealth problem in Bangladesh for many decades. National control efforts in the past have not been successful, with less than half of detected cases being cured. In 1993, a project based on the DOTS (directly observed treatment, short-course) strategy was initiated for a population of approximately one million in a rural setting. Following a 78% cure rate in the initial cohort of new smear-positive patients, the project was expanded in phases to cover a rural population of 67 million in OBJECTIVES: Routine programme data on all new sputum smear-positive patients registered in the TB project since its inception until 1996 were analysed. Case finding results are presented until 1996, as are results of sputum smear conversion after 2 months of treatment in new smear-positive patients for the same cohort of patients. Final treatment outcome results were analysed for new smear-positive patients registered up to RESULTS: A total of patients were registered in the project during the 3-year period. Two-thirds of these were new smear-positive cases and 27% were new smear-negative patients. Sputum smear conversion in new smear-positive patients at 2 months was 85%; 5% remained smear-positive, 3% had died and the rest had no sputum examination. Final treatment outcome results in new smear-positive patients registered during showed that 75% were cured, 4% completed treatment but did not have a sputum smear result, 2% remained smear-positive, 6% died, 10% defaulted and 3% were transferred out. CONCLUSION: The DOTS strategy can be successfully implemented in phases in large countries with a high tuberculosis burden. This success is due to decentralizing sputum smear microscopy and treatment delivery services to peripheral health facilities, utilizing the existing primary health care network. High cure rates can be maintained despite rapid expansion of coverage, with proper implementation of the strategy and regular monitoring of reports on case finding, sputum smear conversion and treatment outcome. Case detection needs to be further increased by informing and involving the community in TB control efforts through social mobilization. KEY WORDS: tuberculosis; control; treatment outcome; DOTS strategy BANGLADESH, with a population of 120 million, is the most densely populated country in the world; more than three-quarters of its population lives below the poverty level. 1 The country is caught in a vicious cycle of poverty and ill health. Poverty combined with low literacy rates hampers productivity and growth. Despite these problems, national population and health programmes have made good progress in recent decades. Tuberculosis (TB) is a serious public health problem with grave socio-economic consequences. The Fourth Population and Health Project (FPHP) included TB as a component of Further strengthening of TB and leprosy control services. The FPHP spans a 5- year period ( ) and was financed by the Government of Bangladesh (GoB), the World Bank and the donor consortium of development partners. A national tuberculosis survey in 1966 estimated a prevalence of 500 per smear positive tuberculosis cases. 2 A more recent estimate in 1993, considering all prevalence surveys conducted in the country, indicated an annual incidence rate of 220 per (all TB cases). 3 Based on the estimated incidence, the expected number of new cases in 1995 was , whereas the nation-wide notification was only cases (16% of that estimated). Before 1993, tuberculosis services were provided by chest specialists in specialised TB hospitals and 44 district TB clinics serving a population of approximately million people per district. There were no standardized guidelines for categorisation and treatment of patients, and no systematic reporting of treatment outcome of patients using internationally accepted Correspondence to: Jacob A Kumaresan, MD, Global Tuberculosis Programme, World Health Organization, 20 Avenue Appia, CH-1211 Geneva 27, Switzerland. Tel: ( 41) Fax: ( 41) kumaresanj@who.ch Article submitted 9 January Final version accepted 8 July 1998.

2 Tuberculosis control in Bangladesh 993 definitions. The treatment generally utilized was a standard long-course non-rifampicin-containing regimen. Case finding and cure rates were not routinely evaluated. The average cure rate was estimated to be 25 40% and case finding was less than 10% of that expected. 4,5 Due to the magnitude of the TB burden in the country and the inadequate control efforts, the GoB made TB a high priority in the FPHP. The tuberculosis project The TB project, based on the World Health Organization (WHO) s strategy for effective TB control (DOTS directly observed treatment, short-course), was initiated in The key elements of the strategy included standardized case definitions, priority for detection of smear-positive patients by sputum smear microscopy, short-course chemotherapy regimens directly observed by trained health workers during the intensive phase of treatment, recording and monitoring of patients during treatment, and quarterly evaluation of reports on case finding and treatment outcome of cohorts of patients. The preparatory phase included development of the plan of action, preparation of technical guidelines and manual for TB control, training of personnel at various levels of the system, and provision of sufficient drugs, supplies and other materials to the initial demonstration areas where the DOTS strategy was introduced. In November 1993, the DOTS strategy was implemented in demonstration areas four thanas* covering a population of approximately one million people. Evaluation of treatment outcomes in the first cohort of 67 new sputum smear-positive patients registered in these thanas in early 1994 showed that 92.5% of these patients became smear-negative after 2 months of treatment. Based on the above success, the project expanded in phases with a coverage of 40 thanas (population 10 million) by the end of 1994, 181 thanas (population 45 million) by the end of 1995, and 261 thanas (population 67 million) by the end of During 1993 and 1994, the DOTS strategy was implemented only by the GoB at thana health complexes. In 1995, nongovernmental organisations such as the Bangladesh Rural Advancement Committee (BRAC) 6 and Leprosy Co-ordinating Committee (LCC organizations involved in leprosy control, such as the Damien Foundation and the Danish Bangladesh Leprosy Mission) began implementing the same policies as the project expanded. More recently, the district TB clinics have also become involved in the project, especially in *A thana covers approximately 200 to population. The thana health complex is the basic unit where TB services such as case detection, treatment and follow up are carried out. TB treatment cards, the TB register and laboratory register are maintained by a trained health worker at this unit, under the administrative and technical supervision of the Medical Officer. training and supervision of staff at thana and peripheral health facilities. The technical policies in the TB project are briefly summarised. 5 Patients with symptoms (productive cough for more than 3 weeks sometimes bloodstained and other symptoms such as weight loss, fatigue or night sweats) present themselves or are referred to the thana health complex from peripheral health services for clinical examination by the doctor. Those with suspected tuberculosis submit three sputum specimens for smear examination by Ziehl- Neelsen stain one spot specimen, an overnight specimen and another spot specimen on the following day when the patient returns to the laboratory at the thana health complex. All sputum-positive patients are registered and start treatment immediately. Those with three negative sputum smears undergo chest radiography at the thana health complex (if available) or at the district TB clinics or specialised TB hospitals. A diagnosis of smear-negative pulmonary TB is made by a physician on the basis of radiological abnormalities and clinical features. All smear-positive and smear-negative patients are registered in the tuberculosis register, and have a tuberculosis treatment card, which is maintained at the thana health complex. New adult smear-positive patients are treated with rifampicin, isoniazid, ethambutol, and pyrazinamide daily for 2 months (intensive phase). Those patients with a negative sputum smear at 2 months receive a further 6 months of treatment with isoniazid and thioacetazone daily (continuation phase). If a patient is smear-positive at 2 months, the intensive phase of treatment is continued for the third month, followed by a 5-month continuation phase. All drugs are provided free of charge to the patient. Each new sputum smear-positive patient undergoes a sputum smear examination at the end of the intensive phase (at 2 months), during the continuation phase (at 5 months) and at the end of the continuation phase (at 8 months). Patients who are smear-positive at 5 months are considered treatment failures and receive the retreatment regimen. Final treatment outcomes after 8 months of treatment were defined as follows: cured a patient who completed treatment and had at least two negative smears, one of them at completion of treatment; treatment completed a patient who completed treatment with no sputum smear examination at the end of treatment; failed a patient who remained smear positive or had become smear positive at 5 months or more after commencing treatment; died a patient who died during treatment, regardless of cause; defaulted a patient who, at any time after registration, had not collected drugs for 2 consecutive months or more; transferred out a patient who was transferred to another reporting unit and for whom treatment outcome is unknown. Regular supervision of the programme is con-

3 994 The International Journal of Tuberculosis and Lung Disease ducted by staff from the central and divisional levels. Supervisory visits are comprehensive. Staff carefully review the treatment cards, compare these with the tuberculosis and laboratory registers, and ensure that adequate quantities of drugs, supplies, laboratory reagents and recording/reporting forms are available. The divisional supervisors meet every month with central unit staff to assess programme performance through a critical review of quarterly reports on case finding, smear conversion after the intensive phase and final treatment outcome. Intake of medication by sputum smear-positive patients is directly observed daily during the intensive phase by health workers at thana health complex or at peripheral health units. Patients who cannot come daily to the health facilities receive their treatment from health assistants (government employees) in their villages. Health assistants are provided with one week s supply of drugs at a time for that patient and report to the thana health complex every week to collect a further supply. Another method of direct observation of treatment, mainly followed by NGOs, is using a guardian who is made responsible for the patient and is provided with drug supplies for a week or up to a month the guardian could be a trained community health worker in the village, 6 the supervisor at the work place, a religious leader in the village, a school teacher or a relative of the patient. Daily observation of treatment is recorded on the treatment cards at the thana health complex, or the information is transferred to these cards weekly from records that are kept by the supervisors. During the continuation phase, patients receive treatment once every 2 weeks, or monthly, from the thana health complex for self-medication. METHODS Routine programme data were reviewed from the implementation of the project until A retrospective analysis was conducted on data from tuberculosis registers and reports on case finding, smear conversion and final treatment outcome. The aim was to assess programme performance through analysis of routinely collected data, and no attempts were made to obtain additional information that was missing. Results on case finding data were analysed up to A total of TB cases were registered in the project during this period. Results on sputum smear conversion after 2 months of treatment were analysed for new smear-positive patients registered in the above cohort. Final treatment outcome results were analysed for new smear-positive patients registered up to RESULTS The TB project was initiated in four thanas in November 1993, and was gradually expanded in phases, covering 261 of the 497 thanas by the end of Of these 261 thanas, 171 (66%) were covered by the GoB and the rest by NGOs. Case finding at the thanas increased as the project expanded. Data from the first 10 thanas where the project was initiated indicate that new smear-positive case rates increased from 30/ in 1994 to 40/ in 1996 (range 9 60). A total of patients were registered during the 3-year period (Table 1). New smear-positive patients accounted for 66% of the cases, and new smear-negative patients 27% of the total, indicating that the priority for case finding is the detection of sputum smear-positive patients. Table 2 gives the sputum smear conversion results after 2 months of treatment for the new smear-positive patients registered in the project. Sputum smear conversion in new smear-positive patients at 2 months was 85%, 5% remained smear-positive, 3% died and the rest had no sputum smear examination. The final treatment results are given in Table 3 for new smear-positive patients registered in the project up to The final outcome of treatment in the cohorts of new smear-positive patients registered during shows that 75% were cured, 4% completed treatment, 2% failed, 6% died, 10% defaulted and 3% were transferred out. These results included 95% of patients registered in the cohorts for sputum smear conversion and 92% of patients registered in the cohorts for final treatment outcome. DISCUSSION Bangladesh is a country faced with many challenges, including poverty, illiteracy, natural disasters, and Table 1 Case finding in TB project, Year No. of thanas implementing DOTS New smear Relapse New smear Extra pulmonary Total (%) (66) (2.5) (27) (4.5) (100) Total

4 Tuberculosis control in Bangladesh 995 Table 2 Smear conversion after 2 months of treatment Year Total Smear (%) Smear Died Defaulted Transferred out (92.5) (87) (84) (86) Total (%) (85) lately, political instability. Despite these problems, the TB project has had considerable success over the past years. This is due to the strong political commitment of the GoB and the priority given to TB. The central unit and intermediate levels of the system were strengthened by recruitment of expatriate and national consultants for the project. Staff involved in TB control at all levels were properly trained and supported by regular supervision in the implementation of the DOTS strategy in the field. The rigorous monitoring system for recording the progress of individual patients and the system of regular quarterly reporting on case finding/treatment outcomes identifies weaknesses in programme implementation. This enables managers and supervisors to work with staff in the field to overcome these weaknesses. Prior to 1993, TB control activities were organised in TB hospitals and clinics, largely at district level. Follow up of patients from these centres was extremely difficult, resulting in a large number of defaulters, and as a consequence chronic cases. By implementing the DOTS strategy at thana level utilising existing government health services, the project has succeeded in bringing TB services closer to patients homes. Thus the default rate has fallen to 10% in the project areas. Noting the successes achieved by GoB, NGOs such as BRAC and LCC are now implementing the DOTS strategy. Following initial successes in demonstration areas, they are now expanding their TB services to cover larger populations. This is a good example of co-operation between the public and other sectors for TB control. Reports on sputum smear conversion and final treatment outcome were received from 95% and 92%, respectively, of the cohort of new smear positive patients registered. Failure to achieve complete reporting is due to the fact that some thana reports did not arrive in time for national cohort analysis. The sputum smear conversion and cure rates in new smear positive patients in the project have remained at around 85% and 75%, respectively, despite the rapid expansion of the project and involvement of various partners. This indicates that quality of services can be maintained under field conditions through adequate training and regular supervision of all components of programme delivery. These results are comparable with other shortcourse chemotherapy programmes, for example 71% cured in Mozambique from , and 77% cured in Tanzania from Sputum smear conversion after 2 months of treatment was 86% in Linoning Province, China, in However, cure rates are lower than the national and global targets of 85%, 9 and the results of directly observed shortcourse chemotherapy in China. 10 More effort is necessary to reduce the death and defaulter rates by detecting cases earlier and educating patients and their families on the importance of completing the full course of treatment. Operational research on the determinants of adherence to treatment and various methods to observe treatment intake would help to achieve further success. Such lessons may be learnt from the TB projects implemented by BRAC, where all patients are directly observed by trained community health workers. In the GoB implemented areas, around 30% 50% of patients are receiving directly observed treatment at health facilities, and the remainder from health assistants or guardians in the community. 11 The average case finding rates at thana health complexes have increased to 40/ smear-positive patients during the past 2 years. However, it is much lower than the expected incidence rate of 100/ smear-positive patients, as TB services have recently been introduced at thana level and it is expected that case finding will increase as community Table 3 Final outcome after full course of treatment Year Total Cured Treatment completed Died Failure Default Transferred out Total (%) (100) (75) 421 (4) 612 (6) 183 (2) (10) 297 (3)

5 996 The International Journal of Tuberculosis and Lung Disease Table 4 Training Cadre Course type/duration Total Medical Officers/ Consultants Directors of Health (divisional and district levels) Laboratory technicians Health and sanitary inspectors, statisticians, district store keepers Health assistants, store keepers National and district level management 6 days Tuberculosis orientation course 1 day TB laboratory course 6 days Mid-level supervisor course 2 days Field level supervisors course 1 day Total awareness on the availability of services improves. Increasing case finding to attain the national and global targets of 70% should be the next priority of the programme. Furthermore, community awareness and involvement in the programme will increase the feasibility of directly observing the intake of medication by the patients. With the experience gained thus far, the challenge now is to achieve nation-wide coverage by the TB project while maintaining the quality of services. Adequate training of staff is critical in this respect. Until mid 1997, a total of staff of various cadres in different levels of the system were trained. Table 4 provides a detailed summary of the number of different cadres of staff trained each year. In addition, continuous support has been provided for trained staff through regular supervision and monitoring visits, mainly by expatriate and national consultants recruited for the project. It is now necessary to strengthen and build the capacity of GoB staff with this expertise in order to ensure the sustainability of the project; involvement and co-operation with NGOs is essential in this respect. CONCLUSION The TB project will be expanded to achieve countrywide coverage in the next few years. Collaboration between public and private sectors should be strengthened, learning from the co-operation between the NGOs and the public sector. So far the project has not been introduced in an urban setting, and the role and method of cooperation with the private sector in large cities needs to be studied urgently. During the expansion of the project, high quality should be maintained in the services. Furthermore, the project should explore the potential to integrate functions such as training, supervision and logistics into the sector-wide health and population strategy proposed for the next fiveyear health plan. Based on the experience in Bangladesh, it can be concluded that the DOTS strategy can be implemented in a low-income, high TB incidence country with a large population. Compared with the estimated case finding and cure rates prior to the project, very impressive results have been obtained thus far with the DOTS strategy in a large number of patients. The lessons learnt from this project support the view that all the key elements of the strategy need to be introduced and monitored to ensure the success of the project. Varied approaches to observe patients intake of drugs are necessary, and successful implementation of these approaches in demonstration areas should lead to expansion on a wider scale. The key to monitoring the programme implementation is through analysis of reports on case finding, sputum smear conversion and treatment outcome, for quarterly cohorts of new smear positive patients. The Bangladesh TB project demonstrates that the DOTS strategy can achieve high cure/treatment completion results in varied cultural and geographical settings, as shown previously in Africa and China. 10,12,13 References 1 United Nations Children s Fund. The state of the world s children Oxford University Press, National Tuberculosis Control and Research Project. Report of the tuberculosis survey in Bangladesh Government of the People s Republic of Bangladesh, Dacca, Kumaresan J A, Raviglione M C, Murray C J L. Tuberculosis. In: Murray C J L, Lopez A D, eds. Global Health Statistics Global Burden of Disease and Injury Series. Volume 2. Boston, MA; Harvard University Press 1996: pp Veen J, Becx-Bleumink M. The national tuberculosis programme of Bangladesh: report of a consultancy visit to Bangladesh for the World Bank. The Hague; KNCV, 1990.

6 Tuberculosis control in Bangladesh Bangladesh Ministry of Health and Family Welfare. National guidelines for tuberculosis control (2nd ed), Chowdhury A M R, Chowdhury S, Islam M N, Islam A, Vaughan J P. Control of tuberculosis by community health workers in Bangladesh. Lancet 1997; 350: Murray C J L, de Jonghe E, Chum H J, Nyangulu D S, Salomao A, Styblo K. Cost effectiveness of chemotherapy for pulmonary tuberculosis in three sub-saharan African countries. Lancet 1991; 338: Feng-Zeng Z, Levy M H, Sumin W. Sputum microscopy results at two and three months predict outcome of tuberculosis treatment. Int J Tuberc Lung Dis 1997; 1: Kochi A. The global tuberculosis situation and the new control strategy of the World Health Organization. Tubercle 1991; 72: China Tuberculosis Control Collaboration. Results of directly observed short-course chemotherapy in Chinese patients with smear-positive tuberculosis. Lancet 1996; 347: Chowdhury A M R, Alam A, Chowdhury S A, Ahmed J. Tuberculosis control in Bangladesh. Lancet 1992; 339: Murray C J L, Styblo K, Rouillon A. Tuberculosis in developing countries: burden, intervention and cost. Bull Int Union Tuberc Lung Dis 1990; 65(1): Enarson D A. The International Union Against Tuberculosis and Lung Disease model National Tuberculosis Programmes (editorial). Tubercle Lung Dis 1995; 76: RÉSUMÉ CADRE : Depuis de nombreuses décennies, la tuberculose est un problème majeur de santé publique au Bangladesh. Les efforts nationaux du contrôle n ont pas été couronnés de succès dans le passé, puisque moins de la moitié des cas détectés étaient guéris. En 1993, un projet de tuberculose basé sur la stratégie de DOTS a été mis en route pour une population d environ un million d habitants dans un cadre rural. A la suite de la mise en évidence d un taux de guérison de 78% dans la cohorte initiale de nouveaux patients à bacilloscopie positive, le projet s est élargi par phases pour couvrir une population rurale de 67 millions de personnes en OBJECTIFS : Nous avons analysé les données de routine du programme sur tous les nouveaux cas à bacilloscopie positive enregistrés dans le projet de tuberculose depuis son début, jusqu à Les résultats du dépistage sont présentés jusque 1996, de même que, pour la même cohorte de patients, les résultats de négativation de l expectoration à l examen direct après deux mois de traitement dans les nouveaux cas à bacilloscopie positive. Les résultats finaux du traitement sont analysés pour les nouveaux patients à bacilloscopie positive enregistrés jusque RÉSULTATS : Au total, patients ont été enregistrés dans le projet au cours de la période de 3 ans. Les deux tiers de ces patients étaient positifs à l examen direct, et 27% étaient de nouveaux cas à bacilloscopie négative. Le taux de conversion des frottis d expectoration chez nouveaux cas à bacilloscopie positive est de 85% après 2 mois, 5% restant positifs à l examen direct, 3% étant décédés, et le reste n ayant pas eu d examen de crachat. Les résultats finaux du traitement chez nouveaux patients à bacilloscopie positive enregistrés durant la période , ont été les suivants : 75% de guérison, 4% de traitement achevé sans confirmation bactériologique, 2% d échec (bacilloscopie positive), 6% de décès, 10% d abandon, et 3% de transfert. CONCLUSIONS : La stratégie DOTS peut être appliquée avec succès par phases dans de grands pays avec une lourde endémie tuberculeuse. Ce succès est dû à la décentralisation de l examen microscopique direct de l expectoration et des services d administration du traitement vers les services de soins périphériques, et par l utilisation du réseau de santé primaire existant. Des taux élevés de guérison peuvent être maintenus malgré une expansion rapide de la couverture, avec une mise en oeuvre correcte de la stratégie et un suivi régulier des rapports sur le dépistage, la négativation bactériologique, et l issue du traitement. La détection des cas doit être encore augmentée davantage, en informant et en impliquant la communauté dans les efforts de contrôle de la tuberculose au travers d une mobilisation sociale. RESUMEN MARCO DE REFERENCIA : La tuberculosis es un problema mayor de salud pública en Bangladesh desde hace varias décadas. Los esfuerzos nacionales de control en el pasado no han sido exitosos y menos de la mitad de los casos detectados se curan. En 1993 se inició un programa de TB basado en la estrategia del DOTS para una población rural de alrededor de un millón de habitantes. Después de la constatación de un índice de curación del 78% en una cohorte inicial de pacientes nuevos con baciloscopia positiva, el proyecto se expandió a una población rural de 67 millones en OBJETIVOS : Se analizaron los datos de rutina del programa de todos los pacientes nuevos registrados en el proyecto TB desde su inicio hasta Se presentan los resulados de la detección hasta 1996 y los resultados de la conversión del esputo después de 2 meses de tratamiento en los pacientes nuevos con baciloscopia positiva de la misma cohorte. Se analizó el resultado final del tratamiento de los enfermos con baciloscopia positiva hasta RESULTADOS: En el período de 3 años se registraron en el proyecto un total de pacientes. Los dos tercios de estos pacientes eran casos nuevos, con baciloscopia positiva, y el 27% eran casos nuevos con baciloscopia negativa. La conversión del esputo en pacientes nuevos con baciloscopia positiva fue

7 998 The International Journal of Tuberculosis and Lung Disease del 85% a los 2 meses, permanecieron positivos el 5%, hubo el 3% de muertes y el resto sin control del esputo. El resultado final del tratamiento mostró que en pacientes nuevos con baciloscopia positiva en , el 75% había curado, el 4% completó el tratamiento sin control del esputo, el 2% siguió siendo positivo, el 6% murió, el 10% abandonó y el 3% fue transferido. CONCLUSIÓN : La estrategia DOTS puede ser implementada en grandes países con una elevada carga de TB. Este éxito se debe a la descentralización de los servicios de examen baciloscópico del esputo y de administración del tratamiento hacia la periferia, utilizando los centros primarios de salud de la red existente. Se pueden mantener altas tasas de curación a pesar de la rápida expansión de la cobertura, con una buena explicación de la estrategia y un buen control de los informes sobre la detección de casos, la conversión del esputo y el resultado del tratamiento. La detección de los casos puede aún ser mejorada informando e involucrando a la comunidad en los esfuerzos de control de la TB a través de la movilización social.