Elsevier

The Lancet

Volume 357, Issue 9257, 3 March 2001, Pages 664-669
The Lancet

Articles
Effectiveness of the direct observation component of DOTS for tuberculosis: a randomised controlled trial in Pakistan

https://doi.org/10.1016/S0140-6736(00)04129-5Get rights and content

Summary

Background

DOTS is the control strategy for tuberculosis promoted by WHO. Pakistan is currently developing its National Tuberculosis Programme, and requires guidance on types of direct observation of treatment appropriate for the local conditions. We did a randomised trial to assess the effectiveness of different packages for tuberculosis treatment under operational conditions in Pakistan.

Methods

We enrolled 497 adults with new sputum-positive tuberculosis. 170 were assigned DOTS with direct observation of treatment by health workers; 165 were assigned DOTS with direct observation of treatment by family members; and 162 were assigned self-administered treatment. The trial was done at three sites that provide tuberculosis services strengthened according to WHO guidelines for the purposes of the research, with a standard daily short-course drugs regimen (2 months of isoniazid, rifampicin, pyrazinamide, and ethambutol, followed by 6 months of isoniazid and ethambutol). The main outcome measures were cure, and cure or treatment completion. Analysis was by intention to treat.

Findings

Within the strengthened tuberculosis services, the health-worker DOTS, family-member DOTS, and self-administered treatment strategies gave very similar outcomes, with cure rates of 64%, 55%, and 62%, respectively, and cure or treatment-completed rates of 67%, 62%, and 65%, respectively.

Interpretation

None of the three strategies tested was shown to be superior to the others, and direct observation of treatment did not give any additional improvement in cure rates. The effectiveness of direct observation of treatment remains unclear, and further operational research is needed.

Introduction

Tuberculosis remains the commonest cause of death in adults in less developed countries—south Asia being the region worst affected. Numbers of cases are increasing due to population growth, HIV, and inadequate treatment. Patients' compliance with treatment is commonly poor, and non-completion of a course can lead to relapse, possibly with drug-resistant bacilli. Multidrug resistance is an increasing problem.1

WHO and the International Union Against Tuberculosis and Lung Disease advocate use of the DOTS strategy to control tuberculosis, and recommend that no rifampicin-containing (short-course) tuberculosis treatment should be given without direct observation of treatment.2, 3 The DOTS strategy is made up of five operational components (together with government commitment to tuberculosis control): diagnosis and follow-up through sputum microscopy; use of standardised short-course drug regimens; regular uninterrupted supply of drugs; accurate cohort recording and analysis; and direct observation of treatment by people responsible to the health services (which, in most settings, refers to health-facility staff). The DOTS package with outpatient direct observation of treatment has been shown to be successful through many observational studies in South Africa,4 China,5 Bangladesh,6, 7 and elsewhere.8 All components other than direct observation of treatment have been used for many years, and are non-controversial. The direct observation of treatment component is intended to address patients' non-compliance, and has two important elements: observation of every scheduled dose of medication a patient takes during at least the initial 2-month intensive phase of treatment, and the giving of positive encouragement to patients to ensure treatment completion.

There is some doubt as to whether direct observation of treatment is essential, or indeed effective, in reducing patients' non-compliance. Such a strategy imposes burdens on patients and health services through prolonged admission or frequent attendance at clinics. In rural and periurban areas, there are substantial financial costs and time-costs involved in travelling to a clinic daily for 2 months to receive treatment.9 Furthermore, patients can be reluctant to attend repeatedly because of the stigma attached to tuberculosis in most countries.9, 10 These additional burdens on patients can only be defended if direct observation of treatment improves cure rates and is not a barrier to care-seeking; there is currently little evidence to suggest that this is the case.11 Results from the only two randomised controlled trials of direct observation of treatment published to date paint a confusing picture. In Thailand, direct observation of treatment was shown to give significantly higher cure rates than self-administered treatment,12 but the direct observation included home visits by health workers—a strategy that is not feasible in most developing countries. In South Africa, no additional benefit was obtained with twice weekly observed treatment, compared with weekly follow-up visits to clinics,13 but results from this study remain controversial14, 15 because of the high numbers of patients refusing to join the trial.

In Pakistan, tuberculosis services have been neglected for years, and treatment completion is low. The National Tuberculosis Programme has adopted DOTS as policy, but the specifics have not yet been addressed. We therefore undertook a randomised controlled trial in three sites in Pakistan to assess the effectiveness of different packages for tuberculosis treatment under the operational conditions proposed by the National Tuberculosis Programme. These packages, although planned, have not yet been implemented in Pakistan—ie, within services strengthened to represent operational conditions appropriate to developing countries that have implemented national programmes.

Tuberculosis patients in Pakistan come from all agegroups, both sexes, and all walks of life. Since poverty is wide spread, most patients are very poor: the majority are daily labourers or housewives. There is substantial stigma towards those with tuberculosis, and this stigma is particularly strong for women. Unmarried women in particular experience many cultural difficulties in accessing health services. Thus it is important to devise treatment strategies that impose little additional financial burden on patients, keep the impact of stigma to a minimum, and give patients improved access to care.

Section snippets

Patients

Three diagnostic centres were selected for the trial: the Federal Tuberculosis Centre, Rawalpindi (where the National Tuberculosis Programme is based); the Red Crescent clinic in Sahiwal; and the Gujranwala Tuberculosis Centre. Sites were selected mainly for convenience, but were regarded as typical of tuberculosis treatment centres in Pakistan. All sites served a mix of urban and rural patients, but since urban Rawalpindi was a WHO-sponsored “demonstration” site, patients in the demonstration

Results

497 patients were enrolled in the trial; no eligible patients refused to participate. The numbers of patients enrolled at Rawalpindi, Gujranwala, and Sahiwal were 165, 155, and 177, respectively. The numbers of patients enrolled in the health-worker DOTS, family-member DOTS, and self-administered groups were 170, 165, and 162, respectively (figure). There was little difference in the distribution of demographic factors of enrolled patients by trial group (table 1).

27 (16%) of patients assigned

Discussion

The main findings were that there were no significant differences in cure rates and cure and completion rates between the three strategies tested. There was no additional benefit from direct observation of treatment over and above service strengthening, whether supervision was by health workers or family members. These were unexpected results; we envisaged that appropriate direct observation of treatment strategies would show increases in cure rates, compared with self-administered treatment.

Cited by (193)

  • MDR-TB in Pakistan: Challenges, efforts, and recommendations

    2022, Annals of Medicine and Surgery
    Citation Excerpt :

    PPM model led to increased case notifications in the age groups of <15 years and >54 years [19]. It also initiated Directly Observed Treatment, Short course (DOTS) in all the provinces according to WHO guidelines in which the patients are given short course drugs along with careful observation of treatment by the health care workers [20]. In a recent conference, namely End TB Summit held on January 18th, 2021, president, Dr Arif Alvi addressed the ongoing TB crisis by mentioning the number of TB cases detected annually estimated to be around 350 000 and emphasized on how finesse across the globe could be brought together to bringing together to help Pakistan put an end to TB by 2030 [21].

  • Individualizing Tuberculosis Therapy

    2017, Individualized Drug Therapy for Patients: Basic Foundations, Relevant Software and Clinical Applications
  • A systematic review of non-pharmacological interventions to improve therapeutic adherence in tuberculosis

    2019, Heart and Lung
    Citation Excerpt :

    Treatment duration was generally between six and nine months with widely varying interventions. However, they can be grouped into several categories: education (training for non-health workers or volunteers on tuberculosis, educational programmes on patient adherence, training health professionals),58,63,64,69,72,89 psychological support (self-help groups and self-esteem counselling),54,74 new technologies (messages to the mobile phone of the patient and medication event monitoring systems),53,55,56,78 DOT (by family members, volunteers, health workers and teachers),57,67,68,71,75,77,83,84,86,87 incentives (cash, food vouchers and transportation),61,62,89 improvements in access to healthcare services (decentralise treatment centres and supervise health centres),65 with the control group in most cases receiving the usual treatment according to the protocol of the country. Some of the papers presented more than one intervention or a combination of interventions compared with the control group.56,58,78,83,84

View all citing articles on Scopus
View full text