Faculty of Health Sciences, Stellenbosch University, Francie van Zijl Drive, Tygerberg 7505, South Africa; Health Systems Research Unit, South African Medical Research Council, Norwegian Knowledge Centre for the Health Services, Francie van Zyl Drive, Parrow 7505, South Africa; Cochrane Centre, South African Medical Research Council, Francie van Zyl Drive, Parrow 7505, South Africa
Lutge, E., Faculty of Health Sciences, Stellenbosch University, Francie van Zijl Drive, Tygerberg 7505, South Africa; Lewin, S., Health Systems Research Unit, South African Medical Research Council, Norwegian Knowledge Centre for the Health Services, Francie van Zyl Drive, Parrow 7505, South Africa; Volmink, J., Faculty of Health Sciences, Stellenbosch University, Francie van Zijl Drive, Tygerberg 7505, South Africa, Cochrane Centre, South African Medical Research Council, Francie van Zyl Drive, Parrow 7505, South Africa
Background: Poverty undermines the adherence of patients to tuberculosis treatment. A pragmatic cluster randomized controlled trial was conducted to investigate the extent to which economic support in the form of a voucher would improve patients' adherence to treatment, and their treatment outcomes. Although the trial showed a modest improvement in the treatment success rates of the intervention group, this was not statistically significant, due in part to the low fidelity to the trial intervention. A qualitative process evaluation, conducted in the final few months of the trial, explained some of the factors that contributed to this low fidelity.Methods: In-depth interviews were conducted with patients who received vouchers, nurses in intervention clinics, personnel in shops who administered the vouchers, and managers of the TB Control Programme. These interviews were analyzed thematically.Results: The low fidelity to the trial intervention can be explained by two main factors. The first was nurses' tendency to 'ration' the vouchers, only giving them to the most needy of eligible patients and leaving out those eligible patients whom they felt were financially more comfortable. The second was logistical issues related to the administration of the voucher as vouchers were not always available for patients on their appointed clinic dates, necessitating further visits to the clinics which they were not always able to make.Conclusions: This process evaluation identifies some of the most important factors that contributed to the results of this pragmatic trial. It highlights the value of process evaluations as tools to explain the results of randomized trials and emphasizes the importance of implementers as 'street level bureaucrats' who may profoundly affect the way an intervention is administered.Trial registration: Current Controlled Trials ISRCTN50689131, registered 21 April 2009.The trial protocol is available at the following web address: http://www.hst.org.za/publications/study-protocol-economic-incentives-improving-clinical-outcomes-patients-tb-south-africa. © 2014 Lutge et al.; licensee BioMed Central Ltd.
adult; article; clinical article; controlled study; female; health economics; human; male; patient compliance; qualitative research; randomized controlled trial; social justice; South Africa; treatment outcome; tuberculosis; tuberculosis control; catering service; economics; financial management; health care cost; health personnel attitude; income; medication compliance; patient selection; poverty; sample size; time; treatment outcome; tuberculosis; tuberculostatic agent; Adult; Antitubercular Agents; Attitude of Health Personnel; Female; Financing, Personal; Food Supply; Health Care Costs; Health Expenditures; Humans; Income; Male; Medication Adherence; Patient Selection; Poverty; Qualitative Research; Sample Size; South Africa; Time Factors; Treatment Outcome; Tuberculosis