Mapping tuberculosis treatment intervention trials in Africa

ID: 

236

Session: 

Poster session 2

Date: 

Monday 17 September 2018 - 12:30 to 14:00

All authors in correct order:

Hohlfeld A1, Mathebula L1, Pienaar E1, Abrams A1, Lutje V2, Kredo T1, Ndwandwe D3
1 Cochrane South Africa, South African Medical Research Council, Tygerberg, South Africa
2 Cochrane Infectious Diseases Review Group, Liverpool School of Tropical Medicine, UK
3 Cochrane South Africa, South Africa
Presenting author and contact person

Presenting author:

Ameer Hohlfeld

Contact person:

Ameer Hohlfeld
Abstract text
Background: Mycobacterium tuberculosis (TB) poses a substantial burden in sub-Saharan Africa. Randomised controlled trials (RCTs) are the gold standard for evaluating the effectiveness of an intervention. Despite the many TB-related RCTs conducted in Africa, the problem remains. This study aimed to describe all published African TB trials to inform patients, researchers, policymakers and funders about the breadth of interventions that has been evaluated and the quality of their conduct to inform future research agendas.
Method: This is a cross-sectional study of published TB trials conducted in at least one African country. We searched PubMed, Embase and the Cochrane Library (April 2015) using the validated Africa search filter. We screened records for eligibility and extracted data using a prepiloted data extraction form. Extracted data included: country, publication dates, ethics statement, trial registration number, participants' age range and the methodological quality of the data using Cochrane 'Risk of bias' criteria. Data were analysed descriptively using MS Excel.
Results: We identified 7401 studies, 132 of which were eligible for inclusion. Trials were published from 1952 to 2015. Most trials were conducted in South Africa (n = 59), followed by Uganda (n = 22), and Tanzania (n = 15). Most interventions were drug therapies (n = 98), nutrition supplementation (n = 12), and directly observed treatment (n = 8). Fifty-one trials reported trial registry numbers. Forty did not provide an ethics statement. Sixty-four trials included adults only, while four included only children. Overall, we found that methodological quality was poorly reported: 53/132 trials reported adequate sequence generation, 28/132 trials reported adequate allocation concealment and 32/132 trials reported blinding of the provider. The main funders were pharmaceutical industries and international agencies.
Conclusion: Patients participating in trials may expect the results to inform policy and practice, however poorly reported research may not be adequate to meet this need. By mapping TB trials conducted in Africa, we identified gaps in research and methodological flaws in the reporting. Funders and researchers should ensure the conduct of higher quality trials addressing complex issues of how best to implement the available effecti

Relevance to patients and consumers: 

Patients and healthcare consumers participating in randomised controlled trials may expect favourable results to inform policy and practice. This is not always, especially when studies are poorly reported or consist of poor methodological quality. Consequently, their findings may be considered less valid.