Publication bias and time to the publication of randomised trials conducted in sub-Saharan Africa: a cross-sectional study.

Background: Ideally, evidence-based decisions about healthcare interventions should be informed by access to up-to-date information from all relevant RCTs, making it essential that the reports are published soon after study completion. However, studies have consistently shown that between 25 and 50% of clinical trials remain unpublished or are only published many years after completion. The WHO has noted a slow but steady increase in the number of RCTs since the mid-2000s, particularly in sub-Saharan Africa (SSA). However, the extent of publication bias of SSA RCTs remains unknown. Therefore, our study objectives were to assess (1) the proportion of completed RCTs from SSA that have been published and (2) the time from completion to publication.

Methods: This cross-sectional study, consisting of a retrospective analysis of registered SSA RCTs, aims to report the proportion of completed and terminated SSA RCTs registered in ClinicalTrials.gov and the Pan African Clinical Trials Registry (PACTR) and their time to publication.

Results: Our search yielded 7896 records, of which 3026 RCTs met our inclusion criteria for analysis. We identified journal publications for 1983 (65.5%) RCTs. The overall median time to publication from the primary completion date was 34.2 months (95% CI: 32.4 to 35.5).

Conclusions: Overall, we found a substantial proportion (34.5%) of unpublished SSA RCTs. Moreover, the median time to publication from primary completion was 34.2 months. The persistence of publication bias threatens the integrity of evidence-based healthcare practice, particularly given that consumers depend on peer-reviewed journal publications as conventional and trusted sources to stay informed. Our findings underscore the importance of continued research to test and implement preventative strategies to mitigate publication bias.