Applications and implementation considerations for stepped-wedge designs in sub-Saharan Africa: a systematic review.

Abstract

Introduction: Stepped-wedge design (SWD) has gained prominence as a versatile research methodology, particularly in public health and implementation science, due to its ability to balance ethical concerns with methodological rigor. This systematic review aims to evaluate the implementation and effectiveness of SWD in sub-Saharan African (SSA) research contexts, focusing on the types of interventions, primary outcomes, and the unique geographic and cultural factors influencing the studies.

Methodology: A systematic review protocol was developed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) under identification number CRD42024530774. A comprehensive search strategy was employed to identify studies conducted in SSA using SWD from January 2000 to March 2024 across five electronic databases (PubMed, Web of Science, CINAHL, PsycINFO, and Cochrane Library), along with Google Scholar and citation tracking. Studies were included if they utilized SWD in SSA settings and reported relevant public health, clinical, or social interventions. Data were extracted on study characteristics, SWD implementation details, statistical methods, and sample size calculations. A total of 85 studies were included after screening 873 titles and abstracts and conducting full-text reviews of 93 articles.

Results: The 85 studies included in the review spanned a wide range of health domains, including HIV/AIDS, maternal and child health, tuberculosis, and malaria, conducted across diverse SSA settings such as hospitals, communities, and schools. The studies involved a total of 1,895,788 participants, with sample sizes ranging from 17 to 780,000. Most studies (84.7%) were facility-based, while 15.3% were community-based. The number of clusters per study varied, with some studies using as few as four clusters, while others utilized up to 54 clusters. The number of steps ranged from two to twelve, depending on the complexity and scale of the intervention. Sample size calculations were often based on expected changes in primary outcomes, with many studies assuming an intra-cluster correlation coefficient to account for clustering effects. The SWD was primarily chosen to address ethical concerns, logistical challenges, and resource limitations. The review highlights significant variability in study designs, interventions, and outcomes, reflecting the adaptability of SWD to different contexts and challenges.

Conclusion: The SWD has been effectively utilized in SSA research to evaluate a wide range of interventions across diverse settings, demonstrating its flexibility and suitability for addressing complex public health challenges. However, the review also identifies challenges related to study duration, logistical implementation, randomization processes, and statistical analysis, suggesting the need for careful planning and methodological rigor in future studies using SWD. The findings provide valuable insights for researchers and policymakers seeking to optimize the use of SWD in resource-limited settings, ensuring that interventions are both effective and ethically implemented.

Systematic reviews registration: https://www.crd.york.ac.uk/PROSPERO/view/CRD42024530774, PROSPERO CRD42024530774.