"background": "Community health centres are a cornerstone of primary healthcare delivery in South Africa, yet systematic evaluations of their adoption and impact are methodologically heterogeneous. This complicates the synthesis of evidence for policy. ", "purpose and objectives": "This meta-analysis aims to methodologically evaluate the application of difference-in-differences (DiD) models in measuring adoption rates of community health centre systems, assessing their design robustness and inferential validity. ", "methodology": "A systematic search identified studies employing DiD to evaluate health centre adoption. Studies were appraised for methodological rigour, focusing on model specification, parallel trends assumption testing, and handling of serial correlation. The core DiD estimator was formalised as Y{it = \0 + \1 + \2 + \3 (\) +, with inference based on cluster-robust standard errors. ", "findings": "Only 35% of included studies formally tested the parallel trends assumption. Among those that did, a significant proportion (approximately 40%) found visual or statistical evidence suggesting its violation, potentially biasing adoption rate estimates. The methodological quality score had a median of 6 out of 10 (IQR: 4–7). ", "conclusion": "The application of DiD in this context is often methodologically inconsistent, with frequent lapses in key diagnostic checks, threatening the credibility of reported adoption effects. ", "recommendations": "Future research must rigorously test and report on the parallel trends assumption, employ more robust estimation techniques like event-study designs, and improve transparency in pre-intervention data reporting. ", "key words": "difference-in-differences, parallel trends, health systems evaluation, primary healthcare, methodological review, adoption metrics", "contribution statement": "This study provides the first methodological synthesis and quality assessment of DiD applications in South African
Thandiwe Nkosi (Wed,) studied this question.