African Forensic Medicine | 15 September 2009
Methodological Assessment of Quasi-Experimental Designs in South African Community Health Centre Systems,
N, o, m, o, n, d, e, K, h, u, m, a, l, o, ,, S, i, p, h, o, M, s, i, m, a, n, g, ,, N, o, k, u, b, o, n, g, i, s, e, n, i, N, k, a, b, i, n, d, e
Abstract
Quasi-experimental designs are increasingly used in medical research to evaluate interventions without random assignment but with careful design control. A comprehensive search strategy was employed in electronic databases and grey literature. Studies were screened based on predefined inclusion criteria related to design features, context, and outcome measures. The review identified a trend towards using difference-in-differences (DID) models to assess yield improvement across various health centre settings. While DID is commonly used in the reviewed studies, variability exists in how it was applied and interpreted. Future research should consider methodological consistency and robustness when employing DID for yield improvement assessments in community health centres. Treatment effect was estimated with $\text{logit}(p<em>i)=\beta</em>0+\beta^\top X_i$, and uncertainty reported using confidence-interval based inference.