Abstract
{ "background": "Community health centres are pivotal for primary care delivery in sub-Saharan Africa, yet evidence on the efficacy of different operational models for improving agricultural yield and nutritional outcomes remains fragmented. Previous syntheses have not systematically appraised the methodological rigour of field trials in this context.", "purpose and objectives": "This meta-analysis aimed to evaluate the methodological quality of randomised field trials assessing yield optimisation interventions and to quantify the pooled effect size of structured community health system models on measurable yield outcomes.", "methodology": "We systematically searched for published and grey literature reporting randomised controlled trials of community health centre-led interventions targeting yield improvement. Methodological quality was assessed using a modified Cochrane Risk of Bias tool. A random-effects meta-regression model, $\\hat{\\theta}i = \\mu + \\epsiloni + \\deltai$, where $\\deltai \\sim N(0, \\tau^2)$, was fitted to estimate the overall standardised mean difference (SMD). Heterogeneity was quantified using the $I^2$ statistic.", "findings": "The pooled analysis of 18 eligible trials (total \(n=12\),450 plots) showed a moderate, significant positive effect on yield (\(SMD = 0\).42, 95% CI: 0.28 to 0.56). However, high heterogeneity was observed ($I^2 = 78\\%$), which was partly explained by variability in trial design quality; only 33\\% of studies were judged as low risk for performance bias. Subgroup analysis indicated integrated agro-nutrition programmes had a larger effect size (\(SMD = 0\).61) compared to singular input distribution models.", "conclusion": "While community health centre-based interventions demonstrate a statistically significant positive impact on yield, the evidence base is constrained by considerable methodological heterogeneity and prevalent risks of bias, limiting definitive conclusions.", "recommendations": "Future trials should employ more rigorous blinding procedures and standardised outcome measurement for agricultural yields. Programme design should prioritise integrated service delivery