Community health centres in Tanzania have been established to improve access to healthcare services, yet their effectiveness varies widely across different locations and populations. A multilevel regression model was employed to analyse data collected from multiple levels (individual, family, community) within a hierarchical structure. Robust standard errors were used to account for potential sources of heteroscedasticity in the analysis. The multilevel regression analysis revealed that socio-economic status significantly influenced patient adherence rates at the individual level (β = -0.75, p < 0.05). The study demonstrates the utility of multilevel regression for understanding complex healthcare systems and identifying factors affecting service delivery. Policy recommendations include targeting interventions to improve patient adherence in low-income communities to enhance overall health outcomes. Treatment effect was estimated with $\text{logit}(p_i)=\beta_0+\beta^\top X_i$, and uncertainty reported using confidence-interval based inference.