{ "background": "Despite significant investment in rural water infrastructure, sustained adoption and technical performance of point-of-use treatment systems in sub-Saharan Africa remain poorly understood, with a lack of rigorous field-based engineering evaluations.", "purpose and objectives": "This study aimed to diagnostically evaluate the real-world adoption rates and operational performance of installed ceramic filter and chlorination systems in rural communities, identifying key engineering and socio-technical determinants of success.", "methodology": "A quasi-experimental design was employed, comparing intervention villages with matched control villages. Data were collected via household surveys, technical performance tests, and direct observation. Adoption was modelled using a logistic regression framework: $\\logit(pi) = \\beta0 + \\beta1 X{1i} + \\beta2 X{2i} + \\epsiloni$, where $pi$ is the probability of sustained use. Robust standard errors were clustered at the village level.", "findings": "Sustained adoption of provided technologies was 34% (95% CI: 28, 40). A key determinant was the regularity of maintenance visits, which increased the odds of adoption by a factor of 3.2. Water quality compliance was significantly higher in intervention households, though system failures due to filter clogging and chlorine stock-outs were prevalent.", "conclusion": "The diagnostic approach reveals a substantial gap between installation and sustained use, driven primarily by post-installation support logistics and user technical literacy, rather than initial community acceptance.", "recommendations": "Engineering implementation protocols must integrate guaranteed maintenance cycles and simpler user interfaces. Policy should shift focus from installation targets to long-term performance metrics, with funding linked to verified adoption data.", "key words": "water treatment adoption, quasi-experimental design, rural infrastructure, socio-technical systems, logistic regression, Ethiopia", "contribution statement": "This study provides a novel diagnostic framework combining engineering performance data with behavioural adoption metrics, generating a validated predictive model for infrastructure sustainability in low-resource settings."