Purpose <p>This study examines the influence of key service quality dimensions physical environment, “care responsiveness, and patient-centered care on patient loyalty, and investigates the mediating role of patient satisfaction in public hospitals in Sidama Regional State, Ethiopia. The region was selected as a representative setting of public healthcare delivery in resource-constrained environments, where service quality challenges are particularly evident.</p> Design/methodology/approach <p>A facility-based cross-sectional design was employed using data collected from 382 patients attending public hospitals in Sidama Regional State. The sample size was justified using an a priori statistical power analysis suitable for structural equation modeling, ensuring adequate power to detect medium effect sizes. Respondents were selected through convenience sampling. Data were collected using a structured questionnaire adapted from validated scales. The study applied Partial Least Squares Structural Equation Modeling (PLS-SEM) using SmartPLS to test both direct and mediating relationships. Reliability and validity were assessed through Cronbach’s alpha, composite reliability, average variance extracted (AVE), and discriminant validity criteria.</p> Findings <p>The results show that physical environment (<i>β</i> = 0.211, p &lt; 0.001) and care responsiveness (<i>β</i> = 0.479, p &lt; 0.001) have significant positive effects on patient loyalty. Similarly, physical environment (<i>β</i> = 0.457, p &lt; 0.001) and care responsiveness (<i>β</i> = 0.288, p &lt; 0.001) significantly influence patient satisfaction. Patient-centered care has a significant direct effect on loyalty (<i>β</i> = 0.225, p &lt; 0.001) but an insignificant effect on satisfaction (<i>β</i> = 0.077, p &gt; 0.05). Patient satisfaction significantly predicts patient loyalty (<i>β</i> = 0.112, p &lt; 0.05) and partially mediates the relationship between service quality and loyalty. The model explains a substantial proportion of variance in patient loyalty (R² = 0.802) and satisfaction (R² = 0.548), with care responsiveness emerging as the strongest predictor.</p> Originality/value <p>This study contributes to the literature by addressing both theoretical and methodological gaps. Theoretically, it integrates Expectation–Confirmation Theory and Social Exchange Theory to explain the cognitive and relational mechanisms linking service quality to loyalty. Methodologically, it applies a PLS-SEM mediation model in a public healthcare context, which remains underexplored in Sub-Saharan Africa. Empirically, it provides context-specific evidence from Ethiopian public hospitals, offering practical insights for improving service quality and strengthening patient-centered healthcare delivery in resource-constrained settings.</p>