Modifiable service-delivery factors, not geography, drive patient satisfaction in rural Sierra Leone: a district-comparative cross-sectional household survey of 679 facility users

Background Patient-reported satisfaction is a core tracer of health-system responsiveness in Universal Health Coverage (UHC) monitoring, yet its determinants in rural Sierra Leone are poorly characterised. We quantified overall and domain-specific satisfaction and identified modifiable predictors across three rural border districts. Methods We conducted a cross-sectional, population-based household survey in October 2024 in Kailahun, Kambia and Pujehun districts, using a two-stage cluster design (chiefdoms sampled with probability proportional to size; households sampled at random). A validated 29-item instrument measured overall satisfaction and eight patient-experience domains on five-point Likert scales. The primary outcome was the five-level single-item overall satisfaction rating. We fitted a multivariable proportional-odds ordinal logistic regression, with 95% confidence intervals (CIs) obtained by a cluster bootstrap resampling the 20 chiefdom clusters. Robustness was assessed with binary and composite-outcome sensitivity models. Results Of 750 respondents, 679 (90.5%) had used a formal health facility in the previous 12 months and formed the analytic sample (510 [75.1%] female; mean age 28.2 years [SD 13.8]). The instrument showed high internal consistency (Cronbach = 0.87 for the five core domains). Overall, 375/679 (55.2%) were satisfied or very satisfied, ranging from 185/244 (75.8%) in Kambia to 110/224 (49.1%) in Kailahun and 80/211 (37.9%) in Pujehun ({chi}{superscript 2} = 70.8; p<0.001). In the adjusted model, staff attitude was the strongest predictor of higher satisfaction (adjusted odds ratio [AOR] 2.74, 95% CI 2.03-3.70; p<0.001 per one-point increase), followed by waiting-time satisfaction (AOR 1.89, 1.39-2.46) and medicine availability (AOR 1.43, 1.16-2.10). Travel-time category, facility type and sex were not independently associated. Large district disparities persisted after adjustment: relative to Kambia, the AOR for higher satisfaction was 0.27 (0.15-0.50) in Kailahun and 0.33 (0.20-0.66) in Pujehun. Adjusted probabilities of high satisfaction were 0.72, 0.48 and 0.43, respectively. Conclusions Respectful provider behaviour, shorter waits and reliable medicine supply, all amenable to district-level management, were the dominant and actionable drivers of patient satisfaction, whereas geographic distance was not. Persistent between-district gaps call for tailored quality-improvement in Kailahun and Pujehun. Institutionalising routine patient-experience measurement would strengthen accountability for people-centred care and support equitable progress towards UHC.