Results Socio-demographic characteristics of clients In total, 927 clients were interviewed with a response rate of 78.3%. More than one third of those interviewed were aged 19–30 (35.8%), while those aged 31–40 made up 14.7% and 9.5% were 51 years or older. Majority (63.5%) were married or cohabiting, 39.0% of the clients had primary education as the highest level of educational attainment and 19.6% had no formal education at all. Over half (50.4%) of the respondents were peasant farmers. Most (65.0%) of the respondents were Protestants (Table 2). Table 2 Socio-Demographic Characteristics of respondents 1Lowest overall satisfaction rate 2Lowest rating on payment dimension 3Lowest rating on rights dimension ** Factors shown in univariate model had LRT p-value < 0.10 * Wide confidence interval due to small sample size Type of services sought by clients Most (55.2%) of the clients interviewed sought services from the outpatients department on the interview days. In-patients were also interviewed; 21.5% in medical, 9% in maternity, 6.3% in pediatric, and 3.5% in surgical department. There were also a substantial number of clients that had sought laboratory services (16.4%), family planning and ANC services (12.3%), and HIV/STI services (10.1%) as shown in Figure 1. Figure 1 Type of services sought by clients on interview day. Distance between Client's home and the Health Facility visited The majority of clients' (64.0%), lived between 0–4 kms from their homes to the facility they visited on the day of interview, 15.7% between 5–9kms, 5.2% between 10–14 kms, 2.5% had travelled between 15–19 kms to reach the health facility. Notably, 12.6% had travelled 20 kms or more to reach the health facilities they visited. For 46% of clients (N=426), there was a nearer health facility to their home than the one they visited at the day they were interviewed. Reasons given for not visiting the nearest facility were; overall poor quality of services, (18.7%), unavailability of drugs (18.4%), unavailability of staff (10.2%), cost (6.7%). Satisfaction with the overall level care seeking experiences was 84.2% (Table 3). For most of the client satisfaction dimensions, there was a satisfaction rate of over 70.0%. This pattern was consistent for almost all dimensions. However, confidentiality was rated highest at 90.8%. The lowest client satisfaction ratings were for the dimensions relating to rights and payments. The amount charged and the explanation of charges provided were particularly identified as issues of dissatisfaction. In addition, clients felt that they were not fully made aware of their rights and entitlements and were not satisfied that they were facilitated with information to make choices and decisions that suited them. Table 3 Client satisfaction with the different care dimensions assessed. Overall satisfaction increased with age up to the age 40 and reduced beyond this age. The single/never married had lower satisfaction levels compared to the married, widowed and divorced. Satisfaction levels were generally lower in Health Centre IV and hospitals compared to the lower level centres. In regard to satisfaction of payment for services, females were more satisfied than males, satisfaction levels increased with education but dropped after secondary education to a level below those with no education. Those in formal employment and the unemployed had the lowest levels of satisfaction for payment (<50%) while satisfaction levels were fairly the same across all other occupational groups. Protestants and Catholics had lower satisfaction with payments compared to the Muslims and other religion. In terms of rights, the divorced had lower levels of satisfaction compared to the widowed, married and never married/single. Muslims had a lower rate of satisfaction with the rights dimension compared to Protestants, Catholics and other religions (Table 4). Table 4 Association between client satisfaction and demographic characteristics Across all the three dimensions, diocese was associated with satisfaction although this varied (Table 5). Table 5 Overall client satisfaction, satisfaction with payments and rights by Diocese A diocese is an administrative unit under the stewardship of a bishop within the church structure and consists of parishes. The units relating to the diocese do not necessarily correspond to the government district administrative structures and in Uganda, most of the dioceses combine two or more of the government administrative districts. The Church of Uganda has 35 dioceses 13 and as the technical health arm of the Church of Uganda, SDA and other Protestant churches, UPMB works within the diocese structures in terms of defining stewardship of the member health facilities majority of whom are founded by the Church of Uganda. Therefore in this study, the dioceses' correspond to a group of protestant church-founded health facilities within the administrative units and are members of UPMB. For the other protestant churches that do not have dioceses, the study grouped these by the church affiliation. The following dioceses had low overall satisfaction rates; South Rwenzori, Nebbi, Karamoja, Mukono and Lango. In regard to satisfaction with the payments dimension; Kumi, Lango, South Rwenzori, Mbale, Mukono, Kigezi, Karamoja and Luwero diocese had lower satisfaction rates. Concerning the rights dimension; Busoga, lango, Kigezi, kitgum, South Rwenzori, Namirembe, Luwero and Mukono dioceses had lower satisfaction. In general, the following dioceses had high dissatisfaction across all the satisfaction dimensions (overall satisfaction, payment, rights); South Rwenzori, Mukono and Lango. Table 6 shows the factors associated with client dissatisfaction with health services in UPMB facilities. Overall satisfaction In the multivariate analysis the patients' dioceses were significantly associated with overall dissatisfaction. The dioceses of Nebbi (OR=16.4), South Rwenzori (OR=5.65), Karamoja (OR=3.74), Lango (OR=3.38), Mukono (OR=2.93) had higher levels of dissatisfaction compared to North Kigezi. There was some evidence of association with marital status, single/never married were 3.05 times more likely to be dissatisfied compared to widowed. Clients attending HCIII were less likely to be dissatisfied compared to HCII (OR=0.51). The association with age observed in the univariate model disappeared in the final model. Dissatisfaction with payments dimension The dioceses association with dissatisfaction was also observed in payments but with a different set of dioceses added to Nebbi (OR=4.49), South Rwenzori (OR=4.82), Karamoja (OR=11.28), Lango (OR=5.59) and Mukono (OR=3.13); Kumi (OR=3.32), Soroti (OR=2.40), Kigezi (OR=2.92), South Rwenzori (OR=4.48), Mbale (OR=2.62), Luwero (OR=2.23) were the additional dioceses where clients showed higher odds of dissatisfaction compared to those in North Kigezi. Post-secondary education (OR=1.79), being formally employed (OR=2.78) or unemployed (OR=3.34), attendance at a hospital (OR=2.15) were also associated with high dissatisfaction levels with payments. Religion and gender were only associated with dissatisfaction at the univariate analysis. Dissatisfaction with the rights dimension In addition to South Rwenzori (OR=4.56), Mukono (OR=2.40) and Lango (OR=9.43), the patients in the following dioceses also showed higher odds of dissatisfaction with the rights dimension; Busoga (OR=3.00), Kigezi (OR=2.82), Kitgum (OR=12.40), Northern Uganda (OR= 3.00), Namirembe (OR=3.18), and Luwero (OR=5.57) compared to North Kigezi. Clients from Rwenzori were less likely to be dissatisfied compared to North Kigezi. There was some evidence of association with marital status, as the divorced/separated had higher odds of dissatisfaction (OR=2.54) in respect to rights compared to the widowed.