Socio-demographic and institutional predictors of inpatients’ post-discharge stay in referral hospitals in Kisumu County, Kenya

Abstract

Globally, inpatients continue to unnecessarily prolong their stay in referral hospital wards upon their medical discharge. This causes congestion in the wards, hospital reinfection, relapse, death of PDS inpatients and financial burden to the hospital management. Existing literature linked post discharge stay (PDS) to economic reasons. However, even with the introduction of Universal Health Coverage (UHC), waivers of medical bills, and free maternal health care in Kenya, reports still show PDS cases in the country, especially in referral hospitals such as Jaramogi Oginga Odinga Teaching and Referral Hospital (JOOTRH) and Kisumu County Referral Hospital (KCRH) in Kisumu county. However, it was unclear whether inpatients’ socio-demographic characteristics and institutional factors influenced PDS. This study aimed to investigate the socio-demographic and institutional predictors of inpatients’ PDS in referral hospitals in Kisumu County, Kenya. The specific objectives were to determine the influence of inpatient’s demographic characteristics on PDS, to establish the influence of social support on PDS, and to assess the influence of institutional factors on PDS. The study was guided by the social-ecological model proposed by McLeroy, et al., (1988). The study adopted a correlational cross-sectional research design and used mixed methods of data collection. Hospital records estimated that 200 inpatients experienced PDS in the two facilities per month, out of which a sample of 133 was calculated using Yamane’s (1967) formula. A stratified sampling technique was used to select inpatients in the 14 wards, after which systematic random sampling was used to reach the individual PDS inpatients for interviews. Key informant interview was used to collect qualitative data from 10 key informants who were purposively selected from the staff returns report (2019) while an in-depth interview was used to collect data from 13 PDS inpatients. To establish the predictors of PDS, a binary logistic regression analysis was used for the three objectives where p-values <0.05 was considered statistically significant Odds ratios and 95% confidence intervals were reported to show the magnitude and influence of PDS. Thematic analysis was used to analyze qualitative data and quantitative results were corroborated with verbatim quotations. The findings established that demographic characteristics of PDS inpatients namely age (P-value 0.01), gender (P-value 0.03), marital status (P-value 0.02), and nature of the illness (P-value <0.0001) were key demographic predictors of PDS. The parental status of children, religion, educational level, and employment status of respondents was not statistically significant to the study. The estimated logistic regression results indicated that social factors including living arrangement (P-value <0.009), who brought the patients to the hospital (P-value 0.034), visitation during hospitalization (P-value 0.029), social support received from relatives and friends (P-value 0.001) were statistically significant in the study implying that the respondents whose social support was strong were less likely to experience PDS. Institutional delays like waiting for discharge clearance (P-value 0.028), timely information (P-value 0.003), UHC status (P-value 0.01), awaiting tracing (P-value <0.001), and awaiting repatriation (P-value 0.001) were significant predictors of PDS while NHIF status and request for prolonged stay were not significant predictors of PDS. The study recommends hospital management mitigate delaying processes such as improper social assessment and long discharge processes that escalate the PDS of inpatients in the hospitals by early identification of PDS predictors. Policymakers should also incorporate strategies for reducing PDS cases in the existing health policies and strategic plans considering key socio-demographic and institutional predictors of PDS.