Prediktor Lama Rawat Inap Pasien: Peran Gender, Jenis Penyakit, Pendidikan, Usia, Motivasi Sembuh, dan Status Fungsional

Length of stay (LOS) is a key indicator of hospital service efficiency. This study aims to identify predictors of length of stay, including demographic (gender, age, education), psychosocial (motivation to recover), and clinical (type of disease, functional status) factors in adult patients. A quantitative cross-sectional study was conducted on 171 inpatients in the internal medicine ward at a central referral hospital in Manado, selected using consecutive sampling. Data were collected via medical records, a structured recovery motivation questionnaire, and the Katz Index for functional status. Multivariate analysis was performed using binary logistic regression with the Enter method. The results of the analysis showed that the logistic regression model had excellent fit (p=.000; Hosmer-Lemeshow test p=.986; Nagelkerke R² = 31.6%; prediction accuracy 80.1%). Of all the variables, only two were found to be significant independent predictors of length of hospital stay: functional status (OR=1.301; 95% CI: 1.068–1.586; p=.009) and stroke type compared to diabetes mellitus (OR = 3.923; 95% CI: 1.130–13.623; p = 0.031). The variables of gender, education, age, and motivation to recover were not independently significant ( ). In conclusion, functional status and stroke diagnosis are significant predictors of changes in length of hospital stay. Early functional screening and discharge planning based on clinical risk are highly recommended.