Determinants of hospital length of stay for people with serious mental illness in England and implications for payment systems: a regression analysis
BACKGROUND Admissions lasting over six months (long-stay) occur frequently among patients with schizophrenia in South Korea. AIMS To identify some patient-level and institution-level factors associated with long-stay status of patients with schizophrenia. METHODS This is a retrospective cross-sectional study. We analysed a nationwide population-based reimbursement claims data set consisting of 496,338 claims for 58,287 patients with schizophrenia between 1 January 2005 and 30 June 2006. A two-level random effects logistic regression model was used to identify those factors. RESULTS Age (<20 years (ref), 60-69 (OR 2.000, 95% CI: 1.640-2.438), ≥ 70 (2.068, 1.682-2.543)), male gender (1.192, 1.144-1.242), type of national health insurance plan (national health insurance (ref), Medical Care Aid Type 1 (4.299, 4.024-4.593)), secondary diagnosis (none (ref), psychiatric diagnosis (0.719, 0.666-0.777), non-psychiatric diagnosis (0.918, 0.850-0.991)) and type of institution (clinic (ref), psychiatric hospital (2.769, 1.507-5.087)) were associated with likelihood of long-stay status. Institutional variable associated with long-stay status included a higher number of beds (1.073, 1.013-1.137). The number of professionals (0.752, 0.646-0.876) showed negative association with long-stay status. CONCLUSIONS Researchers could improve their assessment of long-stay status of patients with schizophrenia by using a two-level analysis including patient-level and institution-level factors. This study suggests that mental health interventions to reduce the long stay of patients with schizophrenia focus on older male patients, those enrolled in a national medical care aid programme and those admitted to psychiatric hospitals.