RESUMO
BACKGROUND: Psychiatric morbidity burden accounts 12.45% of the disease admission burden in Ethiopia; only two referral hospitals are found to manage all cases. The aim of this study is to assess the predictors of patient outcomes. METHOD: A 3 years retrospective patients' cards, charts and medical notes review in psychiatry case admission department of Jimma university teaching and training specialised hospital was conducted. All the admitted cases included in this study. Bivariate and multivariable logistic regression analyses were conducted to identify independent predictors of outcomes. RESULT: Among 402 study participants, the majority of them 301 (74.9%), were improved from their mental illnesses. First to eight grades completed study participants were found to be 1.34 times more likely improved mental illness than not able to read or write [AOR = 1.34, 95% CI (1.18-2.78), P < 0.009)]. The probability of improving from mental illness on married study participants was found 2.81 times more likely than single study participants [AOR = 2.81, CI (1.90-4.50), P < 0.043]. First time admitted cases improved 2.82 times more likely than those having a previous admission history [AOR = 2.82, CI (2.05-3.17), P < 0.05]. Duration of stay from 31 to 44 days showed more likely than from 1 to 20 days on patient improvement, [AOR = 1.88, CI (1.42-2.65), P < 0.034]. However, the hospital stay above 44 days does not show any statistical association with patient's medical improvement. CONCLUSION: Married, better educated, and the hospital stay of one to one-and-half month predicts better health outcome. Thus, this study suggests, psychiatric case management needs the collaborative care of the family in concurrence with counselling and guidance with enough time to better-off patients' outcomes. Our findings are useful in designing and improving-patient services for psychiatric patient programs and focused health communication and counselling strategies in relation to psychoactive substances in Ethiopia.