Discharge against medical advice in pediatric wards at tertiary center in Addis Ababa, Ethiopia: A retrospective cross-sectional survey.

Background: Children are at high risk of problems related to discharge against medical advice (DAMA). Because they are not part of the decisions their best interests may be violated. This study is aimed to determine the prevalence, clinical outcomes, and factors associated with discharge against medical advice. Methods: We cross-sectionally described 123 admissions in which caretakers decided to discharge against medical advice. A mixed data collection method from the patient's charts and telephone interviews was employed. A pretested semi-structured questionnaire tool was used. The data was manually cleaned and analyzed using SPSS software version 25, USA. The frequency and percentage of categorical data were calculated, as well as the mean, median, SD, and IQR of the continuous variable. A multivariate regression analysis was performed, with an adjusted odds ratio of 95% CI and a statistical significance of 0.05. Results: Hospital prevalence of DAMA was 1.42%. The median age was 11(±59) months—eighty-one percent of the caretakers signed before leaving the hospital. The median hospitalization was 7 days (IQR=13 days). The majority of the caretakers were discharged during working hours (71%). Discharge was registered in all seasons: winter (31%), spring (28%), summer (23%), and autumn (18 %). Hemato-oncologic conditions, infectious, and neonatal problems were common. Most of the cases were reported from pediatric emergency and neonatology wards. Patients’ poor clinical response and the caretaker's financial constraints were the main reasons for signing DAMA. Death was significant when the DAMA occurred in fast-improving cases and infants; (AOR=6.909, 95% CI-2.191-21.782), and AOR=1.3, 95% CI -0.48-3.3) respectively. Conclusion: DAMA in the Pediatric wards of Tikur-Anbessa Specialized Hospital was similar to the reported data elsewhere. However, death was very high which could be due to insufficient fol-low-up after DAMA.