744 Correlates to improvement in sleep in acute mania

Abstract

Abstract Introduction Sleep disruption and reduced sleep duration are potential triggers and also core symptoms of a manic episode. Our understanding of how sleep duration changes during the course of a manic episode and its potential association with symptom improvement is limited. We examined the natural course of sleep duration and its association with time to discharge and other clinical and demographic factors in patients with mania. Methods This was a retrospective study conducted in patients admitted to an acute care psychiatric unit with a manic episode. Sleep duration was determined based on observer report as logged by nursing staff. Sleep duration at admission and discharge were determined by averaging the total sleep time on day 2/3 of hospitalization and day 3/2 preceding discharge date. We obtained data on possible confounders including antipsychotic (chlorpromazine equivalents), benzodiazepine (diazepam equivalents) and other hypnotic medication doses administered at admission and discharge. We examined the associations between the change in sleep duration from admission to discharge with length of hospitalization and other clinical and demographic characteristics. Results The sample consisted of 35 patients (54.3% male) aged 32 ± 9.96 years with an average length of hospitalization of 20.63 ± 18.62 days. The mean sleep duration on admission was 6.23 ± 1.77 hours and was 7.45 ± 1.49 hours on discharge, with a mean change of 1.23 ± 1.93 hours. The change in sleep duration was positively correlated with length of hospitalization (r=0.42; p=0.01). Other clinical factors including benzodiazepine or antipsychotic dose on admission, age, sex, and use of mood stabilizers were not correlated with the change in sleep duration. Conclusion There was a substantial improvement in the total sleep duration in patients with mania over the course of hospitalization. Overall, the change in sleep duration was only correlated to the length of stay and did not appear to be impacted by other clinical and demographic characteristics. Support (if any):