Objective. To identify variables that predict length of stay (LOS) in an acute psychogeriatric ward; quantify psychiatric, medical and social factors that contribute to LOS; and identify factors that prolong LOS.
Design. Prospective audit of admissions over 12 months. Setting. Acute psychogeriatric ward in teaching hospital. Patients. 73 consecutive patients with a total of 88 admissions. Main outcome measures. Main psychiatric, medical and social reasons for admission; Physical Health Schema; total LOS; number of days attributable to psychiatric, medical, social and treatment (non-response and adverse reactions) reasons; DSM-IV diagnosis; global clinical outcome; discharge living arrangements; discharge psychotropic medication.
Main results. Presence of depressive symptoms was the most common psychiatric reason for admission and carer stress the main social reason for admission. Principal diagnosis of major depression and carer stress were signi®cant predictors of LOS and accounted for 24.3% of the variance ( p < 0.0001). Physical health and psychiatric comorbidity were not able to predict LOS. The proportion of psychiatric, medical and social reasons involved in LOS varied signi®cantly by principal diagnosis ( p < 0.001). Almost 42% of bed days were attributable to delaysÐ for medical, social (eg placement problems), treatment (adverse events, non-response) or hospital system reasons.
Conclusions. The quanti®cation of factors contributing to LOS in an acute psychogeriatric unit allows for more accurate identi®cation of issues that can be addressed to improve the eciency of bed utilization.