Factors Affecting Length of Stay for Observation Patients

Objectives The objective of this study was to determine whether the addition of a case manager and a physician advisor to the observation unit would decrease the length of stay (LOS) of observation patients. Study design This retrospective, observational study for observation patients was conducted in 2017. Methods At a tertiary-care, medium-sized, urban, community hospital, the LOS for all observation patients in 2017 (2, 981 clinical decision unit [CDU] patients and 1,248 non-cohort patients) was studied. Interventions studied were the addition of unit-based case manager and physician advisor to observation patient treatment teams. Results Patients assigned to the CDU had a shorter LOS than scattered patients, p < 0.0005. After the data was controlled for changes in LOS on inpatients using analysis of covariance (ANCOVA), none of the interventions resulted in statistically significant effects on LOS for CDU or scattered patients. Season, day of the week, the month of the year, and the presence of residents/medical students did not have any effect on LOS. Patients arriving at night had significantly shorter LOS than those arriving during the day or evening, p = 0.035 and p = 0.029, respectively. Conclusions Placing observation patients in a single unit is effective for decreasing LOS. The addition of case managers or physician advisors may not be an effective strategy to address the LOS. The presence of trainees does not hinder patient flow.