BACKGROUND: Information on outcomes of acute inpatient care in routine psychiatric practice is scant. In particular, it is uncertain to what extent short hospitalization can produce clinically meaningful changes. AIM: Our aim was to estimate the symptomatic outcome in a representative sample of patients admitted for short treatment to general hospital psychiatric units in Italy. METHODS: Patients were assessed at admission and discharge using 24-item Brief Psychiatric Rating Scale (BPRS). Reliable change index was calculated to estimate the proportion of change attributable to measurement error and a cut-off score of 38 was adopted to identify the patients who showed clinically significant change. RESULTS: Average length of stay was 5.7 days. Mean BPRS score dropped from 53.2 on admission to 41.5 at discharge, showing statistically significant improvement with an effect size of 0.80. However, reliable change was achieved by 24.7% of patients and clinically meaningful change by 13.6%. CONCLUSIONS: Reliance on statistical significance and effect size overestimates treatment effects, whereas reliable and clinically significant change index provides a conservative way to assess outcome. Few patients showed relevant improvement after a brief admission.
BACKGROUND: Information on outcomes of acute inpatient care in routine psychiatric practice is scant. In particular, it is uncertain to what extent short hospitalization can produce clinically meaningful changes. AIM: Our aim was to estimate the symptomatic outcome in a representative sample of patients admitted for short treatment to general hospital psychiatric units in Italy. METHODS:Patients were assessed at admission and discharge using 24-item Brief Psychiatric Rating Scale (BPRS). Reliable change index was calculated to estimate the proportion of change attributable to measurement error and a cut-off score of 38 was adopted to identify the patients who showed clinically significant change. RESULTS: Average length of stay was 5.7 days. Mean BPRS score dropped from 53.2 on admission to 41.5 at discharge, showing statistically significant improvement with an effect size of 0.80. However, reliable change was achieved by 24.7% of patients and clinically meaningful change by 13.6%. CONCLUSIONS: Reliance on statistical significance and effect size overestimates treatment effects, whereas reliable and clinically significant change index provides a conservative way to assess outcome. Few patients showed relevant improvement after a brief admission.