Improving disposition outcomes for patients in a geriatric skilled nursing facility

To evaluate an intervention to improve discharge disposition from a skilled nursing unit (SNU).

Historical control comparison of discharge disposition before and after implementation.

Fifty-bed SNU.

All patients admitted from acute care hospitals to a SNU between June 2008 and May 2010.

Physician admission procedures were standardized using a template, patients with three or more hospital admissions over the prior 6 months received palliative care consultations, and multidisciplinary root-cause analysis conferences for patients transferred back to the hospital acutely were conducted bimonthly to identify problems and improve processes of care.

Patients' discharge disposition (i.e., acute care, long-term care, home, or death) before and after implementation were compared.

Discharge dispositions were determined for all 1,725 patients admitted during the study; 862 patients before (June-May 2008) and 863 during (June 2009-May 2010) the intervention. Discharge dispositions were significantly differently distributed across the two periods (P=.03). Readmission to acute care declined (from 16.5% to 13.3%, a nearly 20% decline). Multivariable logistic regression, controlling for age, sex, and case-mix index and adjusting for clustering due to repeated admissions of individual patients, suggests that, during the intervention period, patients were more likely than during the baseline period to die on the unit in accordance with their wishes than to be transferred out to the hospital (odds ratio=2.45, 95% confidence interval=1.09-5.5).

Interventions such as the ones implemented can lead to fewer hospital transfers for SNUs.