OBJECTIVES
To test the effectiveness of a multicomponent care transition intervention targeted at hospitalized patients, aged 75 years and older, at high risk for hospital readmissions, return emergency department (ED) visits, and related complications.
DESIGN
Implementation as a quality improvement program with propensity‐matched preintervention and concurrent comparison groups over a 12‐month period.
SETTING
A 400‐bed community teaching hospital.
PARTICIPANTS
Patients, aged 75 years and older, admitted to non–intensive care unit beds who met specific high‐risk criteria. The intervention group included 202 patients, and the concurrent and preintervention comparison groups included 4142 and 4592 patients, respectively.
MEASUREMENTS
Primary outcomes were 30‐day hospital readmissions and returns to the ED; 7‐day readmissions and ED visits were secondary measures.
RESULTS
Among the 202 patients enrolled in the “Safe Transitions for At‐Risk Patients” (“STAR”) program, 37 (18.3%) were readmitted within 30 days, in contrast to 14.3% and 14.6% in the concurrent and preintervention comparison groups, respectively. Rates for 30‐day return ED visits that did not result in hospitalization were 10.9% in the intervention group, and 7.2% and 7.9% in the comparison groups. STAR patients had greater 30‐day ED use than patients in the preintervention comparison group (5.0 percentage points; 95% confidence interval = 0.8‐9.3 percentage points; P = .020). Implementation challenges included suboptimal involvement of the participating hospital and post–acute care organizations and a relatively high proportion of patients who did not receive the intervention as planned, despite agreeing to participate before leaving the hospital.
CONCLUSION
A multicomponent care transitions intervention targeting high‐risk patients, aged 75 years and older, admitted to a community teaching hospital was not effective in reducing 30‐ or 7‐day readmissions or return ED visits. Our implementation experience offers many lessons for future programs for similar high‐risk geriatric populations. J Am Geriatr Soc 67:2634–2642, 2019