Thirty-Day Hospital Readmissions in a Care Transitions Program for High-Risk Older Adults

To describe the causes of 30-day hospital readmissions among high-risk older adults during implementation of a multicomponent care transitions program.

Secondary analysis of data from the evaluation of a multicomponent care transitions program for hospitalized high-risk older adults.

A 400-bed community teaching hospital.

Patients aged 75 and older admitted to non-intensive care unit beds who met specific criteria for high risk of complications and hospital readmissions. The intervention group included 202 patients, of whom 37 were readmitted to the hospital as an inpatient or on observation status within 30 days of discharge.

Root-cause analyses on each readmission were conducted by hospital physicians and post-acute care (PAC) organization staff. Additional data were collected by trained project staff using the medical record and postdischarge telephone or in-person follow-up visits. These data were reviewed and adjudicated among the authors, and each readmission was rated with unanimous agreement as "preventable," "possibly preventable," or "not preventable."

No significant differences were found in demographic and clinical characteristics of intervention patients readmitted versus those not readmitted. A higher proportion of the 37 patients who were readmitted did not have a postdischarge visit than the 165 patients who were not readmitted (15 [41%] vs 45 [27%]; P = .11). Among the 37 readmissions, 14 (38%) were rated as not preventable, 14 (38%) as possibly preventable, and 9 (24%) as preventable. Readmissions were rated as preventable or possibly preventable for a variety of reasons that provide insight into how care transitions programs for high-risk older adults might be made more effective.

Root-cause analyses of hospital readmissions among high-risk older adults by hospital physicians and PAC providers can identify strategies that might enhance the effectiveness of care transitions interventions in this complex population. J Am Geriatr Soc 68:1307-1312, 2020.