Gaps in Primary Care Follow-Up After Hospital Discharge Among Medicare Beneficiaries.

Background: Timely primary care follow-up after hospitalization is recommended to monitor recovery and coordinate care. Whether follow-up differs for vulnerable populations, such as those with frailty and those discharged to skilled nursing facilities (SNF) prior to returning home, is not known.

Methods: Retrospective cohort study using a 100% sample of traditional Medicare beneficiaries discharged from hospital to home or from hospital to SNF and then home, between 2010 and 2022. The primary outcome was the receipt of a primary care visit within 30 days of return to home, measured overall and stratified by disposition (discharged home vs. to SNF then home) and by frailty (defined by a claims-based frailty index). Multivariable logistic regression models were used to estimate changes in outcomes over time, overall and stratified by disposition and frailty.

Results: The cohort included 94,248,326 discharges (80.1% age ≥ 65 years, 55.1% female, 36.7% frail) of which 21.5% were discharged to SNF and then home. Between 2010 and 2022, primary care follow-up increased from 51.5% to 57.5% for patients discharged directly home and from 24.3% to 28.4% for patients discharged to SNF then home. In adjusted analyses, compared to those discharged directly home, patients discharged to SNF and then home had an 8.2% point (pp) (95% CI, -8.5 to -7.9) lower predicted probability of ambulatory follow-up in 2022. Among patients discharged directly home, no difference was evident in follow-up between frail and non-frail patients (54.6% vs. 54.1%); difference 0.4 pp (95% CI, -0.1 to 1.0). In contrast, among patients discharged to SNF then home, frail patients had a lower predicted probability of follow-up (42.8% vs. 48.9%); difference - 6.1 pp (95% CI, -7.0 to -5.2).

Conclusions: Frail patients and patients requiring a short-term SNF stay after hospitalization are less likely to receive timely follow-up upon return to home than other patient groups.