Perioperative Primary Care Utilization and Postoperative Readmission, Emergency Department Use, and Mortality in Older Surgical Patients

BACKGROUND: Postdischarge primary care follow-up is associated with lower readmission rates after medical hospitalizations. However, the effect of primary care utilization on readmission has not been studied in surgical patients. METHODS: Retrospective cohort study of Medicare beneficiaries aged 65 and older undergoing major inpatient diagnostic or therapeutic procedures (n = 3,552,906) from 2017 through 2018, examining the association between postdischarge primary care visits within 14 days of discharge (primary exposure), and Annual Wellness Visits in the year prior (secondary exposure), with 30-day unplanned readmission (primary outcome), emergency department visits, and mortality (secondary outcomes). RESULTS: Overall, 9.5% (n = 336,837) had postdischarge visits within 14 days, 2.9% (n = 104,571) had Annual Wellness Visits in the year preceding the procedure, 9.5% (n = 336,401) were readmitted, 9% (n = 319,054) had emergency department visits, and 0.6% (n = 22,103) of the cohort died within 30 days. Our fully adjusted propensity-matched proportional hazards Cox regression analysis showed that postdischarge visits were associated with a 5% lower risk of readmission (hazard ratio HR, 0.95, 95% confidence interval CI, 0.93–0.97), 43% higher risk of emergency department use (HR, 1.43, 95% CI, 1.40–1.46) and no difference in mortality risk (HR, 0.98, 95% CI, 0.90–1.06), compared with not having a visit within 14 days of discharge. In a separate set of regression models, Annual Wellness Visits were associated with a 9% lower risk of readmission (HR, 0.91, 95% CI, 0.88–0.95), 45% higher risk of emergency department utilization (HR, 1.45, 95% CI, 1.40–1.49) and an 18% lower mortality risk (HR, 0.82, 95% CI, 0.75–0.89) compared with no Annual Wellness Visit in the year before the procedure. CONCLUSIONS: Both postdischarge visits and the Medicare Annual Wellness Visit appear to be extremely underutilized among the older surgical population. In those patients who do utilize primary care, compared with propensity-matched patients who do not, our study suggests primary care use is associated with modestly lower readmission rates. Prospective studies are needed to determine whether targeted primary care involvement can reduce readmission.