Correction to “Changes in Skilled Nursing and Home Health Admissions Associated With Medicare Payment Reforms and the COVID-19 Pandemic”

R. A. Prusynski, N. E. Leland, A. Humbert, A. Dahal, C. Brown, H. Amaravadi, D. Saliba, T. M. Mroz, “Changes in Skilled Nursing and Home Health Admissions Associated With Medicare Payment Reforms and the COVID-19 Pandemic,” Journal of the American Geriatrics Society 73, no. 2 (2025): 592–601, https://doi.org/10.1111/jgs.19322.

The authors have noticed that some inpatient rehabilitation stays were not removed from the hospital cohort. This requires a change to all descriptive statistics and analytic results.

Abstract

Methods: Using 100% of Medicare fee-for-service data, we conducted adjusted interrupted time series analyses of 30,507,466 hospital stays.

Results: At baseline, an average 18.3% of hospitalized beneficiaries were admitted to SNFs and 14.2% to HHAs. Post-COVID, admissions to SNF declined to 15.8% of patients being discharged from hospitals and HHA admissions increased to 18.6%.

RESULTS

Our sample included 30,507,466 Medicare fee-for-service acute hospital stays between January 2018 and October 2021. Of the 785,626 average monthly hospital discharges during the baseline period, 18.3% were admitted to SNF within 3 days and 14.2% were admitted to HHA within 14 days (Figure 1). During the COVID-19 period, SNF admissions decreased to an average of 15.8% and HHA admissions increased to 18.6%. PDPM implementation in October 2019 was associated with an immediate 0.29 pp decline in SNF admissions (95% CI −0.43, −0.16). However, the negative baseline trend reversed, resulting in 0.36 pp more SNF admissions per month after PDPM compared to baseline trends, for a 0.43pp-increase in SNF admissions by the end of Period 2 compared to the projected rate based on pre-PDPM trends.

COVID-19 was associated with an immediate 2.89 pp reduction in SNF admissions (95% CI −3.08, −2.69) compared to the baseline projected rate, and was not significantly associated with any trend changes in SNF admissions.

For HHA admissions, the baseline trend was an adjusted monthly increase of 0.04 pp (95% CI 0.01, 0.06). PDPM was significantly associated with a 0.34 pp immediate decline (95% CI −0.49, −0.18) and a 0.38 pp monthly decline in HHA admissions, for a total 1.01 pp reduction in HHA admissions by the end of Period 2 compared to the baseline projection. There was no immediate change in HHA admissions the month after PDGM implementation. However, PDGM was associated with a 1.10 pp monthly increase in HHA admissions compared to baseline, for a total 0.90 pp higher rate at the end of Period 3 compared to the projection from baseline. COVID-19 was associated with an immediate 3.66 pp increase in HHA admissions (95% CI 3.42, 3.99) but was not significantly associated with any trend changes compared to baseline projections.

Tables

In the revised Table 1, all descriptive statistics were recalculated using the smaller cohort.

In the revised Table 2, all coefficients, confidence intervals, and p-value statistics now reflect results of the analysis using the smaller cohort.

Figure 1 has been recreated using the smaller cohort.

We apologize for these errors.