Association of Medicare Advantage Enrollment Growth and Changes in Home Health Supply, 2011–2022

Medicare Advantage (MA) enrollment has grown rapidly over the last two decades. MA insurers have a strong incentive to reduce their health care spending because they are paid on a risk-adjusted per capita basis. One area they have targeted in reducing health expenditures has been home health [1-4]. MA plans have a number of tools to try to reduce home health care use, including the use of prior authorization, restrictive networks, and cost-sharing [1]. MA plans also have flexibility in the prices that they pay HHAs and may reimburse at lower levels than the administratively set prices in traditional Medicare (TM) [5]. In recent years, the number of home health agencies has declined substantially while MA enrollment has grown rapidly [5]. We investigated whether rising MA enrollment is related to declining home health supply. Our objective was to examine the association between county-level changes in MA penetration and county-level changes in home health supply and use from 2011 to 2022. We calculated the number of home health agencies serving each county in 2011 and 2022 using the Outcome and Assessment Information Set (OASIS) linked to beneficiary county from the Medicare Master Beneficiary Summary File (MBSF). OASIS contains assessments that must be completed by all Medicare-certified HHAs and includes all Medicare patients, regardless of whether they are enrolled in MA or TM. In addition to home health supply, we calculated three measures of home health use: the number of starts of care, the number of post-acute starts of care, and the number of community-initiated starts of care. We expressed all values per 1000 Medicare beneficiaries. We also used publicly available data on county MA penetration, changes in economic factors, and urbanicity. More details on measures and the sample can be found in the Supplement. For each county, we calculated the change in the MA penetration and the number of home health agencies per 1000 beneficiaries by subtracting the 2011 value from the 2022 value. We used a binned scatter plot to visually characterize this relationship. We also used ordinary least squares regressions of the change in home health supply and change in home health use on the change in MA penetration, adjusting for changes in the share of people in poverty, the median income, and the unemployment rate as well as the county's urbanicity as measured by its Rural Urban Continuum Code. Counties with greater growth in MA penetration had larger declines in the number of HHAs per 1000 beneficiaries (Figure 1). Adjusting for other factors, a 10-percentage point larger increase in MA penetration was associated with 0.54 (95% CI: −0.65, −0.42) fewer HHAs per 1000 beneficiaries, an 11.5% decline relative to the baseline mean (Table 1). A 10-percentage point larger increase in MA penetration was associated with 6.08 (95% CI: −9.20, −2.95) fewer starts of care per 1000 beneficiaries and 5.53 (95% CI: −0.810, −0.296) fewer community-admitted starts of care, declines of 5.7% and 13.7% relative to the baseline means. However, there was no statistically significant relationship between changes in MA penetration and changes in the use of post-acute care home health. In this study, we found that US counties with greater growth in MA enrollment had greater declines in home health supply and use. Our study was limited by the fact that there may be unobservable variables related to both changing MA penetration and home health supply and by the fact that changes in the number of home health agencies is an imperfect measure of access to care, for example because agencies are different sizes. As MA continues to grow, it may reshape the home health industry and beneficiary access to home health care. Concept and design (J.M., D.J.M.); statistical analysis (J.M., J.S.); data interpretation (all authors); drafting manuscript (J.M.); review of manuscript (all authors). This research was supported by the National Institute on Aging (5P01AG027296). Drs. Kosar and Rahman also received funding from the National Institute on Aging (1R01AG089051-01). This work was supported by the National Institute on Aging, 1R01AG089051-01, 5P01AG027296. The authors declare no conflicts of interest. Data S1: Supporting Information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.