Table 2. Estimated Changes in Medicare Shared Savings Program ACO-Assigned Beneficiaries, Utilization, and Expenditure Associated With Participation of FQHCs From Difference-in-Differences Analyses.
Outcome | Baseline valuea | Estimated change (95% CI) | P value | Pretrend P valueb |
---|---|---|---|---|
Assigned beneficiaries, No. of person-years | ||||
Total | 16 818.1 | 9594.1 (2249.4 to 16 938.8) | .01 | .68 |
Dual eligible | 1095.5 | 872.9 (345.9 to 1399.8) | <.001 | .35 |
Disabled | 2107.1 | 1137.6 (390.1 to 1885.1) | <.001 | .59 |
Racial and ethnic minoritizedc | 2241.9 | 1350.8 (447.4 to 2254.1) | <.001 | .68 |
Utilization, No. per 1000 person-years | ||||
FQHC and RHC visits | 302.6 | 494.5 (254.0 to 735.0) | <.001 | .99 |
Total primary care visits | 10 391.9 | 325.8 (−143.4 to 795.0) | .17 | .44 |
Outpatient emergency department visits | 691.1 | 6.7 (−43.5 to 56.9) | .79 | .64 |
Short-term acute care hospital discharges | 282.4 | 10.4 (−3.6 to 24.4) | .15 | .38 |
Skilled nursing facility discharges | 59.1 | 5.0 (−0.6 to 10.6) | .08 | >.99 |
Per-capita expenditure, $ | 10 559.5 | 169.0 (−157.6 to 495.5) | .31 | .47 |
Quality, % | ||||
Influenza immunization | 71.9 | 5.9 (1.4 to 10.4) | .01 | .44 |
Tobacco use screening and cessation intervention | 76.8 | 11.8 (3.7 to 20.0) | .004 | .73 |
Screening for depression and follow-up plan | 63.4 | 8.9 (0.5 to 17.4) | .04 | .87 |
Colorectal cancer screening | 67.1 | 1.3 (−2.4 to 5.1) | .48 | .79 |
Breast cancer screening | 71.8 | −0.8 (−3.8 to 2.1) | .57 | .48 |
Depression remission at 12 mo | 10.4 | 3.2 (−4.5 to 10.8) | .42 | .89 |
Diabetes control (glycated hemoglobin ≤9%) | 83.6 | 0.9 (−2.8 to 4.7) | .62 | .07 |
Controlling high blood pressure | 72.2 | 0.8 (−1.6 to 3.1) | .52 | .65 |
Abbreviations: ACO, accountable care organization; FQHC, federally qualified health center; RHC, rural health clinic.
Means of outcomes across all pretreatment years among ACOs that included FQHCs during 2016 to 2022.
Pretrend tests report the P value from a joint F test that the estimated coefficients of up to 3 years before FQHC inclusions are equal and not statistically significant from zero.
Includes all beneficiaries who are American Indian or Alaska Native, Asian and Pacific Islander, Black, Hispanic, other (no additional information available for this group from the Centers for Medicare & Medicaid Services data), and unknown.