Table 4:
Changes in Care for Dually Eligible Beneficiaries Associated with Medicare ACO Participation, 2009–2017
| Annual Outcomes | Any Mental Illness | Bipolar, Schizophrenia, or other Psychotic Disorders | Depression w/o Bipolar, Schizophrenia, or other Psychotic Disorders |
|---|---|---|---|
| Annual spending ($) (95% CI) | |||
| Total | 30 (−215 – 275) | 184 (−98 – 465) | −63 (−323 – 197) |
| Acute care | −11 (−118 – 96) | 52 (−63 – 66) | −38 (−156 – 80) |
| SNF | 56 (13 – 98)* | 76 (17 – 134)* | 45 (−3 – 93) |
| Psychiatric hospital | −4 (−17 – 10) | −15 (−44 – 15) | 1 (−3 – 5) |
| Utilization (/1000 beneficiaries) (95% CI) | |||
| Emergency department visits | −15.5 (−37.9 – 6.9) | −0.1 (−29.9 – 29.7) | −29.5 (−48.6 – −10.4)** |
| Acute care/critical access admissions | 3.3 (−5.2 – 11.8) | 8.6 (−1.0 – 18.2) | 0.2 (−9.9 – 10.2) |
| Ambulatory E&M visits | 191.1 (−2.2 – 384.5) | 393.9 (155.2 – 632.6)** | 60.0 (−145.6 – 265.5) |
| Prevention Quality Indicator | 1.1 (−1.7 – 3.9) | 2.4 (−0.5 – 5.4) | 0.7 (−2.6 – 3.9) |
| 7 day follow-up (%) (95% CI) | 1.2 (0.6 – 1.8)*** | 1.2 (0.6 – 1.9)*** | 0.7 (−0.3 – 1.8) |
| 30 day follow-up (%)(95% CI) | 1.3 (0.8 – 1.8)*** | 1.3 (0.7 – 1.9)*** | 1.6 (0.4 – 2.9)** |
| Readmission proportion (%) (95% CI) | −0.1 (−0.2 – 0.1) | 0.0 (−0.2 – 0.2) | −0.1 (−0.2 – 0.0) |
| Person-years | 21,571,635 | 8,861,127 | 8,920,924 |
SOURCE: Authors’ analysis of Medicare claims and enrollment data for 2009–2017. NOTES: Acute care includes critical access hospitals. CI = confidence interval, SNF = skilled nursing facility, E&M = evaluation and management. Observations in Hospital Service Areas below 25th percentile of ACO penetration excluded. Year 2012 observations for beneficiaries attributed to Medicare Shared Savings Program ACO contracts that were implemented in April or July of 2012 excluded. Prevention Quality Indicators are rates of potentially preventable hospitalizations (indicator 90) defined by the Agency for Healthcare Research and Quality. Except where otherwise noted, estimates are per beneficiary-year. “Any Mental Illness” column includes some beneficiaries who are represented in neither the “Bipolar, Schizophrenia, or other Psychotic Disorders” nor the “Depression w/o Bipolar, Schizophrenia, or other Psychotic Disorders” columns. Person-years for follow-up (readmission) were 900,209 (5,520,211) for any mental illness, 820,986 (2,205,144) for bipolar schizophrenia or other psychotic disorders and 72,953 (2,491,328) for depression without bipolar, schizophrenia or other psychotic disorders.
p<0.05
p<0.01
p<0.001