Michigan is one of several states that expanded Medicaid coverage under the Affordable Care Act (ACA). The “Healthy Michigan Plan”, implemented April 2014, provides coverage through Medicaid to adults with incomes up to 138% of the federal poverty level and requires a health risk assessment and cost sharing by enrollees.
Early results suggest that the Michigan Plan has been successful. Within one year of expansion, 600,000 new adults had enrolled.(1) Primary care service utilization increased 6% following expansion, and participation in health risk assessments are more double that of private health insurance plans.(2) Less well known is the association between Medicaid expansion and the use and outcomes of cardiovascular revascularization.
We evaluated Michigan's Medicaid expansion as it relates to access and outcomes for 7,558 coronary artery bypass grafting (CABG) operations at 33 hospitals, and 45,183 percutaneous coronary interventions (PCI) at 47 hospitals. We excluded patients >=65 years and those with Medicare, non-Medicaid state or Federal government coverage (e.g., military), or non-US insurance. Clinical data, from each institution’s participation in CABG and PCI collaboratives, represents procedures performed between April 2012 – March 2014 and April 2014 – March 2016 (24 months before/after expansion).
We evaluated changes in access (i.e., number of procedures) by insurance type (private, Medicaid, uninsured), and used risk adjustment models for CABG(3) and PCI(4) to calculate adjusted mortality (CABG: operative mortality; PCI: during the index admission) before and after expansion. G-computation was utilized to estimate the relationship between Medicaid expansion and the number of patients presenting quarterly for revascularization within insurance coverage type (Medicaid, private, uninsured). A generalized estimating equation (GEE) model using autoregressive covariance structure was fitted, including insurance type, calendar quarter, Medicaid expansion, and interaction terms (two-way and three-way interactions among quarter, Medicaid expansion, and insurance coverage type) as predictors.
The GEE (Q) model was used to generate counterfactual (i.e., if Medicaid expansion had not occurred) predicted observations for the 8 quarters occurring after expansion to estimate the causal effect of Medicaid expansion on access. Bootstrapping with 1000 replicates drawn randomly with replacement from patient data was used to generate empirical 95% confidence intervals around causal effect estimates. We additionally assessed changes in procedural appropriateness, acute kidney injury (AKI) and length of stay across time and procedure.
We estimate a 103.8% (95CI 45.8% – 182.5%), increase in Medicaid patients presenting for CABG following expansion, and a 59.6% (95%CI 41.5% – 72.1%) decrease in uninsured patients, and 8.3% (95%CI 0.2% – 20.1%) decrease in private insurance patients (Table 1). We estimate a 44.5% (95CI 24.6% – 69.9%) increase in Medicaid patients presenting for PCI following expansion, and a 53.2% (95%CI 46.1% – 59.3%) decease in uninsured patients, and 16.3% (95%CI 10.2% – 22.4%) decrease in private insurance patients. Changes in CABG and PCI representation among the Medicaid and uninsured mostly occurred in the first quarter after expansion.
Table 1.
Volume and Outcomes of Revascularization Procedures by Insurance Coverage over Time.
| Percent of Annual Volume | Mortality (% yes) | Acute Kidney Injury (% yes) | Length of Stay (median) | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Medicaid | Uninsured | Private | Medicaid | Uninsured | Private | Medicaid | Uninsured | Private | Medicaid | Uninsured | Private | ||
| Coronary Arteiy Bypass Grafting | Before (n=3838) | 13% | 12% | 75% | 1.3 | 1.2 | 0.86 | 14.7 | 10.2 | 9.2 | 6 | 5 | 5 |
| After (n=3720) | 21% | 4% | 75% | 1.3 | 1.4 | 0.93 | 13.9 | 8.7 | 10.3 | 6 | 5 | 5 | |
| Percutaneous Coronary Intervention | Before (n=22670) | 10% | 14% | 76% | 1.6 | 1.9 | 0.72 | 4.2 | 2.9 | 1.9 | 3 | 3 | 3 |
| After (n=22513) | 19% | 5% | 76% | 1.4 | 3.2 | 0.86 | 3.6 | 2.4 | 2.1 | 4 | 3 | 2 | |
For both CABG and PCI, Medicaid expansion was not significantly associated with mortality, AKI or length of stay, and there was no significant effect modification by insurance coverage (p>0.05, for all main effect as well as interactions between Medicaid expansion and types of insurance). Mortality was similar before and after expansion: CABG (0.95% vs. 1.03%) and PCI (1.07% vs. 0.98%), including among Medicaid patients: CABG (1.27% vs. 1.31%) and PCI (1.63% vs. 1.35%). Finally, expansion was not associated with significant changes in appropriateness for PCI (before: 88%; after: 91%) or CABG (before: 89%; after: 90%) -- similar patterns were seen across payers.
This study, reflecting Michigan’s statewide experience, evaluates changes in access and outcomes for patients receiving cardiovascular revascularization. Similar to Davis and colleagues who evaluated insurance coverage following Michigan’s Medicaid expansion among hospitalized patients, most of the change in access to revascularization occurred within the first quarter following expansion.(5) Michigan’s Medicaid expansion, which has been associated with reductions in predicted risk of morbidity and mortality among CABG patients, was associated changes in the demographic of those using specialty cardiovascular services (i.e., reduction in uninsured patients undergoing coronary revascularization), absent increases in mortality.
Acknowledgments
Funding/Support: The project was approved by the University of Michigan Institutional Review Board. Dr. Likosky received a grant from the Agency for Healthcare Research and Quality. Support for the Michigan Society of Thoracic and Cardiovascular Surgeons Quality Collaborative and the Blue Cross Blue Shield of Michigan Cardiovascular Consortium Percutaneous Coronary Intervention Quality Improvement Initiative is provided by the Blue Cross and Blue Shield of Michigan and Blue Care Network as part of the BCBSM Value Partnerships program. This work was also supported by funding from the National Institute of Aging (Grant No. P01AG019783-0751).
Role of the Sponsor: None of the funders had a role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.
Abbreviations
- ACA
Affordable Care Act
- CABG
coronary artery bypass grafting
- PCI
percutaneous coronary interventions
Footnotes
Disclosures: Drs. Likosky and Gurm, Mr. Seth and Ms. He had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Dr. Gurm reports a consulting role with Osprey Medical. None of the other authors reports any financial disclosures with industry.
References
- 1.Tipirneni R, Rhodes KV, Hayward RA, et al. Primary care appointment availability and nonphysician providers one year after Medicaid expansion. Am J Manag Care. 2016;22:427–31. [PubMed] [Google Scholar]
- 2.Udow-Phillips M, Lausch K, Shigekawa E, Hirth R, Ayanian J. The Medicaid Expansion Experience In Michigan. Health Affairs. 2015 [Google Scholar]
- 3.Shahian DM, O'Brien SM, Filardo G, et al. The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 1--coronary artery bypass grafting surgery. The Annals of thoracic surgery. 2009;88:S2–22. doi: 10.1016/j.athoracsur.2009.05.053. [DOI] [PubMed] [Google Scholar]
- 4.The Society of Cardvioascular Angiography and Intervntions. PCI Risk Calculator [Google Scholar]
- 5.Davis MM, Gebremariam A, Ayanian JZ. Changes in Insurance Coverage Among Hospitalized Nonelderly Adults After Medicaid Expansion in Michigan. JAMA. 2016;315:2617–8. doi: 10.1001/jama.2016.6303. [DOI] [PubMed] [Google Scholar]