[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Table 1.  
Characteristics of Patients With Coronary Artery Disease Enrolled in Medicare Advantage (MA) and Traditional Fee-for-Service (FFS) Medicare
Characteristics of Patients With Coronary Artery Disease Enrolled in Medicare Advantage (MA) and Traditional Fee-for-Service (FFS) Medicare
Table 2.  
Quality of Care in Eligible Patients With Coronary Artery Disease Enrolled in Medicare Advantage (MA) vs Traditional Fee-for-Service (FFS) Medicare
Quality of Care in Eligible Patients With Coronary Artery Disease Enrolled in Medicare Advantage (MA) vs Traditional Fee-for-Service (FFS) Medicare
Table 3.  
Differences in Intermediate Outcomes in Patients With Coronary Artery Disease Enrolled in Medicare Advantage (MA) vs Traditional Fee-for-Service (FFS) Medicare
Differences in Intermediate Outcomes in Patients With Coronary Artery Disease Enrolled in Medicare Advantage (MA) vs Traditional Fee-for-Service (FFS) Medicare
Table 4.  
Guideline-Based Coronary Artery Disease (CAD) Therapy and Intermediate Outcomes in Patients Enrolled in Medicare Advantage Compared With Fee-for-Service Medicare
Guideline-Based Coronary Artery Disease (CAD) Therapy and Intermediate Outcomes in Patients Enrolled in Medicare Advantage Compared With Fee-for-Service Medicare
1.
Jacobson  G, Damico  A, Neuman  T, Gold  M. Medicare Advantage 2017 spotlight: enrollment market update. https://www.kff.org/medicare/issue-brief/medicare-advantage-2017-spotlight-enrollment-market-update/. Accessed February 9, 2018.
2.
Swensen  S, Mohta  NS, Prewitt  E. Leadership survey: anticipating the Trump administration’s impact on health care. https://catalyst.nejm.org/trump-administration-healthcare-impact/. Accessed September 7, 2018.
3.
Ayanian  JZ, Landon  BE, Zaslavsky  AM, Saunders  RC, Pawlson  LG, Newhouse  JP.  Medicare beneficiaries more likely to receive appropriate ambulatory services in HMOs than in traditional Medicare.  Health Aff (Millwood). 2013;32(7):1228-1235. doi:10.1377/hlthaff.2012.0773PubMedGoogle ScholarCrossref
4.
Ayanian  JZ, Landon  BE, Zaslavsky  AM, Newhouse  JP.  Racial and ethnic differences in use of mammography between Medicare Advantage and traditional Medicare.  J Natl Cancer Inst. 2013;105(24):1891-1896. doi:10.1093/jnci/djt333PubMedGoogle ScholarCrossref
5.
Kumar  A, Rahman  M, Trivedi  AN, Resnik  L, Gozalo  P, Mor  V.  Comparing post-acute rehabilitation use, length of stay, and outcomes experienced by Medicare fee-for-service and Medicare Advantage beneficiaries with hip fracture in the United States: a secondary analysis of administrative data.  PLoS Med. 2018;15(6):e1002592. doi:10.1371/journal.pmed.1002592PubMedGoogle ScholarCrossref
6.
Huckfeldt  PJ, Escarce  JJ, Rabideau  B, Karaca-Mandic  P, Sood  N.  Less intense postacute care, better outcomes for enrollees in Medicare Advantage than those in fee-for-service.  Health Aff (Millwood). 2017;36(1):91-100. doi:10.1377/hlthaff.2016.1027PubMedGoogle ScholarCrossref
7.
Tompkins  C, Higgins  A, Perloff  J, Veselovskiy  G. Population health management in Medicare Advantage. https://www.healthaffairs.org/do/10.1377/hblog20130402.029363/full/. Accessed September 7, 2018.
8.
Rosenzweig  JL, Taitel  MS, Norman  GK, Moore  TJ, Turenne  W, Tang  P.  Diabetes disease management in Medicare Advantage reduces hospitalizations and costs.  Am J Manag Care. 2010;16(7):e157-e162.PubMedGoogle Scholar
9.
Centers for Medicare & Medicaid Services. 2017 Star ratings in Medicare Advantage. https://www.cms.gov/newsroom/fact-sheets/2017-star-ratings. Accessed February 9, 2018.
10.
Jacobson  G, Damico  A, Neuman  T, Gold  M. Medicare Advantage 2015 data spotlight: overview of plan changes. https://www.kff.org/medicare/issue-brief/medicare-advantage-2015-data-spotlight-overview-of-plan-changes/. Accessed September 7, 2018.
11.
Morgan  RO, Virnig  BA, DeVito  CA, Persily  NA.  The Medicare-HMO revolving door: the healthy go in and the sick go out.  N Engl J Med. 1997;337(3):169-175. doi:10.1056/NEJM199707173370306PubMedGoogle ScholarCrossref
12.
Medicare Payment Advisory Committee (MedPAC). Improving risk adjustment in the Medicare program. http://www.medpac.gov/docs/default-source/reports/jun14_ch02.pdf?sfvrsn=0. Accessed September 7, 2018.
13.
US Government Accountability Office. Medicare Advantage: CMS should use data on disenrollment and beneficiary health status to strengthen oversight. https://www.gao.gov/products/GAO-17-393. Accessed September 7, 2018.
14.
Martino  SC, Elliott  MN, Haviland  AM, Saliba  D, Burkhart  Q, Kanouse  DE.  Comparing the health care experiences of Medicare beneficiaries with and without depressive symptoms in Medicare managed care versus fee-for-service.  Health Serv Res. 2016;51(3):1002-1020. doi:10.1111/1475-6773.12359PubMedGoogle ScholarCrossref
15.
Gold  M, Casillas  G. What do we know about health care access and quality in Medicare Advantage versus the traditional Medicare program? http://kff.org/medicare/report/what-do-we-know-about-health-care-access-and-quality-in-medicare-advantage-versus-the-traditional-medicare-program. Accessed September 7, 2018.
16.
Iezzoni  LI.  Assessing quality using administrative data.  Ann Intern Med. 1997;127(8, pt 2):666-674. doi:10.7326/0003-4819-127-8_Part_2-199710151-00048PubMedGoogle ScholarCrossref
17.
Lawson  EH, Louie  R, Zingmond  DS,  et al.  Using both clinical registry and administrative claims data to measure risk-adjusted surgical outcomes.  Ann Surg. 2016;263(1):50-57. doi:10.1097/SLA.0000000000001031PubMedGoogle ScholarCrossref
18.
Maddox  TM, Chan  PS, Spertus  JA,  et al.  Variations in coronary artery disease secondary prevention prescriptions among outpatient cardiology practices: insights from the NCDR (National Cardiovascular Data Registry).  J Am Coll Cardiol. 2014;63(6):539-546. doi:10.1016/j.jacc.2013.09.053PubMedGoogle ScholarCrossref
19.
Newhouse  JP, Price  M, Huang  J, McWilliams  JM, Hsu  J.  Steps to reduce favorable risk selection in Medicare Advantage largely succeeded, boding well for health insurance exchanges.  Health Aff (Millwood). 2012;31(12):2618-2628. doi:10.1377/hlthaff.2012.0345PubMedGoogle ScholarCrossref
20.
Newhouse  JP, Price  M, McWilliams  JM, Hsu  J, McGuire  TG.  How much favorable selection is left in Medicare Advantage?  Am J Health Econ. 2015;1(1):1-26. doi:10.1162/AJHE_a_00001PubMedGoogle ScholarCrossref
21.
Patterson  ME, Hernandez  AF, Hammill  BG,  et al.  Process of care performance measures and long-term outcomes in patients hospitalized with heart failure.  Med Care. 2010;48(3):210-216. doi:10.1097/MLR.0b013e3181ca3eb4PubMedGoogle ScholarCrossref
22.
Bradley  EH, Herrin  J, Elbel  B,  et al.  Hospital quality for acute myocardial infarction: correlation among process measures and relationship with short-term mortality.  JAMA. 2006;296(1):72-78. doi:10.1001/jama.296.1.72PubMedGoogle ScholarCrossref
23.
Nguyen  QD, Peters  E, Wassef  A, Desmarais  P, Rémillard-Labrosse  D, Tremblay-Gravel  M.  Evolution of age and female representation in the most-cited randomized controlled trials of cardiology of the last 20 years.  Circ Cardiovasc Qual Outcomes. 2018;11(6):e004713. doi:10.1161/CIRCOUTCOMES.118.004713PubMedGoogle ScholarCrossref
24.
Medicare Payment Advisory Committee (MedPAC). Report to the Congress: Medicare payment policy. http://www.medpac.gov/docs/default-source/reports/mar17_entirereport.pdf. Accessed September 7, 2018.
Original Investigation
February 20, 2019

Differences in Management of Coronary Artery Disease in Patients With Medicare Advantage vs Traditional Fee-for-Service Medicare Among Cardiology Practices

Author Affiliations
  • 1Department of Medicine, Harvard Medical School, Boston, Massachusetts
  • 2Department of Health Policy and Management, Harvard T. H. Chan School of Public Health, Boston, Massachusetts
  • 3Division of General Internal Medicine, Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts
  • 4Division of Cardiology, Massachusetts General Hospital, Boston
  • 5VA Boston Healthcare System, Boston, Massachusetts
  • 6Boston University School of Public Health, Boston, Massachusetts
  • 7Division of Endocrinology, Department of Medicine, Brigham and Women’s Hospital, Boston, Massachusetts
  • 8Baim Clinical Research Institute, Boston, Massachusetts
  • 9Department of Biostatistics, Boston University, Boston, Massachusetts
  • 10Cardiovascular Division, Department of Medicine, Washington University in St Louis School of Medicine, St Louis, Missouri
JAMA Cardiol. 2019;4(3):265-271. doi:10.1001/jamacardio.2019.0007
Key Points

Question  Are there meaningful differences in evidence-based secondary prevention treatment and intermediate outcomes among patients with coronary artery disease enrolled in Medicare Advantage (MA) vs traditional fee-for-service (FFS) Medicare?

Findings  In this cohort study using data from the Practice Innovation and Clinical Excellence (PINNACLE) registry, we found that patients enrolled in MA had more comorbidities than those in FFS Medicare and were more likely to receive guideline-recommended secondary prevention treatments when eligible. However, we found no differences in intermediate outcomes between patients enrolled in MA and FFS Medicare, including blood pressure and cholesterol levels.

Meaning  These findings suggest that MA plans may drive improvements in process-based quality measures for Medicare beneficiaries, although this may not translate into better patient outcomes over FFS Medicare.

Abstract

Importance  One-third of Medicare beneficiaries are enrolled in Medicare Advantage (MA), Medicare’s private plan option. Medicare Advantage incentivizes performance on evidence-based care, but limited information exists using reliable clinical data to determine whether this translates into better quality for patients with coronary artery disease (CAD) enrolled in MA compared with those enrolled in traditional fee-for-service (FFS) Medicare.

Objective  To determine differences in evidence-based secondary prevention treatments and intermediate outcomes among patients with CAD enrolled in MA vs FFS Medicare.

Design, Setting, and Participants  In this observational, retrospective, cohort study, deidentified data from patients 18 years or older diagnosed as having CAD between January 1, 2013, and May 1, 2014, at cardiology practices participating in the Practice Innovation and Clinical Excellence (PINNACLE) registry were studied, including 35 563 patients enrolled in MA and 172 732 enrolled in FFS Medicare. Data were analyzed from March to July 2018.

Exposures  Medicare Advantage enrollment.

Main Outcomes and Measures  Medication prescription patterns among eligible patients and intermediate outcomes, including blood pressure and low-density lipoprotein cholesterol.

Results  Of the 35 563 patients with CAD enrolled in MA, 20 193 (56.8%) were male, and the mean (SD) age was 76.7 (7.6) years; of the 172 732 patients with CAD enrolled in FFS Medicare, 100 025 (57.9%) were male, and the mean (SD) age was 77.5 (8.0) years. Patients enrolled in MA were younger, less likely to be white, and more likely to be female and to have heart failure, diabetes, and chronic kidney disease compared with those enrolled in FFS Medicare. Compared with FFS Medicare beneficiaries, MA beneficiaries were more likely to receive secondary prevention treatments, including β-blockers (80.6% vs 78.8%; P < .001), angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers (70.7% vs 65.1%; P < .001), and statins (68.4% vs 64.5%; P < .001). Patients enrolled in MA were also more likely to receive all 3 medications when eligible (48.9% vs 40.4%; P < .001). After adjustment, MA beneficiaries had higher odds of receiving guideline-recommended therapy compared with FFS Medicare beneficiaries for β-blockers (odds ratio, 1.10; 95% CI, 1.04-1.17; P = .002), angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers (odds ratio, 1.13; 95% CI, 1.08-1.19; P < .001), and all 3 medications (odds ratio, 1.23; 95% CI, 1.001-1.50; P = .047). There were no significant differences in intermediate outcomes between those enrolled in MA and FFS Medicare, including systolic and diastolic blood pressure and low-density lipoprotein cholesterol levels.

Conclusions and Relevance  Among patients with CAD in the PINNACLE registry, MA beneficiaries had more comorbidities than FFS Medicare beneficiaries and were more likely to receive secondary prevention treatments. However, this did not translate into differences in intermediate outcomes. These findings suggest that MA plans may drive improvements in process-based quality measures for Medicare beneficiaries, although this may have a limited effect on improving patient outcomes over FFS Medicare.

Introduction

Medicare Advantage (MA) is Medicare’s managed care alternative in which Medicare beneficiaries can elect to enroll in private insurance plans rather than in traditional fee-for-service (FFS) Medicare. Over the past decade, enrollment in MA has grown substantially, with private plans covering 33% of Medicare beneficiaries in 2017, up from 22% in 2008.1 Additionally, there is increased interest in MA under current political leadership, as Congress and the White House have expressed support for more market-driven health reform.2

Prior studies using administrative claims data have suggested that beneficiaries enrolled in MA receive higher-quality care than beneficiaries in FFS Medicare, although limited data exist on clinical outcomes.3-6 Proponents of MA argue this is because of several program features. First, patients enrolled in MA plans receive additional support—for example, in the form of disease management, coaching, or navigator services—as part of their plan benefits.7,8 Second, since plans are required to submit quality data to Medicare and can receive bonuses for high performance, they have financial incentives to follow evidence-based guidelines and to achieve better blood pressure and lipid control.9,10

However, earlier reports suggest that differences between patients enrolled in MA and FFS Medicare are because of selection effects—namely, that healthier, more engaged beneficiaries elect to enroll in MA, while sicker ones remain in FFS Medicare.11,12 Furthermore, a 2017 report by the US Government Accountability Office13 found that sicker patients are more likely to disenroll from MA plans because of concerns about access to care and restrictions from seeing preferred physicians and hospitals. Other work found that patients with mental illness enrolled in MA are more likely to report difficulty getting appropriate care and medications and, as a result, rate their experience worse than those in FFS Medicare.14 Therefore, there is concern that the quality of care for patients with long-term conditions, such as coronary artery disease (CAD), may be compromised under MA.

As the number of MA enrollees and interest in this program continues to expand, there is a growing need to understand whether quality of care differs between MA and FFS Medicare beneficiaries. Prior studies attempting to answer this question have largely relied on claims data and have been limited by a lack of high-quality, granular clinical data with which to compare patient characteristics and performance.15 On the other hand, clinical registries are generally considered a more valid, reliable data source for quality measurement than administrative claims data.16,17 One reason is that they use clinical personnel to ensure standardization of the data collection, while claims do not and therefore are subject to differences in coding and billing practices.

The National Cardiovascular Disease Registry Practice Innovation and Clinical Excellence (PINNACLE) registry presents a unique opportunity to assess whether MA beneficiaries differ in terms of their demographic characteristics and comorbidities and whether they receive higher-quality care than beneficiaries in FFS Medicare.18 In this article, we sought to answer the following key questions. First, are there clinically meaningful differences in characteristics of patients with CAD enrolled in MA compared with FFS Medicare? Second, is enrollment in a MA plan associated with significant differences in the receipt of evidence-based secondary prevention treatments among patients with CAD? Finally, do patients enrolled in MA experience better intermediate outcomes, including lipid control and blood pressure, compared with patients with FFS Medicare?

Methods
Data

This study used data from the PINNACLE registry, which is a national outpatient-based cardiac quality improvement registry of patients seen in cardiology practices across the United States. The PINNACLE registry was started by the American College of Cardiology Foundation in 2008 and includes detailed patient-level information, including demographic characteristics, comorbidities, vital signs, medications, contraindications to medications, and laboratory values. Cardiology practices voluntarily chose to participate in this registry. Data are collected by either paper-based report forms or by exporting information from practice’s electronic health records into the PINNACLE registry. Data collection is standardized through written definitions and uniform data entry. Data quality checks are performed periodically to ensure accuracy. Because of the deidentified nature of the data and the fact that all data are held at the analytic center and not released to researchers, a waiver of written informed consent and authorization for this study was granted by Chesapeake Research Review Incorporated.

Study Population

We first identified all patients 18 years or older diagnosed as having CAD between January 1, 2013, and May 1, 2014. We defined CAD as those with a history of myocardial infarction (MI), percutaneous coronary intervention, or coronary artery bypass grafting. Patients missing key characteristics (ie, data on age, sex, or comorbidities) were excluded. We then selected only patients with either MA or FFS Medicare coverage. To confirm enrollment in MA vs FFS Medicare, we linked the data set to the Medicare Beneficiary Summary File. Patients were classified as being enrolled in MA if they had at least 1 month of enrollment. As a sensitivity analysis, we examined patients who were continuously enrolled in MA or FFS Medicare for the entire study period from 2013 to 2014. Patient characteristics, including age, sex, race/ethnicity, and dual eligibility for Medicare and Medicaid, were collected at the last patient encounter. Patient’s long-term conditions were recorded at any interval during the study period.

Outcomes

We evaluated patient eligibility for guideline-recommended CAD therapy. These include (1) β-blocker therapy in those with prior MI or left ventricular ejection fraction (LVEF) less than 40%, (2) angiotensin-converting enzyme (ACE) inhibitors or angiotensin II receptor blockers (ARBs) for those with LVEF less than 40% and/or diabetes, and (3) statin therapy in patients with CAD with a low-density lipoprotein cholesterol (LDL-C) concentration of 100 mg/dL or greater. Patients with documented reasons for not being prescribed any of the specified medications or referrals were excluded from the analyses of that particular metric. If there were no documented reasons for the lack of prescription, the patient was considered eligible to receive medication. We also calculated a combined prescription rate in patients eligible for all 3 medications. Finally, we also assessed referral to cardiac rehabilitation. Rates for each specific performance measure were calculated by dividing the number of eligible patients prescribed appropriate medication for a given quality measure by the number of eligible patients. We also examined intermediate outcomes, including systolic and diastolic blood pressure (millimeters of mercury) and LDL-C concentration (milligrams per deciliter).

Statistical Analyses

We first compared differences in patient characteristics of beneficiaries with MA vs FFS. We used t tests for continuous variables and χ2 tests for categorical variables. Second, we compared the rate of eligible patients enrolled in MA and FFS Medicare receiving guideline-recommended prescriptions for each individual medication as well as whether they were referred to cardiac rehabilitation if appropriate. We used multivariable hierarchical logistic regression models with patient characteristics as fixed effects and practice sites as a random effect to account for correlation of patients within the same practice. Additionally, we compared a combined prescription rate for patients eligible to receive all 3 medications and repeated the model. For the intermediate outcomes that are continuous variables, we used hierarchical linear regression models adjusted for patient characteristics, including age, sex, race/ethnicity, current tobacco use, and the presence of the following comorbidities: heart failure, dyslipidemia, diabetes, hypertension, peripheral vascular disease, stroke or transient ischemic attack, angina, atrial fibrillation or flutter, chronic liver disease, and chronic kidney disease. As a sensitivity analysis, we repeated our models for patients who were continuously enrolled in MA or FFS Medicare for the entire study period.

All P values were 2-sided, and statistical significance was set at a P value less than .05. Analyses were performed using SAS version 9.4 (SAS Institute) by analysts at the Baim Institute for Clinical Research.

Results
Patient Characteristics

There were 279 278 patients enrolled in Medicare diagnosed as having CAD from January 2013 to December 2014 in the PINNACLE data set. Of these, we were able to link 224 175 patients with CAD enrolled in Medicare (80.3%) to the Medicare denominator file to determine whether they were enrolled in MA or FFS Medicare; 35 563 were enrolled in MA, and 172 732 were enrolled in FFS Medicare. On average, compared with FFS Medicare beneficiaries, MA beneficiaries were slightly younger (mean [SD] age, 76.7 [7.6] years vs 77.5 [8.0] years; P < .001), more likely to be female (43.2% vs 42.1%; P < .001), less likely to be white (69.8% vs 73.7%; P < .001), more likely to use tobacco products (13.7% vs 12.6%; P < .001), and more likely to be dually eligible for Medicare and Medicaid (5.7% vs 3.4%; P < .001). Patients enrolled in MA were also more likely to have heart failure (36.6% vs 34.6%; P < .001), diabetes (31.6% vs 28.0%; P < .001), peripheral vascular disease (14.2% vs 11.7%; P < .001), and chronic kidney disease (6.6% vs 5.6%; P < .001) and less likely to have atrial fibrillation or flutter (31.1% vs 33.1%; P < .001) (Table 1). Patients enrolled in MA were also slightly more likely to have had percutaneous coronary intervention within the past year (3.6% vs 2.5%; P < .001), MI within the past year (6.4% vs 5.4%; P < .001), or MI at any point (33.0% vs 30.4%; P < .001).

Differences in CAD Secondary Prevention Treatments and Intermediate Outcomes

Among individuals eligible for each secondary prevention treatment, MA beneficiaries were more likely to receive β-blockers (80.6% vs 78.8%; P < .001), ACE inhibitors or ARBs if they had diabetes or an LVEF of 40% or lower (70.7% vs 65.1%; P < .001), statins if they had LDL-C levels of 100 mg/dL or greater (68.4% vs 64.5%; P < .001), and referral to cardiac rehabilitation if eligible (6.4% vs 5.5%; P < .001) (Table 2). When indicated, MA beneficiaries were more likely to receive all 3 prescriptions than those enrolled in FFS Medicare (48.9% vs 40.4%; P < .001). Of note, very few patients across each category had a documented contraindication (eTable 1 in the Supplement). There were no significant differences in systolic blood pressure, diastolic blood pressure, or LDL-C levels between MA and FFS Medicare beneficiaries (Table 3).

In our sensitivity analysis, we examined differences between patients continuously enrolled in MA (n = 23 947) and patients continuously enrolled in FFS Medicare (n = 165 161) during our study period. We found similar results. Continuously enrolled MA beneficiaries had higher rates of receiving appropriate CAD therapy (eTable 2 in the Supplement).

Association of MA Enrollment With Guideline-Recommended Therapy and Intermediate Outcomes

After multivariate adjustment for comorbidities and patient characteristics, patients with CAD enrolled in MA had higher odds of receiving β-blockers when indicated (odds ratio, 1.10; 95% CI, 1.04-1.17; P = .002) and ACE inhibitors or ARBs if they had diabetes or reduced LVEF (odds ratio, 1.13; 95% CI, 1.08-1.19; P < .001). Eligible patients enrolled in MA also had higher odds of receiving all 3 medications compared with those enrolled in FFS Medicare (odds ratio, 1.23; 95% CI, 1.001-1.50; P = .047). There were no differences in the odds of receiving either a statin or referral to cardiac rehabilitation. Lastly, after multivariate adjustment, differences in intermediate outcomes remained statistically nonsignificant between MA and FFS Medicare beneficiaries (Table 4).

In our sensitivity analyses of patients who were continuously enrolled in MA or FFS Medicare, our results were qualitatively similar. Continuously enrolled MA beneficiaries were significantly more likely to be prescribed β-blockers and ACE inhibitors or ARBs (eTable 3 in the Supplement). They were numerically more likely to receive statins or all 3 medications when indicated, although this did not reach statistical significance. When examining intermediate outcomes, there were no significant differences between continuously enrolled MA vs FFS Medicare beneficiaries for systolic blood pressure or LDL-C levels (eTable 4 in the Supplement). There was a small difference in diastolic blood pressure, with continuously enrolled MA beneficiaries having a lower diastolic blood pressure compared with FFS Medicare beneficiaries.

Discussion

In an analysis of a national registry of Medicare beneficiaries with CAD, we found differences between patients enrolled in MA vs traditional FFS Medicare. Patients enrolled in MA were slightly younger and had more comorbidities than those enrolled in FFS Medicare. Additionally, MA beneficiaries were more likely to have received guideline-recommended CAD therapy for secondary prevention compared with FFS Medicare beneficiaries. However, we found no meaningful differences in intermediate outcomes, including blood pressure and lipid control, between patients enrolled in MA vs FFS Medicare. To our knowledge, this study represents the first national comparison study examining quality of care using detailed clinical data among patients enrolled in Medicare by insurance status.

Our first finding is that patients with CAD enrolled in MA in the PINNACLE registry had a higher burden of most comorbidities than their FFS Medicare counterparts. This is in significant contrast to earlier reports showing that MA plans tend to attract healthier beneficiaries than FFS Medicare.11,12 However, these older studies comparing MA vs FFS Medicare relied on administrative claims, while we used clinical registry data built from the electronic health record. In theory, registry data obtained from the electronic health record should be much more consistent across payers, leading to more accurate ascertainment of comorbidities and a more appropriate comparison. Additionally, registry data contains more granular clinical information, including information on severity of disease, like class of congestive heart failure. While other recent work has also suggested that the differences in disease severity between MA and FFS Medicare may be narrowing,19,20 some have thought that the change is solely an artifact of coding; because MA plans get higher payments for caring for complex patients, they are incentivized to ensure optimal coding of all patients’ clinical comorbidities on an annual basis, which some have postulated has led to upcoding. Our findings would suggest that the narrowing is real and that, at least within the CAD population, MA beneficiaries are as sick or sicker than FFS Medicare beneficiaries.

Our second finding is that patients enrolled in MA were more likely to receive guideline-recommended CAD therapy than those enrolled in FFS Medicare. Receipt of guideline-based CAD treatment may be higher among MA beneficiaries for several reasons. First, managed care plans often directly incentivize the delivery of evidence-based treatments that have been shown to improve quality outcomes. These guideline-based treatments overlap with multiple Healthcare Effectiveness Data and Information Set measures, which are the basis on which MA plans receive star ratings and quality bonuses from Medicare. For example, Healthcare Effectiveness Data and Information Set measures include assessments of the proportion of patients with hypertension whose blood pressure was well controlled, comprehensive care for patients with diabetes (including use of ACE inhibitors and ARBs), rates of persistent treatment with β-blockers following acute MI, and cholesterol management for patients with cardiovascular conditions. This undoubtedly creates additional incentives for MA plans to focus on improving these measures, particularly given the fact that many plans use their star rating in marketing materials to drive enrollment. Other reasons for better performance on process measures may include the additional benefits offered by MA plans (eg, disease management programs and pharmacist medication reconciliation services) or simply that a greater share of MA beneficiaries are enrolled in Part D drug benefits compared with FFS Medicare beneficiaries and therefore may be better able to access and afford any prescribed medications. These findings may be of interest to policymakers hoping to use more market-driven strategies, such as private insurance coverage, to drive improvements in quality in Medicare.

Finally, while we find statistically significant, meaningful differences in process measures, we do not find significant differences in intermediate clinical outcome measures, including blood pressure and cholesterol control. Small effects that might be meaningful at the population level would also not be detected in our sample. Additionally, process measures are not always strong indicators of outcomes.21,22 Among the older adult population, a group often underrepresented in randomized clinical trials targeting coronary artery disease and heart failure,23 guideline-recommended therapies may not be as effective as they are in the populations that participated in the trials. To the extent this is true, this may indicate that the process measures currently being collected are inadequate to drive improvements in outcomes in an older Medicare population. Therefore, policymakers might temper their expectations on using MA plans, especially as enrollment continues to grow, as a means for improving patient outcomes for Medicare-enrolled patients. However, since MA plans outperformed FFS Medicare on process measures, one strategy the US Centers for Medicare and Medicaid Services could pursue is revising the measures it focuses on for financial incentives under MA to ones that are more closely tied to outcomes in the Medicare population.

Whether or not our findings of higher quality with similar intermediate outcomes represent high-value care under MA compared with FFS Medicare is an important area for future work. Since the passage of the Affordable Care Act, the Centers for Medicare and Medicaid Services began reducing payment to MA plans to be much more like those in FFS Medicare. However, the latest report from the Medicare Payment Advisory Commission (MedPAC)24 reported that MA payments to care for Medicare beneficiaries are still higher at 103% compared with the FFS Medicare costs.

Our analysis is in line with a literature showing higher quality on process measures for MA beneficiaries compared with FFS Medicare beneficiaries.3-6 Building on prior work, ours is one of few studies examining variation in clinically meaningful outcomes for CAD between patients enrolled in MA and FFS Medicare using a robust set of clinical data that does not rely on billing codes.

Limitations

Our study has several limitations. First, the PINNACLE registry is a voluntary registry, so participating practices may differ from those that did not join. Relative to the national average, patients in the PINNACLE registry are also less likely to be dually eligible for Medicare and Medicaid and live in areas with higher median incomes. Further, it is possible that participation in the PINNACLE registry may have influenced quality of care, and therefore the high rates of guideline-based therapy we documented may not reflect practices more broadly. Second, we determined whether patients were enrolled in MA vs FFS Medicare based on their first year of data in the study period. We also defined MA participation as any patient enrolled in MA for at least 1 month. It is possible that some patients may have switched plans, which would bias our results to the null owing to potential misclassification of binary treatment assignment. However, rates of switching plans between MA and FFS Medicare are low,19 and therefore, it is unlikely that the results would be substantially impacted by this, as was confirmed by our sensitivity analyses, which showed qualitatively similar results. Third, as with all observational studies, we can only report associations and not prove a causal relationship between enrollment in MA and performance on quality measures. However, given the practical limitations of performing randomized clinical trials at scale in the real-world setting to test the effect of MA on quality, we are limited to observational studies. In the future, the Centers for Medicare and Medicaid Services may consider, on a limited basis, further exploring a study design to better test the effect of MA on quality. Fourth, we were unable to examine clinical outcomes such as mortality because these data are not available for MA beneficiaries. Finally, we were unable to determine how long patients had been on the prescribed CAD therapy (eg, length of treatment with a statin and levels of cholesterol), which has implications for differences in intermediate outcomes.

Conclusions

Using high-quality clinical registry data, we found that patients enrolled in MA had a slightly higher comorbidity burden and received higher quality care as measured by process measures than their traditional FFS Medicare counterparts. However, intermediate outcomes did not vary significantly between the 2 groups. As MA continues to enroll a higher proportion of beneficiaries each year, it will be important to monitor both quality and outcomes of care to determine whether these patterns ultimately lead to better outcomes in Medicare.

Back to top
Article Information

Accepted for Publication: December 21, 2018.

Corresponding Author: Jose F. Figueroa, MD, MPH, Division of General Internal Medicine, Department of Medicine, Brigham and Women’s Hospital, 42 Church St, Cambridge, MA 02138 (jfigueroa@hsph.harvard.edu).

Published Online: February 20, 2019. doi:10.1001/jamacardio.2019.0007

Author Contributions: Mr Song had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Figueroa, Blumenthal, Feyman, Turchin, Joynt Maddox.

Acquisition, analysis, or interpretation of data: Figueroa, Frakt, Turchin, Doros, Gao, Song, Joynt Maddox.

Drafting of the manuscript: Figueroa, Blumenthal, Gao, Joynt Maddox.

Critical revision of the manuscript for important intellectual content: Figueroa, Blumenthal, Feyman, Frakt, Turchin, Doros, Song, Joynt Maddox.

Statistical analysis: Frakt, Doros, Gao, Song.

Obtained funding: Figueroa.

Administrative, technical, or material support: Figueroa, Song.

Study supervision: Figueroa, Blumenthal, Frakt, Joynt Maddox.

Conflict of Interest Disclosures: Dr Figueroa is partly funded by grant KL2 TR002542 from the National Center for Advancing Translational Sciences. Dr Blumenthal is the medical director of Devoted Health, which is a Medicare Advantage health plan. Dr Joynt Maddox has a contract with the US Department of Health and Human Services. No other disclosures were reported.

References
1.
Jacobson  G, Damico  A, Neuman  T, Gold  M. Medicare Advantage 2017 spotlight: enrollment market update. https://www.kff.org/medicare/issue-brief/medicare-advantage-2017-spotlight-enrollment-market-update/. Accessed February 9, 2018.
2.
Swensen  S, Mohta  NS, Prewitt  E. Leadership survey: anticipating the Trump administration’s impact on health care. https://catalyst.nejm.org/trump-administration-healthcare-impact/. Accessed September 7, 2018.
3.
Ayanian  JZ, Landon  BE, Zaslavsky  AM, Saunders  RC, Pawlson  LG, Newhouse  JP.  Medicare beneficiaries more likely to receive appropriate ambulatory services in HMOs than in traditional Medicare.  Health Aff (Millwood). 2013;32(7):1228-1235. doi:10.1377/hlthaff.2012.0773PubMedGoogle ScholarCrossref
4.
Ayanian  JZ, Landon  BE, Zaslavsky  AM, Newhouse  JP.  Racial and ethnic differences in use of mammography between Medicare Advantage and traditional Medicare.  J Natl Cancer Inst. 2013;105(24):1891-1896. doi:10.1093/jnci/djt333PubMedGoogle ScholarCrossref
5.
Kumar  A, Rahman  M, Trivedi  AN, Resnik  L, Gozalo  P, Mor  V.  Comparing post-acute rehabilitation use, length of stay, and outcomes experienced by Medicare fee-for-service and Medicare Advantage beneficiaries with hip fracture in the United States: a secondary analysis of administrative data.  PLoS Med. 2018;15(6):e1002592. doi:10.1371/journal.pmed.1002592PubMedGoogle ScholarCrossref
6.
Huckfeldt  PJ, Escarce  JJ, Rabideau  B, Karaca-Mandic  P, Sood  N.  Less intense postacute care, better outcomes for enrollees in Medicare Advantage than those in fee-for-service.  Health Aff (Millwood). 2017;36(1):91-100. doi:10.1377/hlthaff.2016.1027PubMedGoogle ScholarCrossref
7.
Tompkins  C, Higgins  A, Perloff  J, Veselovskiy  G. Population health management in Medicare Advantage. https://www.healthaffairs.org/do/10.1377/hblog20130402.029363/full/. Accessed September 7, 2018.
8.
Rosenzweig  JL, Taitel  MS, Norman  GK, Moore  TJ, Turenne  W, Tang  P.  Diabetes disease management in Medicare Advantage reduces hospitalizations and costs.  Am J Manag Care. 2010;16(7):e157-e162.PubMedGoogle Scholar
9.
Centers for Medicare & Medicaid Services. 2017 Star ratings in Medicare Advantage. https://www.cms.gov/newsroom/fact-sheets/2017-star-ratings. Accessed February 9, 2018.
10.
Jacobson  G, Damico  A, Neuman  T, Gold  M. Medicare Advantage 2015 data spotlight: overview of plan changes. https://www.kff.org/medicare/issue-brief/medicare-advantage-2015-data-spotlight-overview-of-plan-changes/. Accessed September 7, 2018.
11.
Morgan  RO, Virnig  BA, DeVito  CA, Persily  NA.  The Medicare-HMO revolving door: the healthy go in and the sick go out.  N Engl J Med. 1997;337(3):169-175. doi:10.1056/NEJM199707173370306PubMedGoogle ScholarCrossref
12.
Medicare Payment Advisory Committee (MedPAC). Improving risk adjustment in the Medicare program. http://www.medpac.gov/docs/default-source/reports/jun14_ch02.pdf?sfvrsn=0. Accessed September 7, 2018.
13.
US Government Accountability Office. Medicare Advantage: CMS should use data on disenrollment and beneficiary health status to strengthen oversight. https://www.gao.gov/products/GAO-17-393. Accessed September 7, 2018.
14.
Martino  SC, Elliott  MN, Haviland  AM, Saliba  D, Burkhart  Q, Kanouse  DE.  Comparing the health care experiences of Medicare beneficiaries with and without depressive symptoms in Medicare managed care versus fee-for-service.  Health Serv Res. 2016;51(3):1002-1020. doi:10.1111/1475-6773.12359PubMedGoogle ScholarCrossref
15.
Gold  M, Casillas  G. What do we know about health care access and quality in Medicare Advantage versus the traditional Medicare program? http://kff.org/medicare/report/what-do-we-know-about-health-care-access-and-quality-in-medicare-advantage-versus-the-traditional-medicare-program. Accessed September 7, 2018.
16.
Iezzoni  LI.  Assessing quality using administrative data.  Ann Intern Med. 1997;127(8, pt 2):666-674. doi:10.7326/0003-4819-127-8_Part_2-199710151-00048PubMedGoogle ScholarCrossref
17.
Lawson  EH, Louie  R, Zingmond  DS,  et al.  Using both clinical registry and administrative claims data to measure risk-adjusted surgical outcomes.  Ann Surg. 2016;263(1):50-57. doi:10.1097/SLA.0000000000001031PubMedGoogle ScholarCrossref
18.
Maddox  TM, Chan  PS, Spertus  JA,  et al.  Variations in coronary artery disease secondary prevention prescriptions among outpatient cardiology practices: insights from the NCDR (National Cardiovascular Data Registry).  J Am Coll Cardiol. 2014;63(6):539-546. doi:10.1016/j.jacc.2013.09.053PubMedGoogle ScholarCrossref
19.
Newhouse  JP, Price  M, Huang  J, McWilliams  JM, Hsu  J.  Steps to reduce favorable risk selection in Medicare Advantage largely succeeded, boding well for health insurance exchanges.  Health Aff (Millwood). 2012;31(12):2618-2628. doi:10.1377/hlthaff.2012.0345PubMedGoogle ScholarCrossref
20.
Newhouse  JP, Price  M, McWilliams  JM, Hsu  J, McGuire  TG.  How much favorable selection is left in Medicare Advantage?  Am J Health Econ. 2015;1(1):1-26. doi:10.1162/AJHE_a_00001PubMedGoogle ScholarCrossref
21.
Patterson  ME, Hernandez  AF, Hammill  BG,  et al.  Process of care performance measures and long-term outcomes in patients hospitalized with heart failure.  Med Care. 2010;48(3):210-216. doi:10.1097/MLR.0b013e3181ca3eb4PubMedGoogle ScholarCrossref
22.
Bradley  EH, Herrin  J, Elbel  B,  et al.  Hospital quality for acute myocardial infarction: correlation among process measures and relationship with short-term mortality.  JAMA. 2006;296(1):72-78. doi:10.1001/jama.296.1.72PubMedGoogle ScholarCrossref
23.
Nguyen  QD, Peters  E, Wassef  A, Desmarais  P, Rémillard-Labrosse  D, Tremblay-Gravel  M.  Evolution of age and female representation in the most-cited randomized controlled trials of cardiology of the last 20 years.  Circ Cardiovasc Qual Outcomes. 2018;11(6):e004713. doi:10.1161/CIRCOUTCOMES.118.004713PubMedGoogle ScholarCrossref
24.
Medicare Payment Advisory Committee (MedPAC). Report to the Congress: Medicare payment policy. http://www.medpac.gov/docs/default-source/reports/mar17_entirereport.pdf. Accessed September 7, 2018.
×