Medicare coverage did not necessarily lead to increased diagnosis of chronic conditions.
ABSTRACT
Objectives: The near-universal access to Medicare coverage at age 65 years improves access to care. However, little is known about whether Medicare eligibility promotes the diagnosis of chronic diseases. We examined the effects of Medicare eligibility at age 65 years on the diagnosis of chronic conditions.
Study Design: Using data from the 2007-2019 Medical Expenditure Panel Survey, we employed a regression discontinuity design.
Methods: Our sample includes 43,620 individuals aged 59 to 71 years. Our primary outcomes were diagnoses of 19 chronic conditions. Using a regression discontinuity design, we exploited the discontinuity in eligibility for Medicare at age 65 years and compared individuals just before and after age 65 years.
Results: Medicare eligibility at age 65 years led to significant increases in having any coverage or Medicare coverage: 8.8 percentage points (95% CI, 8.4-9.2) and 78.1 percentage points (95% CI, 74.9-81.4), respectively. However, there were no or small changes in the diagnosis of chronic conditions at age 65 years. Specifically, there were no significant changes in the diagnoses of 17 chronic conditions, and the changes were minor in magnitude. Significant changes were observed only in the diagnosis of stroke and cancer, at –0.6 percentage points (95% CI, –1.0 to –0.2) and –1.7 percentage points (95% CI, –2.8 to –0.6), respectively.
Conclusions: Our findings suggest that Medicare coverage did not necessarily lead to increased diagnosis of chronic conditions. Further research is necessary to explore the underlying mechanisms behind this observation.
Am J Manag Care. 2024;30(2):96-100. https://doi.org/10.37765/ajmc.2024.89497
Takeaway Points
Health insurance coverage has the potential to improve health by increasing access to and use of care and reducing financial barriers to health care. In particular, Medicare is of great interest to policy makers because it is a significant public insurance program in the United States, providing coverage for approximately 60 million adults who either are older or have disabilities. Generally, individuals become eligible for Medicare at the age of 65 years. Evidence suggests that the near-universal access to Medicare coverage at age 65 years improves care access, increases health care utilization, and decreases out-of-pocket spending.1-6
However, little is known about the impact of Medicare eligibility at age 65 years on the diagnosis of chronic diseases. On one hand, Medicare coverage can help individuals who were previously uninsured to receive timely diagnosis of chronic diseases. Prior work shows that uninsured adults with cancer experience delayed diagnosis compared with insured adults with cancer.7 This is particularly relevant to older adults, as delayed diagnosis can lead to overall worse health outcomes, such as the progression of disease to advanced stages, complications, and reductions in quality of life.8 On the other hand, the effect of gaining Medicare coverage on the diagnosis of chronic diseases may not be significant. This is because some individuals may have already had access to health services that led to the diagnosis of chronic disease through other types of health insurance coverage such as private health insurance. Additionally, some individuals may experience difficulty transitioning to new insurance coverage.9,10 Many challenges can make this transition difficult, such as lack of understanding of the program, difficulty navigating the system, or language barriers. As a result, individuals who face difficulties transitioning health coverage may be less likely to receive a chronic disease diagnosis after Medicare enrollment. To fill this gap in the literature, we examined the effects of Medicare eligibility at age 65 years on the diagnoses of 19 chronic conditions.
METHODS
Data
We employed a repeated cross-sectional study design using data from the 2007-2019 Medical Expenditure Panel Survey (MEPS). The MEPS is a nationally representative survey of the US civilian noninstitutionalized population. The MEPS collects data from 2 main sources: the Household Component and the Medical Provider Component. The Household Component collects data from individual household members through questionnaires, and the Medical Provider Component collects data from a sample of providers who provided medical care to MEPS Household Component respondents. The Household Component data include demographic characteristics, health conditions, health insurance coverage, and health care utilization. The Medical Provider Component data include dates of visits/services, use of health care services, and diagnosis and procedure codes for medical visits/encounters. For our study, we merged 2 data sets from the MEPS: the full-year consolidated data files from the Household Component and the medical conditions files from the Medical Provider Component.
Sample
We first identified respondents aged 45 to 85 years from the MEPS. Using a data-driven method that automatically selected an optimal age bandwidth,11 we then selected a bandwidth of 6 around aged 65 years, the Medicare eligibility start age. Thus, our study sample includes individuals aged 59 to 71 years. Individuals aged 65 years were not included in the analysis because individuals become eligible for Medicare the first day of the month they turn 65, indicating that they might not be eligible for Medicare for a full year. Although some individuals were included in the data over the course of multiple years, we treated each year’s data as independent observations. This suggests that our sample is a repeated cross-sectional one, not a cohort.
Outcomes
Our primary outcomes were diagnoses of chronic conditions. We included 19 chronic conditions commonly encountered in clinical practice based on the criteria developed by a multiple chronic condition group within HHS’ Office of the Assistant Secretary for Health (arthritis, asthma, cancer, cardiac arrhythmias, chronic kidney disease, chronic obstructive pulmonary disease, congestive heart failure, coronary artery disease, dementia, depression, diabetes, hepatitis, HIV, hyperlipidemia, hypertension, osteoporosis, schizophrenia, stroke, and substance use disorders). The MEPS collects information on self-reported diagnoses, but these data can be biased by factors such as memory limitations. To address this, chronic conditions were identified using the International Classification of Diseases, Ninth Revision (ICD-9) and Tenth Revision (ICD-10) codes from the MEPS medical conditions files as part of the Medicare Provider Component. The medical conditions and procedures reported by the MEPS related to disease conditions were recorded by an interviewer as verbatim text and then coded according to ICD-9 and ICD-10. We also included health insurance coverage (any coverage and Medicare coverage) and health care utilization (the number of hospitalizations, outpatient visits, and emergency department visits) as secondary outcomes.
Covariates
We included the following individual-level characteristics from the survey as control variables: race and ethnicity, sex, employment status, marital status, education, family income, and Census region of residence.
Statistical Analysis
To evaluate the effect of Medicare eligibility, we used a regression discontinuity design, which exploits a change in Medicare eligibility at age 65 years and compares individuals aged 59 to 64 years with individuals aged 66 to 71 years. Regression discontinuity addresses selection bias in comparing individuals with and without an intervention by exploiting an arbitrary cutoff in program eligibility, which offers transparent visual evidence of changes in the outcome at the cutoff.12,13 The regression discontinuity approach relies on the assumption that individuals within a narrow bandwidth around the cutoff have similar observed and unobserved characteristics other than the likelihood of obtaining Medicare coverage. To observe a causal effect of Medicare eligibility, it is required to find a discontinuity in the outcome variable at the age cutoff. This suggests that some outcomes increase with age, but there may be no effects if there were no discontinuities at the age cutoff.
We first compared sample characteristics between individuals aged 59 to 64 years and aged 66 to 71 years. We then used a parametric regression discontinuity model with a quadratic age trend while adjusting for individual-level covariates.5,14 Our primary focus was on the coefficient of the age-65 dummy variable, which represents the immediate impact of transitioning into Medicare eligibility on the outcome of interest. We also plotted outcomes by age to visually inspect the discontinuities in the outcome at the age eligibility cutoff. To examine the robustness of our findings, we conducted a series of sensitivity analyses using different age bandwidths (5 and 7 years around age 65), study samples (including individuals aged 65 years, excluding Medicaid beneficiaries and excluding the uninsured), and alternative model specifications (accounting for a linear age trend and without adjusting for survey weights). Finally, we ran placebo tests using cutoffs above and below 65 years to see whether the results were unique to the 65-year cutoff.
For all analyses, we included year-fixed effects and clustered the SEs within individuals. We used survey weights to adjust sample characteristics to be representative of the US population.
RESULTS
We included a total of 43,620 individuals aged 59 to 71 years (Table). Demographic and socioeconomic characteristics were largely similar between individuals younger and older than 65 years except for employment status (eAppendix Table 1 [eAppendix available at ajmc.com]).
Medicare eligibility at age 65 years led to an increase of 8.8 percentage points (95% CI, 8.4-9.2) in the probability of having any insurance coverage and 78.1 percentage points (95% CI, 74.9-81.4) in the probability of having Medicare coverage (Table). However, there were no or small changes in the diagnosis of chronic conditions at age 65 years. Specifically, there were no significant changes in the diagnoses of 17 chronic conditions. Significant changes were observed only in the diagnosis of cancer and stroke (–1.7 percentage points [95% CI, –2.8 to –0.6] and –0.6 percentage points [95% CI, –1.0 to –0.2], respectively). This pattern was also found in our unadjusted analyses (Figure). There were no significant changes in health care utilization.
Although there were some changes in magnitude, our results from sensitivity analyses showed that there were no or small changes in the diagnosis of chronic conditions (eAppendix Table 2). Also, our placebo tests showed almost no significant changes in the diagnosis of chronic conditions at ages 63, 64, 66, and 67 years (eAppendix Table 3).
DISCUSSION
We found that the near-universal access to Medicare coverage at age 65 years resulted in significant increases in health insurance coverage but that there were no or small changes in chronic condition diagnosis among the entire population. This finding is consistent with prior research showing small changes in the diagnosis of chronic conditions among newly enrolled Medicaid beneficiaries.15 There may be multiple explanations for this observation. First, a significant number of adults had existing insurance coverage. Additionally, the Affordable Care Act expanded access to preventive care by requiring health insurance plans to cover a range of preventive services without any cost-sharing requirements.16 These could have led to early screening for and diagnosis of chronic conditions prior to Medicare eligibility.
Gaining Medicare coverage may lead to earlier diagnosis of chronic diseases but only among the uninsured. This is because the uninsured may be less likely to have a usual source of care and seek preventive care, which can lead to delays in diagnosis.7 Indeed, we found that the uninsured had lower chronic condition diagnoses than the privately insured before Medicare eligibility at age 65 years. However, estimating the causal effect of Medicare eligibility on chronic condition diagnosis among the uninsured was challenging with the current data. The MEPS provides longitudinal data, which allow us to track individuals who transition from being uninsured to having Medicare insurance. However, the sample size was small, raising potential concerns about the reliability of the findings.
Our findings suggest that Medicare coverage may play a limited role in the diagnosis of chronic conditions. However, because our findings only measure short-term effects, caution is needed when interpreting the results. It is possible that the impact of Medicare coverage on the diagnosis of chronic conditions may vary over the long term because patients may require time to establish a relationship with a new health care provider and access the necessary resources to manage their conditions effectively. Because the prevalence of chronic conditions tends to increase with age, it remains critical to prioritize timely detection as well as effective management of these conditions. Thus, CMS should prioritize investment in primary care to enhance its capacity to deliver timely and high-quality care for those with chronic diseases.17 This strategy is critical not only for improving health outcomes but also for ensuring the sustainability of the Medicare program in the long run.
Limitations
Our study has several limitations. First, the validity of the regression discontinuity counts on several assumptions. The study design assumed that the outcome variables of interest would evolve smoothly with age in the absence of Medicare eligibility at age 65 years. However, life changes such as retirement may have different impacts on health status. Additionally, it is assumed that individuals cannot manipulate Medicare eligibility at age 65 years. Individuals, however, may react in anticipation of gaining Medicare coverage by delaying health care use, although very limited evidence shows delayed care until age 65 years.18 Third, our sample is a repeated cross-sectional sample, meaning that it is possible to observe changes in the sample composition over time. However, this is unlikely to affect our findings unless there was a substantial change in the population at age 65 years. Fourth, we did not examine all chronic conditions or multimorbidities, and thus our findings may not generalize to individuals with multiple conditions or rare illnesses. Therefore, further research is needed to confirm our findings. Finally, the Medicare program has different payment systems from other programs, which may lead to substantial differences in practice patterns. These differences could influence diagnosis of chronic conditions. Additionally, there may be substantial heterogeneity within the Medicare program, such as Medicare Advantage or accountable care organizations.
CONCLUSIONS
In this repeated cross-sectional study using a regression discontinuity design, we showed that the near-universal access to Medicare coverage at age 65 years resulted in increases in health insurance coverage but that there were no or limited changes in the diagnosis of chronic conditions among the entire population. These findings suggest that Medicare coverage may have a limited impact on the short-term effects of chronic condition diagnosis.
Author Affiliations: Department of Health Policy and Management, College of Health Science, and BK21 FOUR R&E Center for Learning Health Systems, Korea University (SP), Seoul, Republic of Korea; Division of General Internal Medicine, Massachusetts General Hospital (FOM), Boston, MA.
Source of Funding: This work was partly supported by a National Research Foundation of Korea grant funded by the Korean government (MSIT) (No. RS-2023-00219289).
Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (SP); acquisition of data (SP); analysis and interpretation of data (SP, FOM); drafting of the manuscript (SP, FOM); critical revision of the manuscript for important intellectual content (SP, FOM); statistical analysis (SP); and supervision (SP).
Address Correspondence to: Sungchul Park, PhD, Department of Health Policy and Management, College of Health Science, and BK21 FOUR R&E Center for Learning Health Systems, Korea University, 145 Anam-ro, Seongbuk-gu, Seoul, Republic of Korea, 02841. Email: sungchul_park@korea.ac.kr.
REFERENCES
1. Card D, Dobkin C, Maestas N. The impact of nearly universal insurance coverage on health care utilization: evidence from Medicare. Am Econ Rev. 2008;98(5):2242-2258. doi:10.1257/aer.98.5.2242
2. Scott JW, Neiman PU, Tipirneni R, Fan Z, Ayanian JZ. Changes in out-of-pocket spending and catastrophic health care expenditures associated with Medicare eligibility. JAMA Health Forum. 2021;2(9):e212531. doi:10.1001/jamahealthforum.2021.2531
3. Aggarwal R, Yeh RW, Dahabreh IJ, Robertson SE, Wadhera RK. Medicare eligibility and healthcare access, affordability, and financial strain for low- and higher-income adults in the United States: a regression discontinuity analysis. PLoS Med. 2022;19(10):e1004083. doi:10.1371/journal.pmed.1004083
4. Card D, Dobkin C, Maestas N. Does Medicare save lives? Q J Econ. 2009;124(2):597-636. doi:10.1162/qjec.2009.124.2.597
5. Park S, Ortega AN, Chen J, Bustamante AV. Effects of Medicare eligibility and enrollment at age 65 among immigrants and US-born residents. J Am Geriatr Soc. 2023;71(9):2845-2854. doi:10.1111/jgs.18380
6. Wallace J, Jiang K, Goldsmith-Pinkham P, Song Z. Changes in racial and ethnic disparities in access to care and health among US adults at age 65 years. JAMA Intern Med. 2021;181(9):1207-1215. doi:10.1001/jamainternmed.2021.3922
7. Zhao J, Han X, Nogueira L, et al. Health insurance status and cancer stage at diagnosis and survival in the United States. CA Cancer J Clin. 2022;72(6):542-560. doi:10.3322/caac.21732
8. Weissman JS, Stern R, Fielding SL, Epstein AM. Delayed access to health care: risk factors, reasons, and consequences. Ann Intern Med. 1991;114(4):325-331. doi:10.7326/0003-4819-114-4-325
9. Sommers BD, Gourevitch R, Maylone B, Blendon RJ, Epstein AM. Insurance churning rates for low-income adults under health reform: lower than expected but still harmful for many. Health Aff (Millwood). 2016;35(10):1816-1824. doi:10.1377/hlthaff.2016.0455
10. McIntyre A, Shepard M. Automatic insurance policies — important tools for preventing coverage loss. N Engl J Med. 2022;386(5):408-411. doi:10.1056/NEJMp2114189
11. Imbens G, Kalyanaraman K. Optimal bandwidth choice for the regression discontinuity estimator. Rev Econ Stud. 2011;79(3):933-959. doi:10.1093/restud/rdr043
12. Lee DS, Lemieux T. Regression discontinuity designs in economics. J Econ Lit. 2010;48(2):281-355. doi:10.1257/jel.48.2.281
13. Venkataramani AS, Bor J, Jena AB. Regression discontinuity designs in healthcare research. BMJ. 2016;352:i1216. doi:10.1136/bmj.i1216
14. Park S, Wadhera RK, Jung J. Effects of Medicare eligibility and enrollment at age 65 years on the use of high-value and low-value care. Health Serv Res. 2023;58(1):174-185. doi:10.1111/1475-6773.14065
15. Baicker K, Taubman SL, Allen HL, et al; Oregon Health Study Group. The Oregon Experiment — effects of Medicaid on clinical outcomes. N Engl J Med. 2013;368(18):1713-1722. doi:10.1056/NEJMsa1212321
16. Fox JB, Shaw FE. Clinical preventive services coverage and the Affordable Care Act. Am J Public Health. 2015;105(1):e7-e10. doi:10.2105/AJPH.2014.302289
17. Chang CH, Stukel TA, Flood AB, Goodman DC. Primary care physician workforce and Medicare beneficiaries’ health outcomes. JAMA. 2011;305(20):2096-2104. doi:10.1001/jama.2011.665
18. Barcellos SH, Jacobson M. The effects of Medicare on medical expenditure risk and financial strain. Am Econ J Econ Policy. 2015;7(4):41-70. doi:10.1257/pol.20140262
Despite Record ACA Enrollment, Report Reveals Underinsured Americans Are in Crisis
November 21st 2024Despite significant progress in expanding health insurance coverage since the Affordable Care Act (ACA) was enacted, millions of Americans still face critical gaps in access to and affordability of health care.
Read More
Exploring Racial, Ethnic Disparities in Cancer Care Prior Authorization Decisions
October 24th 2024On this episode of Managed Care Cast, we're talking with the author of a study published in the October 2024 issue of The American Journal of Managed Care® that explored prior authorization decisions in cancer care by race and ethnicity for commercially insured patients.
Listen
How English- and Spanish-Preferring Patients With Cancer Decide on Emergency Care
November 13th 2024Care delivery innovations to help patients with cancer avoid emergency department visits are underused. The authors interviewed English- and Spanish-preferring patients at 2 diverse health systems to understand why.
Read More
Frameworks for Advancing Health Equity: Community Wellness Centers
September 27th 2024Delia Orosco, MS, director of Community Wellness Centers at Inland Empire Health Plan, shares insight into innovative initiatives provided by the wellness centers and their new mobile mammogram clinics.
Listen
Geographic Variations and Facility Determinants of Acute Care Utilization and Spending for ACSCs
November 12th 2024Emergency department (ED) visits and hospitalizations for ambulatory care–sensitive conditions (ACSCs) among Medicaid patients constitute almost 40% of all ED visits and hospitalizations, with lower rates observed in areas with greater proximity to urgent care facilities and density of rural health clinics.
Read More
Pervasiveness and Clinical Staff Perceptions of HPV Vaccination Feedback
November 11th 2024This article used regression analyses to quantify how clinical staff perceive provider feedback to improve human papillomavirus (HPV) vaccination rates and determine the prevalence of such feedback.
Read More