Recent Cohorts Aging Into Medicare Use More Counseling and Psychotherapy Than Past Cohorts

ABSTRACT

Objectives: Despite recent policy interest in improving access to mental health care in Medicare, little is known about how demand for care will change among the Medicare population as newer cohorts age into the program. We documented the growing rate of counseling and psychotherapy use in the decade prior to turning age 65 years among subsequent cohorts aging into Medicare. We characterized how this growth varied across demographic groups, income levels, and mental and physical health status.

Study Design: We present trends using data from the 2002-2017 Medical Expenditure Panel Survey Household Component.

Methods: We categorized individuals into 5-year Medicare entry cohorts based on the year they turned age 65 years. Our outcome was an indicator for having a visit for counseling or psychotherapy in a given year. Employing a probit regression, we characterized visit rates across 5-year cohorts, presenting both unadjusted and covariate-adjusted results. We ran stratified regressions by subpopulations.

Results: Our sample included 54,666 individuals aged 55 to 64 years, weighted to be nationally representative. The cohort aging into Medicare between 2021 and 2025 was 88% (95% CI, 57%-119%) more likely to have a counseling or psychotherapy visit between the ages of 55 and 64 years compared with the cohort that gained eligibility for Medicare between 2006 and 2010 at the same age. Growth in utilization was pervasive across many subpopulations.

Conclusions: Our findings suggest that more recent cohorts aging into Medicare seek significantly more counseling and psychotherapy than prior cohorts. This increased utilization is pervasive across subpopulations, suggesting that plans must prepare to accommodate the needs of new Medicare entrants.

Am J Manag Care. 2024;30(11):e337-e340. https://doi.org/10.37765/ajmc.2024.89631

Takeaway Points

• In the decade prior to aging into Medicare, recent cohorts entering the program utilized counseling and psychotherapy services at higher rates compared with prior cohorts at the same age.
• Increased counseling and psychotherapy utilization among more recent cohorts was not concentrated in a specific subpopulation of sex, race and ethnicity, income, education level, or physical or mental health status.
• These findings suggest that demand for mental health care among Medicare beneficiaries may increase in the coming years, meaning that increased investment in mental health care among traditional Medicare and Medicare Advantage plans may be necessary.

Although 20% to 30% of adults 65 years or older report symptoms of anxiety or depression,¹ recent estimates suggest less than a quarter of Medicare beneficiaries with mental health symptoms receive treatment.² This low level of utilization may be in part due to structural access barriers in the Medicare program, such as low mental health provider participation in traditional Medicare (TM)³ and narrow mental health provider networks in Medicare Advantage (MA).⁴ Recent policy efforts have aimed to facilitate mental health care access, especially for counseling services. As of 2024, the Mental Health Access Improvement Act allows marriage and family therapists and mental health counselors to be reimbursed under TM.⁵ In MA, CMS has proposed new network adequacy requirements for outpatient behavioral health facilities.⁶

As regulators and insurers consider mental health care access policies, attention must be paid to how these policies may affect current Medicare enrollees and future cohorts entering the program. Over the past 2 decades, mental health care utilization has increased among the working-age population⁷ and public perceptions about some mental health conditions have changed.⁸ Consequently, we may expect demand for mental health care services among Medicare beneficiaries to increase in the coming years, potentially outpacing the current service capacity.

Little is known about how demand for counseling services has changed among cohorts entering Medicare over time or how these changes vary by demographic, socioeconomic, or health status. This study describes utilization of counseling and psychotherapy services among subsequent cohorts aging into Medicare over 2 decades and explores utilization trends across subpopulations.

METHODS

Data and Sample

The study used data from the Medical Expenditure Panel Survey Household Component (MEPS-HC), a nationally representative household survey of the US civilian noninstitutionalized population.⁹ For more details on the MEPS-HC, see eAppendix Section 1 (eAppendix available at ajmc.com).

We restricted our analysis to survey years 2002 to 2017. This period ends before survey redesign in 2018, allowing for consistent coding of utilization measures. See eAppendix Section 1 for more details. We further restricted to individuals aged 55 to 64 years when interviewed and whose 65th birthday occurs between 2006 and 2025.

Measure of Counseling and Medicare Entry Cohort

Our outcome is an indicator for whether an individual had a psychotherapy or mental health counseling visit in an office or outpatient setting in a given calendar year. In eAppendix Section 2, we describe the creation of this variable and its advantages relative to other measures of mental health care utilization for this study.

We identified the year that each individual turned 65 by birth year. We categorized our sample into 5-year Medicare entry cohorts: individuals turning 65 in 2006-2010, 2011-2015, 2016-2020, and 2021-2025.

Analysis

We fit a probit regression model for whether an individual had a counseling or psychotherapy visit within the year as the outcome. Our primary independent variables were indicators for whether an individual was in a given 5-year Medicare entry cohort. We calculated marginal probabilities of having a counseling or psychotherapy visit for each cohort. We used linear regression to formally test for a positive relationship between an individual’s year turning 65 and their likelihood of having a counseling or psychotherapy visit.

To better understand how utilization changed within subpopulations, we repeated our Medicare entry cohort analysis separately by sex, race and ethnicity, income, education, and reported mental and physical health. For each subpopulation, we also used linear regression to formally test for a positive relationship between the year an individual turns 65 and their likelihood of counseling or psychotherapy use in eAppendix exercises.

By pooling our analysis across multiple years of data, our main, unadjusted 5-year cohort-based approach followed guidance in MEPS-HC documentation on analyzing trends in outcomes.¹⁰ As sensitivity analyses, we followed past longitudinal analyses of the MEPS-HC in employing rich controls to mitigate potential confounding influences of survey redesigns.¹¹ For details, see eAppendix Section 3. Analyses were conducted using Stata 17.0 (StataCorp). This study was deemed not human subjects research and exempt from review by the University of Southern California institutional review board.

Study Variables

We measured age as of the end of the survey year. MEPS-HC reports household income as a percentage of the federal poverty level (FPL), categorized as less than 100%, 100% to 125%, 125% to 200%, 200% to 400%, and 400% or greater. We used 4 categories of self-reported race and ethnicity: Hispanic, non-Hispanic White (hereafter, “White”), non-Hispanic Black (hereafter, “Black”), and a remainder category of smaller racial groups. Education was characterized using the highest degree obtained: no degree, General Educational Development or high school, bachelor’s, or other degree. MEPS-HC reports individuals’ residential census regions: Northwest, Midwest, South, and West. We characterized an individual as employed if they reported either having a job or having a job to return to. An individual was defined as having health insurance if they reported having insurance within the survey year. We used mental and physical health as perceived by the household respondent. We provide greater detail in eAppendix Section 2 and average characteristics in eAppendix Tables 1 and 2 among different Medicare entry cohorts.

RESULTS

The share of individuals aged 55 to 64 years who had a counseling or psychotherapy visit each year increased with each successive Medicare entry cohort (Figure 1). The annual prevalence of receiving counseling or psychotherapy nearly doubled from 3.0% among the cohort that entered Medicare in 2006-2010 to 5.6% among those who entered in 2021-2025.

This upward trend in counseling and psychotherapy remained when including age and year fixed effects, demographic controls, socioeconomic controls, insurance status, self-reported mental health status, and smoking status in a linear model of utilization (eAppendix Table 3), although cohort-specific estimates become less precise (eAppendix Figure 1). After employing our full set of controls, each subsequent year an individual turned 65 was associated with a statistically significant 2.3% (95% CI, 0.2%-4.3%) increase in utilization relative to those who turned 65 in 2006.

This increase in counseling and psychotherapy use among cohorts was not concentrated in a single subpopulation (Figure 2). We observed increases in counseling across sex, income, and education groups and among both Black and White survey respondents. Similarly, recent cohorts reporting almost all categories of both mental and physical health exhibited increased utilization compared with past cohorts with similar reported health.

Although most subpopulations exhibited growth in counseling and psychotherapy use, our unadjusted analysis suggests growth was particularly large among those with lower income, Black and White individuals, and those with worse physical and mental health (eAppendix Figures 2 and 3). Although the inclusion of controls largely eliminates differences, increases in utilization were still particularly high among White and Black respondents and those with household incomes below 100% FPL (eAppendix Figures 4, 5, and 6).

DISCUSSION

Utilization of counseling or psychotherapy in the decade prior to turning 65 has increased among subsequent cohorts entering Medicare. The annual prevalence of individuals aged 55 to 64 years receiving counseling or psychotherapy nearly doubled from 3.0% to 5.6% among the 2006-2010 and 2021-2025 Medicare entry cohorts, respectively. Among the nearly 4.1 million individuals poised to turn 65 in 2024,12 our estimates suggest 221,400 used counseling or psychotherapy annually prior to age 65 years. By contrast, if utilization rates among these current entrants were the same as the 2006-2010 entry cohort, only 123,000 of the 4.1 million individuals would have utilized care prior to age 65 years. (For more details on this accounting exercise, see eAppendix Section 4.)

Utilization increased across several racial and ethnic groups and education, income, and physical health levels, although with the addition of controls, growth was uniquely robust among Black and White individuals as well as those earning less than 100% FPL. This growth suggests that both TM and MA plans may observe increased demand for counseling services as younger cohorts age into the program.

Only a fraction of current Medicare enrollees with mental health symptoms receive treatment, and utilization is particularly low among MA enrollees compared with similar TM enrollees.² Our findings highlight the importance of addressing mental health care access barriers, such as limited provider participation in TM³ and narrow mental health networks,⁴ low provider prices,¹³ and frequent prior authorization for mental health services in MA.¹⁴

Policy movement to reduce barriers to counseling services has been relatively slow and subject to structural challenges.¹⁵ In 2010, TM and MA began phasing in payment parity for mental health care, and in 2024, CMS expanded the set of mental health providers who can be compensated under Medicare. Although these measures may ease current access issues for some enrollees, they still may be insufficient to accommodate the increasing demand for mental health care among newer cohorts of Medicare enrollees. Moreover, there have been increases in mental health symptoms for older adults during the COVID-19 pandemic.¹⁶

Limitations

Our study has several limitations. First, the MEPS-HC survey design underwent changes, particularly in 2008 and 2013. We followed MEPS guidance for dealing with these changes by pooling estimates over multiple years and testing whether trends were robust to the inclusion of survey year fixed effects and controls.^10,11 Relatedly, our analysis was based on data prior to the COVID-19 pandemic and was thus unable to account for any resulting persistent changes in utilization. Third, the MEPS-HC has a relatively small sample for detecting differences across cohorts, especially within subpopulations. Finally, we could not directly observe respondents’ clinical need for counseling, and thus we cannot contextualize whether utilization levels were appropriate given population need nor the degree to which cross-cohort trends were driven by changes in care-seeking behavior among those needing counseling vs changes in the prevalence of individuals with counseling needs.

CONCLUSIONS

Utilization of counseling or psychotherapy services in the decade prior to entering Medicare has increased among subsequent cohorts aging into the program. This increase in utilization is ubiquitous across subpopulations and suggests that TM and most MA plans can expect to see increased demand for services in the coming years.

Acknowledgments

The authors would like to thank Josephine Rohrer for her research assistance.

Author Affiliations: Leonard D. Schaeffer Center for Health Policy & Economics, University of Southern California (GM, ELD), Los Angeles, CA; Lafayette College (AB), Easton, PA.

Source of Funding: Internal institutional funding.

Author Disclosures: Dr Duffy has provided expert testimony on matters in the hospital and health insurance sectors. Dr McCormack and Dr Biener 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 (GM, ELD, AB); analysis and interpretation of data (GM, AB); drafting of the manuscript (GM, AB); critical revision of the manuscript for important intellectual content (GM, ELD, AB); statistical analysis (GM, ELD).

Address Correspondence to: Grace McCormack, PhD, Leonard D. Schaeffer Center for Health Policy & Economics, University of Southern California, 635 Downey Way, Verna & Peter Dauterive Hall (VPD), Los Angeles, CA 90089. Email: gracemcc@usc.edu.

REFERENCES

1. Panchal N, Saunders H, Rudowitz R, Cox C. The implications of COVID-19 for mental health and substance use. KFF. March 20, 2023. Accessed October 17, 2024. https://www.kff.org/mental-health/issue-brief/the-implications-of-covid-19-for-mental-health-and-substance-use/

2. Park S, Meyers DJ, Jimenez DE, Gualdrón N, Cook BL. Health care spending, use, and financial hardship among traditional Medicare and Medicare Advantage enrollees with mental health symptoms. Am J Geriatr Psychiatry. 2024;32(6):739-750. doi:10.1016/j.jagp.2024.01.014

3. Oh S, McDowell A, Benson NM, Cook BL, Fung V. Trends in participation in Medicare among psychiatrists and psychiatric mental health nurse practitioners, 2013-2019. JAMA Netw Open. 2022;5(7):e2224368. doi:10.1001/jamanetworkopen.2022.24368

4. Zhu JM, Meiselbach MK, Drake C, Polsky D. Psychiatrist networks in Medicare Advantage plans are substantially narrower than in Medicaid and ACA markets. Health Aff (Millwood). 2023;42(7):909-918. doi:10.1377/hlthaff.2022.01547

5. Mental Health Access Improvement Act of 2021, HR 432, 117th Cong (2021). Accessed October 17, 2024. https://www.congress.gov/bill/117th-congress/house-bill/432/text

6. Seshamani M, Jacobs D. Important new changes to improve access to behavioral health in Medicare. CMS. November 6, 2023. Accessed October 17, 2024. https://www.cms.gov/blog/important-new-changes-improve-access-behavioral-health-medicare-0

7. Wang J, Qiu Y, Zhu X. Trends of mental health care utilization among US adults from 1999 to 2018. BMC Psychiatry. 2023;23(1):665. doi:10.1186/s12888-023-05156-2

8. Pescosolido BA, Halpern-Manners A, Luo L, Perry B. Trends in public stigma of mental illness in the US, 1996-2018. JAMA Netw Open. 2021;4(12):e2140202. doi:10.1001/jamanetworkopen.2021.40202

9. MEPS-HC panel design and data collection process. Agency for Healthcare Research and Quality. Accessed February 25, 2024. https://meps.ahrq.gov/survey_comp/hc_data_collection.jsp

10. MEPS HC-209 2018 Full Year Consolidated Data File. Agency for Healthcare Research and Quality; August 2020. Accessed February 25, 2024. https://meps.ahrq.gov/data_stats/download_data/pufs/h209/h209doc.pdf

11. Biener AI, Zuvekas SH, Hill SC. Impact of recent Medicaid expansions on office-based primary care and specialty care among the newly eligible. Health Serv Res. 2018;53(4):2426-2445. doi:10.1111/1475-6773.12793

12. Ansberry C. America has never had so many 65-year-olds: they’re redefining the milestone. Wall Street Journal. February 6, 2024. Accessed March 21, 2024. https://www.wsj.com/health/america-has-never-had-so-many-65-year-olds-theyre-redefining-the-milestone-4383e769

13. Pelech D, Hayford T. Medicare Advantage and commercial prices for mental health services. Health Aff (Millwood). 2019;38(2):262-267. doi:10.1377/hlthaff.2018.05226

14. Freed M, Sroczynski N, Neuman T. Mental health and substance use disorder coverage in Medicare Advantage plans. KFF. April 28, 2023. Accessed February 27, 2024. https://www.kff.org/mental-health/issue-brief/mental-health-and-substance-use-disorder-coverage-in-medicare-advantage-plans/

15. McCormack G, Meiselbach M, Rohrer J. Medicare’s mental health care problem. Health Affairs Forefront. March 4, 2024. Accessed October 17, 2024. https://www.healthaffairs.org/content/forefront/medicare-s-mental-health-care-problem

16. Bafail DA. Mental health issues associated with COVID-19 among the elderly population: a narrative review. Cureus. 2022;14(12):e33081. doi:10.7759/cureus.33081