- Issue: Access to health care, use of services, and patient outcomes can be complicated by many medical and nonmedical factors. People facing complex challenges such as mental illness, housing insecurity, or substance use, however, are not a homogeneous group; different individuals have different needs.
- Goals: To understand the needs of people with very low income — no more than 75 percent of the federal poverty level — who enrolled in Medicaid under Minnesota’s expansion of the program prior to the Affordable Care Act.
- Methods: The authors analyzed data on nondisabled, childless adults in the Minneapolis–St. Paul region who enrolled in Medicaid between 2011 and 2013.
- Findings and Conclusions: Early Medicaid expansion enrollees in urban Minnesota were largely nonwhite, male, and unmarried and had low educational attainment. In this very poor population, rates of homelessness, substance use, and mental illness were very high. More than 25 percent of adults dealt with two or more of these challenges, while 10 percent experienced all three. Providing access to a range of highly integrated health and social services may be the best way to help these individuals.
Federal and state expansion of eligibility for Medicaid to formerly ineligible groups, including childless adults, has brought health coverage to a variety of historically uninsured individuals. Many are homeless, have substance use disorders, mental illnesses, or have been in prison, and many have very complex medical and social needs as a result.1
Once insured, new enrollees tend to have pent-up demand for services.2 Many also “churn” on and off Medicaid, causing interruptions in medical care, lowering adherence to prescribed medications, increasing emergency department use, and worsening self-reported quality of care and health.3
This issue brief focuses on the medical and social needs of very-low-income Medicaid beneficiaries in Minnesota, which was among a handful of states that expanded Medicaid before the Affordable Care Act’s (ACA) expansion of the program in 2014. Starting in 2011, Minnesota opened its Medicaid program to childless adults (ages 21 to 64) with incomes at or below 75 percent of the federal poverty level (in 2017, $12,060 for an individual). By contrast, the ACA expanded eligibility to those with incomes up to 138 percent of the federal poverty level.
Prior work comparing Minnesota’s early expansion population to traditional beneficiaries of Medicaid or the Children’s Health Insurance Program found marked differences between the groups, particularly in terms of their social and behavioral health needs.4 Specifically, early expansion enrollees had a much higher rate of serious mental illness (27% versus 19%), often co-occurring with unstable housing (11% for early expansion enrollees).
Findings: Very Poor Medicaid Beneficiaries Have Complex Medical and Social Needs
To characterize the needs of early Medicaid expansion enrollees in Minnesota, we used enrollment and claims data for those in the Minneapolis–St. Paul metropolitan region, which has the largest number of such enrollees. We analyzed the early expansion population’s demographic characteristics, health conditions, and social and behavioral factors that could complicate their health care and outcomes.
Under Minnesota’s early expansion, about 39,000 individuals enrolled each year from 2011–2013. By contrast, about 43,000 individuals enrolled in 2014 under the ACA expansion. Early expansion enrollees were enrolled for less time: a median 6.8 months in a 12-month period (or 57% of the year), compared with a median of 8.3 months (70% of the year) among later enrollees (Exhibit 1).5
Early enrollees were mainly nonwhite (47%) and male (60%). Nearly one-third (32%) had less than a high school education, and nearly 12 percent preferred a language other than English (Exhibit 2).
Based on medical claims, early enrollees had moderate rates of chronic medical conditions (Exhibit 3). About 10 percent received services for asthma, nearly 13 percent received services for diabetes, 7 percent did so for chronic obstructive pulmonary disease, and nearly 20 percent did so for hypertension.
By comparison, early enrollees had very high rates of behavioral health conditions (Exhibit 4). Thirty-seven percent were diagnosed with anxiety, mood, or schizophrenic disorders (with mood disorders, including depressive and bipolar disorders, the most common). More than 25 percent were diagnosed with substance use disorders (roughly equally divided between alcohol and non-alcohol disorders). Nearly one-fifth were diagnosed with both mental health problems and substance use disorders.
Twenty-eight percent of early expansion enrollees appeared to be homeless, based on addresses they listed at time of Medicaid enrollment.6
Many early enrollees had concurrent problems of unstable housing, major mental illness, and substance use disorder. Over half had at least one of these problems (56%). About 10 percent faced all three challenges, and about 9 percent experienced both substance use disorder and mental illness. Sizable percentages faced homelessness only (11%), substance use only (4%), or mental illness only (15%), without the other problems identified in claims (Exhibit 5).
Implications for Policy and Research
Based on our analysis, very poor Medicaid enrollees have high rates of social and behavioral health needs. The scaling back of coverage or benefits for this population, as could occur under some proposed policy changes, would likely make it harder to meet their needs, particularly those related to mental health conditions and substance use disorders.7 Losing coverage would also make it harder for these adults to manage their chronic physical conditions, especially if they were unstably housed. Even with stable Medicaid coverage, very poor adults would benefit from tailored approaches that strive to meet their overlapping social, physical, and behavioral health needs.
To better serve these adults, health care payment and delivery models should make care more accessible and integrate physical health, behavioral health, and social services. Such efforts could also help reduce confusion about which services are covered and help people make or keep appointments.8 Moreover, integrating preventive and chronic medical care with case management, supportive housing, probation services, and other services could increase rates of early detection and improve management of physical and behavioral health conditions. Doing so would likely entail integration of data systems and eligibility. Comprehensive integration is therefore also likely to reduce duplicate services offered by case managers and health care coordinators, as has begun to happen under accountable care arrangements.9 Such efforts require stable payment that offers flexibility to invest in the staff and support services that could produce long-term health improvements and reduce costs.
In addition, efforts are needed to reduce unnecessary disenrollment. Such efforts may result in cost savings for insurers and improved health for enrollees.10 Very-low-income Medicaid expansion enrollees face a variety of barriers, therefore simplified eligibility and renewal processes are needed to improve their coverage and reduce barriers to care.
Helping very-low-income enrollees maintain coverage and better integrating their social and medical services could support improved population health surveillance and data collection. This could be done by linking data across housing supports, Medicaid and other state insurance programs, social services, criminal justice, and clinical medical and behavioral health.11 Doing so would vastly improve policymakers’ and researchers’ ability to understand and help very-low-income Medicaid enrollees. As highlighted in this report, churn in insurance enrollment results in information loss by removing the most at-risk people from observable data. Linking data sources across different health and social sectors could thus help identify individuals eligible for Medicaid, help them retain continuous coverage, and provide policymakers with actionable evidence.
How We Conducted This Study
This study employed analysis of enrollment and claims data from Minnesota’s Medicaid expansion population in Hennepin and Ramsey counties. Data analyses involved using Stata 12.1 to summarize enrollment and claims information about enrollees. The cohort consisted of anyone with more than one month of expansion eligibility in the seven-county metropolitan region of Minnesota surrounding Minneapolis and St. Paul from 2011–2014 (including the early and later Medicaid expansion). Tabulations and summary analyses were used without adjustment to produce descriptive statistics.
1. Barbara DiPietro and Lisa Klingenmaier, “Achieving Public Health Goals Through Medicaid Expansion: Opportunities in Criminal Justice, Homelessness, and Behavioral Health with the Patient Protection and Affordable Care Act,” American Journal of Public Health, published online Nov. 20, 2013, https://doi.org/10.2105/AJPH.2013.301497; Tami L. Mark et al., “National Estimates of Behavioral Health Conditions and Their Treatment Among Adults Newly Insured Under the ACA,” Psychiatric Services 66, no. 4 (Apr. 2015): 426–29, https://doi.org/10.1176/appi.ps.201400078; and Sachini N. Bandara et al., “Leveraging the Affordable Care Act to Enroll Justice-Involved Populations in Medicaid: State and Local Efforts,” Health Affairs 34, no. 12 (Dec. 2015): 2044–51, https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2015.0668.
2. Thomas DeLeire et al., “Wisconsin Experience Indicates That Expanding Public Insurance to Low-Income Childless Adults Has Health Care Impacts,” Health Affairs32, no. 6 (June 2013): 1037–45, https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2012.1026; and Sarah L. Taubman et al., “Medicaid Increases Emergency-Department Use: Evidence from Oregon’s Health Insurance Experiment,” Science 343, no. 6168 (Jan. 17, 2014): 263–68, https://doi.org/10.1126/science.1246183.
3. Benjamin D. Sommers et al., “Insurance Churning Rates for Low-Income Adults Under Health Reform: Lower Than Expected but Still Harmful for Many,” Health Affairs 35, no. 10 (Oct. 2016): 1816–24, http://www.commonwealthfund.org/publications/in-the-literature/2016/oct/insurance-churning-low-income-adults.
4. Katherine D. Vickery et al., “Medicaid Expansion and Mental Health: A Minnesota Case Study,” Families, Systems, & Health 34, no. 1 (Mar. 2016): 58–63, http://psycnet.apa.org/doiLanding?doi=10.1037%2Ffsh0000186.
5. Differences in share of months enrolled (versus not enrolled) per year, along with incomplete demographic data because of an initial lack of communication between the state health exchange and Medicaid data systems, complicate further comparisons between early and ACA expansion enrollees.
7. At the invitation of the Secretary of the U.S. Department of Health and Human Services in 2017 (https://www.hhs.gov/sites/default/files/sec-price-admin-verma-ltr.pdf), multiple states have sought to use federal waivers to limit access to Medicaid for certain low-income adults. This includes states which expanded Medicaid under the ACA (e.g., Arkansas and Massachusetts). See Jessica Schubel, Proposals to Lower Medicaid Expansion Eligibility Jeopardize Coverage for Low-Income Adults(Center on Budget and Policy Priorities, Dec. 19, 2017), https://www.cbpp.org/research/health/proposals-to-lower-medicaid-expansion-eligibility-jeopardize-coverage-for-low-income. Recent budget proposals, including Graham-Cassidy and presidential budget proposals, also have sought to limit or eliminate the ACA Medicaid expansion in favor of alternatives such as block grants. See Peggy Bailey et al., Health Proposals in President’s Budget Would Reduce Health Insurance Coverage and Access to Care (Center on Budget and Policy Priorities, Feb. 16, 2018), https://www.cbpp.org/research/health/health-proposals-in-presidents-budget-would-reduce-health-insurance-coverage-and.
8. Nathan D. Shippee et al., “An Observational Study of Emergency Department Utilization Among Enrollees of Minnesota Health Care Programs: Financial and Non-Financial Barriers Have Different Associations,” BMC Health Services Research 14 (Feb. 8, 2014): 62, https://doi.org/10.1186/1472-6963-14-62.
9. Shana F. Sandberg et al., “Hennepin Health: A Safety-Net Accountable Care Organization for the Expanded Medicaid Population,” Health Affairs 33, no. 11 (Nov. 2014): 1975–84, https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2014.0648; and Renuka Tipirneni, Katherine D. Vickery, and Edward P. Ehlinger, “Accountable Communities for Health: Moving from Providing Accountable Care to Creating Health,” Annals of Family Medicine 13, no. 4 (2015): 367–69, http://www.annfammed.org/content/13/4/367.full.
10. Ritesh Banerjee, Jeanette Y. Ziegenfuss, and Nilay D. Shah, “Impact of Discontinuity in Health Insurance on Resource Utilization,” BMC Health Services Research10 (July 6, 2010): 195, https://doi.org/10.1186/1472-6963-10-195.
11. Katherine D. Vickery et al., “Cross-Sector Service Use Among High Health Care Utilizers in Minnesota After Medicaid Expansion,” Health Affairs 37, no. 1 (Jan. 2018): 62–69, https://www.healthaffairs.org/doi/10.1377/hlthaff.2017.0991.