Effects of the Affordable Care Act on Private Insurance Coverage and Treatment of Behavioral Health Conditions in Young Adults

Abstract

Objectives. To assess changes in private insurance coverage and behavioral treatment of people aged 19 to 25 years and 26 to 35 years following the 2010 Affordable Care Act (ACA) dependent care provision and 2014 insurance reforms.

Methods. We extracted data from the 2008 to 2016 US National Surveys on Drug Use and Health. We used an adjusted difference-in-differences approach to assess effects of age group on change in coverage. Replications in subgroups with serious psychological distress or substance use disorders also assessed change in mental health and substance use treatment.

Results. Between 2008 to 2010 and 2011 to 2013, the increase in coverage of the younger group significantly differed from the decrease in the older group, but the 2 groups did not significantly differ between 2011 to 2013 and 2014 to 2016 (+3.2 percentage points; P < .001; and +3.8 percentage points; P < .001). Similar trends occurred among the distressed subgroups and the younger but not older substance use subgroup who reported no significant coverage changes in either period. A minority in the distressed (31.4%–45.4%) and substance use (5.1%–8.5%) subgroups received treatment.

Conclusions. Although implementation of the ACA provisions coincided with coverage gains for young adults, challenges persist in engaging those with mental health and substance use problems in treatment.

Young adults are at high risk for being uninsured.¹ Because most lifetime substance use and other mental disorders have their onset by age 25 years,² coverage gaps among young adults have important implications for behavioral health care access. Starting in September 2010, the Affordable Care Act (ACA; Pub L No. 111-148) required private health insurers to allow children to remain as dependents on their parents’ plans up to age 26 years. Over the next few years, people aged 19 to 25 years compared with slightly older adults had greater gains in coverage³ and larger increases in mental health service use,⁴ although whether these gains have continued or diminished is not known.

The dependent care provision was followed in January 2014 by other ACA provisions including offering health insurance for individuals and small businesses through insurance exchanges. All plans sold on exchanges were required to cover some substance use and mental health services. Little is known about the effects of these private health insurance reforms on private insurance coverage and behavioral health treatment of young adults. We compared their effects on people aged 19 to 25 years and 26 to 35 years.

METHODS

We combined data from 2008 to 2016 National Survey on Drug Use and Health (NSDUH) cross-sectional nationally representative surveys of the US population with multistage sampling designs. Interviews were conducted with computer-assisted interviewing. Individuals without a household address, military personnel, and institutional residents were excluded from the sampling frame.

Health insurance coverage was assessed with questions about insurance type. Treatment was assessed separately for substances and mental health problems in any setting in the previous 12 months. Substance use disorders and major depressive episodes were assessed with structured diagnostic interviews according to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), criteria.⁵ Substance use disorders were limited to 4 disorders assessed consistently across the years: alcohol-, cannabis-, cocaine-, and heroin-use disorders. Serious psychological distress was defined by a Kessler-6 score of greater than or equal to 13.⁶ Other variables in the analyses included income, marital status, student status, and employment.

A quasi-experimental difference-in-differences design⁷ compared changes in private insurance coverage and treatment across adults aged 19 to 25 and 26 to 35 years across 3 consecutive 3-year periods: 2008 to 2010, 2011 to 2013, and 2014 to 2016. Changes between the first and second periods reflect changes associated with implementation of the dependent care coverage provision, and those between the second and third periods reflect changes associated with implementation of subsequent ACA insurance reforms. We sought to determine whether the 2010 dependent care provision absorbed pent-up coverage and service-use demands in the younger group or whether the 2014 ACA provisions had parallel effects in both groups. We used multivariable logistic regression models to estimate difference-in-differences models. We adjusted all percentages and models for serious psychological distress, substance use disorders, past-year major depressive episode, low income (< 138% of the federal poverty level according to the US Census Bureau), marital status, student status, and employment.

We conducted separate analyses for insurance coverage and treatment. We assessed insurance coverage in all participants and separately in those with substance use disorders and serious psychological distress. We performed treatment analyses separately in those with substance use disorders (substance use treatment) and serious psychological distress (mental health treatment). SAS-callable SUDAAN release 11.0.1 (Research Triangle Institute, Research Triangle Park, NC) accounted for NSDUH’s complex sample design and weights.

RESULTS

Table A (available as a supplement to the online version of this article at http://www.ajph.org) presents background characteristics of the 2 age groups. Among the study participants, 17.7% of adults aged 19 to 25 years and 11.9% of those aged 26 to 35 years had at least 1 past-year substance use disorder (P < .001) and 18.5% and 13.8% (P < .001), respectively, had serious psychological distress.

From the 2008-to-2010 period to the 2014-to-2016 period, the increase in private insurance for people aged 19 to 25 years (7.7 percentage points; P < .001) was significantly greater than the increase (1.2 percentage points; P = .02) for people aged 26 to 35 years. Between the first and second periods, the coverage increase for the younger group was significantly larger than the increase for the older group. Between the second and third periods, however, the coverage increases between the younger (3.2 percentage points; P < .001) and older (3.8 percentage points; P < .001) groups did not significantly differ (Table B, available as a supplement to the online version of this article at http://www.ajph.org).

We observed a similar pattern in both age groups with serious psychological distress (Table B). Among individuals with selected substance use disorders, a significantly greater overall coverage increase occurred for people aged 18 to 25 years than for people aged 26 to 35 years (difference-in-differences: 9.0 percentage points; 95% confidence interval [CI] = 5.5%, 12.5%). While the younger group with substance use disorders reported significant gains in coverage between the first and second (7.2 percentage points; P < .001) and second and third (2.9 percentage points; P = .007) periods, the older group reported little change in coverage (Figure 1).

FIGURE 1—

Trends in Private Insurance Coverage Among Young Adults With Serious Psychological Distress (SPD) and Substance Use Disorders (SUDs): National Survey of Drug Use and Health, United States, 2008–2016

Note. Both models adjusted for past-year major depressive episode, low income, marital status, student status, and employment. The SPD model also controls for SUDs. Adjusted difference-in-differences: 2008–2010 to 2011–2013: 6.7 percentage points (95% confidence interval [CI] = 2.8, 10.6); Adjusted difference-in-differences: 2011–2013 to 2014–2016: 2.3 percentage points (95% CI = −1.3, 6.1). The SUD model also controls for serious psychological distress. Adjusted difference-in-differences: 2008–2010 to 2011–2013: 8.9 percentage points (95% CI = 5.6, 12.2); adjusted difference-in-differences: 2011–2013 to 2014–2016: 0.9 percentage points (95% CI = −2.2, 4.0).

Among individuals with substance use disorders, there was little change in the percentage of the younger (2008–2010: 5.7%; 2011–2013: 5.1%; and 2014–2016: 5.4%) or older (8.5%, 8.0%, and 8.3%, respectively) groups reporting any treatment (Table C, available as a supplement to the online version of this article at http://www.ajph.org). Among adults with serious psychological distress, there was no overall difference in the percentage of the younger (2008–2010: 31.4%; 2011–2013: 33.2%; and 2014–2016: 32.0%) and older (45.5%, 42.0%, and 44.5%, respectively) groups who received mental health treatment (difference-in-differences: 1.6 percentage points; 95% CI = –1.5%, 4.7%). As compared with the older group with distress, however, the younger group reported a significantly larger increase in mental health treatment between the first and second periods (difference-in-differences: 5.3 percentage points; 95% CI = 1.8%, 8.8%), whereas the reverse was true between the second and third periods (difference-in-differences: −3.7 percentage points; 95% CI = −7.0%, −0.4%) (Table C, available as a supplement to the online version of this article at http://www.ajph.org).

DISCUSSION

Consistent with previous research, implementation of the 2010 dependent care provision coincided with disproportionate gains in private insurance coverage for adults aged 19 to 25 years who were the target of this policy.⁴ Similar trends occurred for people aged 19 to 25 years with common substance use disorders or serious psychological distress. For these groups, gains in insurance following implementation of the dependent care provision in 2010 were extended following the 2014 ACA insurance reforms.

Unlike the younger group, adults aged 26 to 35 years with substance use disorders did not experience gains in insurance following the 2014 provisions. Previous research reveals a strong correlation between substance use problems and lack of health insurance,⁸ especially among nonelderly adults,⁹ and a low level of awareness concerning the health insurance exchanges.¹⁰ These findings suggest a need for tailored outreach and educational efforts to encourage insurance acquisition by young adults with substance use disorders who are no longer age-eligible for coverage under their parents’ plans.

For young adults with substance use disorders, increases in insurance coverage did not translate into increases in substance use treatment. Extensive public financing of specialty treatment of substance use and the frequency with which individuals with substance use disorders report not being ready to start treatment¹¹ may contribute to lack of treatment gains. By contrast, following the dependent coverage provision, a differential increase occurred in mental health treatment by the younger group while the subsequent general ACA health insurance reforms were followed by a differential increase in treatment of the older group, perhaps related to pent-up demand.

This study has some limitations. A change in NSDUH survey design prevented examination of trends in insurance coverage or treatment of people with prescription stimulant-, sedative-, or opioid-use disorders. The NSDUH also does not distinguish plans purchased on exchanges from employer-based plans. Although the design controls for overall temporal effects of the macroeconomic decline, it does not control for possible age-specific responses to the economic decline in coverage acquisition or behavioral treatment.

Young adults are at high risk of substance use and mental health problems and for not having insurance coverage. Recent national trends underscore challenges of using policy levers to promote insurance acquisition as a means to increase behavioral treatment of young people with common mental health problems and substance use disorders.