The Affordable Care Act and Insurance Coverage for Young Adults FREE

To the Editor: The Affordable Care Act (ACA) was designed to expand health insurance to millions of uninsured children and adults. The ACA's dependent coverage provision allows young adults to remain on their parents' insurance plans until they turn 26 years old. This took effect for plans renewing on or after September 23, 2010.

Preliminary data suggest that this policy has already had an effect. The National Health Interview Survey showed an 8.3 percentage-point increase in young adults with insurance between September 2010 and June 2011, approximately 2.5 million adults.¹ Census data and independent surveys showed similar increases,^2- 4 but questions remain: What types of coverage did young adults acquire? Did trends differ for adults not affected by the dependent coverage provision or for particular racial/ethnic groups? Our study examines changes in insurance coverage for young adults from 2005 through 2010.

We used the Current Population Survey's (CPS’s) 2006-2011 data sets, covering calendar years 2005-2010 (which incorporate recent revisions to the survey's insurance estimates). The CPS is an in-person survey of noninstitutionalized civilians aged 15 years or older conducted by the US Census Bureau, using a nationally representative probability sample of approximately 60 000 households monthly, with a 92.4% response rate.⁵ Our sample contained adults aged 19 to 34 years. We compared those aged 19 to 25 years with adults aged 26 to 34 years as controls. A differences-in-differences model was used with linear regressions adjusted for age group and survey year. Results were survey-weighted for national estimates, with standard errors clustered at the household level. The coefficient of interest was the interaction term 2010 × ages 19-25 years, which reflects the net change in coverage for 19- to 25-year-olds in 2010 vs 2005-2009, subtracting out the coverage trend in those aged 26 to 34 years.

Outcomes were the percentage-point changes in those with any insurance, Medicaid, private coverage, employer-provided insurance, and direct-purchase insurance, with the latter 2 divided into “own policy” and “dependent policy.” We conducted analyses stratified by race and ethnicity. All analyses used a 2-tailed significance level of .05, and were conducted using Stata version 12.0 (StataCorp). This project was exempted from institutional review board review under federal guidelines because it used only publicly available deidentified data.

Overall, 247 370 adults aged 19 to 34 years were surveyed. Insurance coverage increased significantly from the 2005-2009 period to 2010 for 19- to 25-year-olds (70.1% to 70.3%) compared with 26- to 34-year-olds (74.7% to 72.0%), a net gain of 2.9% (95% CI, 1.7% to 4.2%; P < .001) (Table 1).

Private insurance increased significantly by 2.8% (95% CI, 1.4% to 4.2%; P < .001), with no significant change in Medicaid (0.7% [95% CI, −0.1% to 1.6%]; P = .10). Among private insurance, there was a significant increase in dependents on employer-provided plans (4.3% [95% CI, 3.2% to 5.5%]; P < .001) and a smaller decline in young adults with policies in their own names (−2.5% [95% CI, −3.6% to −1.3%]; P < .001). White, black, Latino, Asian, and Native American young adults all experienced significant increases in coverage (Table 2).

The implementation of the ACA's dependent coverage provision in 2010 was associated with a significant increase in insurance coverage among young adults. The coverage changes of increased dependent coverage in private insurance, no change in Medicaid, and a decline in own private coverage were consistent with what would be expected under this provision. Insurance gains occurred among all racial and ethnic groups, with gains greatest for minorities.

Our study's main strength is its use of a pre-post design and a control group. Our analysis has 2 primary limitations. First, ours is an observational study, so causality cannot be established. In terms of alternative explanations, while the ACA also created new high-risk insurance pools in 2010, enrollment was only 21 000 nationally by April 2011, too low to explain our findings.⁶ Second, our data capture only part of any potential ACA effect because it took until September 2011 before all plans became subject to the dependent coverage provision. Analyses in subsequent years will be informative regarding the long-term implications of this provision.