People who had gone three years or more without having health insurance made up more than half of the gains in coverage in 2014, the first year of the Affordable Care Act (ACA)’s insurance marketplace and Medicaid expansion.

The study, written by Agency for Healthcare Research and Quality (AHRQ) senior fellows Sandra Decker and Brandy Lipton, was published in the January 2017 issue of Health Affairs.

Using data from the 2013 and 2014 National Health Interview Survey, the researchers analyzed changes in the overall uninsured rate and the changes among the long-term (three years or more), medium-term (one to three years) and short-term (less than a year) uninsured.

Roughly one in 10 (9.4 percent) of nonelderly adults qualified as long-term uninsured in 2013. Once the exchanges and Medicaid expansion came into effect, that number dropped to 7.1 percent, greater than a 0.7 percentage point drop in short-term uninsured and the 1.1 percentage point decline in the medium-term uninsured.

“Most nonelderly adults who gained insurance in 2014 had been uninsured for more than three years,” Decker and Lipton wrote. “The long-term uninsured represented the majority of uninsured people in 2013, before most of the ACA provisions were implemented. However, it was not clear how well they would be represented among the newly insured in 2014, since their propensity to take up insurance might have differed from that of people who had been uninsured for a shorter time.”

The analysis didn’t determine exactly what kind of coverage the long-term uninsured gained in 2014. While the overall uninsured rate went down more in Medicaid expansion states than nonexpansion states, the difference in the decline came from the short-term uninsured.

Having the gains in coverage from people who aren’t used to accessing the healthcare system makes it hard to judge how they’ll use their insurance. Decker and Lipton argued it could mean 2014 would see a one-time spike in utilization as the long-term uninsured use services they may have needed for years.

“This may imply that initial changes in utilization may be larger than subsequent changes,” they wrote. “On the other hand, the long-term uninsured are much less likely to have a usual source of care, compared to those who have been uninsured for a shorter time, and it could take time for the long-term uninsured to find primary care providers and establish regular relationships with them. This may imply that changes in the use of care observed in 2014 may be smaller than subsequent changes.”

Future studies, they concluded, may wish to examine the extent to which the previously long-term uninsured were able to access healthcare providers.