Enroll America--alongside Civis Analytics, a technology and research firm--recently updated its model to find uninsured Americans and where they live.
So to rebuild its formula, Enroll America surveyed more than 8,000 consumers to discuss their current health insurance status, their health insurance status in 2013, their income and their employment status.
It then compared these findings to the previous uninsured model--the old method considered around 500 variables then dropped down to 30 variables to determine the odds that an individual was uninsured. Then, it shared this data with workers in local communities who found that neighborhood data was more useful than individual scores, reports The New York Times.
"We started using the model initially to try to get to people at an individual level, and there was a degree of accuracy there," Gregg Ross, Enroll America's organizing director in North Carolina, one of 11 states where the group made a big push, told The Times.
Based on this update, Enroll America found that the average uninsured score declined 5.1 percentage points from 16.4 in 2013 to 11.3 in 2014.
These findings are on par with other polls' data that determined the number of uninsured Americans decreased by 18 percent in September 2013 to 13.4 percent in June 2014, FierceHealthPayer previously reported.
The new formula also found there was a higher rate of enrollment among those ages 26 to 35, Latinos, women and individuals who lived in Medicaid-expanded states.
What's more, this detailed model paints a clearer picture of who has insurance in America, and who does not, suggests The Times.