The state of Connecticut is currently considering a number of policy options to improve health insurance affordability, access, and equity. However, prior to undertaking such initiatives, the state is interested in better understanding the current distribution of insurance enrollment. To create policies that seek to increase insurance coverage and access to care in underserved communities and reduce racial and ethnic disparities, state policymakers need an accurate picture of the current distributions of insurance enrollment across these dimensions. In particular, the state's All-Payer Claims Database (ACPD) does not contain complete information on race and ethnicity and is not representative of the self-funded group market.
In this study we combine data from the American Community Survey Public Use Microdata Sample (ACS PUMS), which includes information on race and ethnicity, as well as insurance status, with various data sources from the state to provide a fuller picture of insurance enrollment among those under the age of 65 in Connecticut. We also use published estimates of the impact of COVID-19 on insurance enrollment to provide a high-level estimate of how enrollment in Connecticut was affected during the early months of the pandemic.
To produce these estimates, we began with the ACS PUMS data, which are representative at the state level and include enrollment information for broad health insurance categories. We used iterative proportional fitting to reweight the age, gender, race, and ethnicity distributions to match the official Connecticut population estimates produced by the National Center for Health Statistics (NCHS), since those are the official estimates used by the Connecticut Department of Public Health. We combined the reweighted ACS PUMS data with a number of data sources on enrollment in various sources of health insurance in Connecticut in order to produce estimates for more granular insurance categories than are available in the ACS PUMS. For each source, we reweighted the ACS PUMS age and gender distributions by the age and gender distributions for that source from Connecticut. To keep population totals consistent, we did not reweight the total population for each insurance category, but instead the age and gender distributions within that population. We did not use race and ethnicity data from the Connecticut sources, since those fields have known issues of missingness and inaccuracy, as communicated to our team by individuals familiar with the data in Connecticut. Instead we used the race and ethnicity status in the ACS PUMS obtained after reweighting using the Connecticut-specific age and gender distributions. In order to produce high-level estimates of changes in insurance enrollment between 2019 and 2020 (when the major impacts of the COVID-19 pandemic started to become evident in the United States), we applied existing estimates of changes in enrollment in different insurance sources to our 2019 enrollment figures for Connecticut.
We find that insurance enrollment in Connecticut in 2019 was generally high, with 93.5 percent of the population enrolled. Younger individuals—especially those under the age of 18—had the highest rate of coverage by Medicaid compared with other age groups, which is consistent with broader Medicaid eligibility criteria for children. Employer-sponsored insurance (ESI) covered the largest proportion of individuals within each age category, but was highest among those ages 40–49, with over two-thirds of that population enrolled in ESI.
There were substantial differences in insurance coverage by race and ethnicity. Asians had the highest rate of ESI coverage (76.0 percent), compared with 64.0 percent of white individuals, 52.8 percent of black individuals, and 45.4 percent of those of other races. The opposite pattern was true for coverage by Medicaid. Within the individual market, subsidized enrollment was highest among those who are white or of another race (46.6 percent and 55.2 percent, respectively), while 30.8 percent of black enrollees and 35.2 percent of Asian enrollees on the individual market received subsidies. Considering differences by ethnicity (Hispanic versus non-Hispanic), Hispanic individuals had a substantially higher rate of coverage by Medicaid (35.7 percent) compared with non-Hispanics (19.4 percent).
Within the individual market, younger people were more likely to have off-marketplace coverage, which is consistent with the notion that younger, healthier individuals may purchase low-priced plans outside of the state marketplace; the use of student health insurance provided via universities and colleges (which are considered individual insurance plans) may also contribute to off-marketplace individual market enrollment among younger people. Within the ESI market there was little variation by age, gender, race, or ethnicity in the distribution across markets (e.g., small groups versus large groups).
We also produced high-level estimates of the changes in insurance enrollment in Connecticut between 2019 and 2020, which includes the early months of the COVID-19 pandemic. Recent estimates of changes in insurance enrollment during this time period, which we applied to the Connecticut enrollment estimates, suggest that uninsurance actually decreased slightly, while Medicaid coverage increased and private insurance coverage fell.
This study provides the state of Connecticut with estimates of the distributions in enrollment by age, gender, race, and ethnicity in detailed insurance categories. The complex nature of this work highlights the need for better, more detailed health insurance enrollment data for state policymakers to aid in decisionmaking.
This research was jointly funded by Arnold Ventures and the Commonwealth Fund and was carried out within the Payment, Cost, and Coverage Program in RAND Health Care.