Elsa Pearson, MPH, is the policy director of the Partnered Evidence-based Policy Resource Center at VA Boston. She tweets at @epearsonbusph.
Pregnancy is a significant life event, one that typically leads to substantially more interaction with the health care system than average. In the United States (US), pregnant people usually have about one health care visit per month of pregnancy, during which they receive a myriad of services. However, access to high quality prenatal care — and enough of it — is often limited by one’s health insurance coverage.
When the Affordable Care Act was enacted, it established the individual Marketplaces from which those who are ineligible for Medicaid, Medicare, and/or employer-sponsored insurance can purchase coverage. However, pregnancy is not considered a qualifying life event, so an individual cannot just sign up for coverage once they find out they’re pregnant; they must wait until the next open enrollment period or the birth of their child, whichever comes first. Thus, they may be stuck without coverage during pregnancy. This can have a significant impact on access to appropriate prenatal care.
A recent study in Health Affairs looked at Marketplace enrollment patterns for pregnant people and the impact of Marketplace insurance coverage on their health and care utilization.
The authors are Sarah Gordon and Melissa Garrido from Boston University School of Public Health (BUSPH) Health, Law, Policy, and Management Department (HLPM) and VA Boston Healthcare System; Charlotte Alger from BUSPH HLPM; and Eugene Declercq from BUSPH Community Health Sciences Department.
The authors used data from the Pregnancy Risk Surveillance and Monitoring System (PRAMS) from 2016 to 2018. Developed by the Centers for Disease Control and Prevention, PRAMS is a self-reported survey within 40 states and New York City and is representative of 83 percent of all US births. State health departments pull a representative sample of recent births from birth certificate registries and reach out via mail and telephone to the selected mothers. The survey asks respondents about health status and behaviors, health care use, and insurance coverage.
With these data, they studied two questions. First, they assessed how likely pregnant people were to be enrolled in Marketplace insurance coverage preconception, during pregnancy, and/or postpartum. Sample size for this question was 6491 and the authors used simple descriptive analysis techniques.
Second, they studied how Marketplace enrollment impacted individuals’ receipt of prenatal care, such as the number of prenatal visits, receipt of care within the first trimester, and receipt of specific health care services like flu shots and screenings for intimate partner violence and depression. The sample size for this question was 3443, limited to individuals who reported Marketplace coverage during pregnancy. The authors used logistic regression models and inverse probability of treatment weights to conduct these analyses.
For enrollment, the authors found that about one third of respondents had continual Marketplace coverage, from preconception to postpartum. Of those who were only enrolled in the Marketplace preconception, over 70 percent reported Medicaid coverage during pregnancy. Of those who were only enrolled in the Marketplace postpartum, almost 50 percent reported Medicaid coverage and one third reported employer-sponsored insurance coverage during pregnancy.
For impact of enrollment during pregnancy, the authors compared those with continuous coverage (preconception to postpartum) to those who only enrolled in the Marketplace during pregnancy. Those with continuous Marketplace coverage were more likely to have “adequate” or “more than adequate” prenatal care use. (The authors defined these classifications using the Adequacy of Prenatal Care Utilization Index which measures timing and quantity of care.) Those with continuous coverage were also more likely to initiate prenatal care in the first trimester, though over 80 percent of respondents in both groups did so. The authors did not find any significant differences in the likelihood of receipt of particular prenatal services, such as flu shots or social/mental health screenings.
There were several limitations to this study due to the nature of the PRAMS data set. For example, PRAMS is self-reported, subject to both recall bias and response bias. Plus, the survey is not conducted in all states and, thus, assumptions must be made about generalizability. Lastly, PRAMS simply includes a finite set of questions; this is certainly understandable but does limit researchers’ analyses.
With the connection between insurance coverage and access to care clear, several notable policy questions arise from this study. Classifying pregnancy as a qualifying life event is perhaps the most obvious. As mentioned previously, pregnancy is not a qualifying life event, though the birth of a child is. (Only two states have implemented policies to the contrary.) Allowing an individual to sign up for health insurance coverage once pregnant, rather than waiting until birth or the next open enrollment period, could improve access to prenatal care and even improve maternal and child health outcomes.
Another related policy implication is determining what type of insurance is ideal for pregnant individuals. The authors found that individuals without Marketplace coverage often have other types of coverage, at least temporarily. What type of insurance is best or most cost-effective for pregnant people — and the benefits of coverage continuity regardless of type — could be studied further.
The study did not touch on the quality of prenatal care but that is also worth discussion. In the US, pregnant people tend to receive far more prenatal care than other countries but that doesn’t mean the quality is better, nor do maternal health outcomes suggest that’s true. In fact, the US’ maternal health outcomes are some of the worst in the industrial world.
Pregnancy is full of changes, expenses, and challenges. Determining how Marketplace insurance coverage — which has been around for a decade — access to care, and maternal and child health outcomes all interact from preconception to postpartum warrants more study.