General Internal Medicine Fellow Boston Medical Center, United States
Background: Mifepristone is a safe and effective medication for early abortion in combination with misoprostol, and it accounts for an increasing number of abortions in the US. Hundreds of new state-level abortion restrictions were implemented between 2010 and 2020 with unknown impact on mifepristone uptake.
Objectives: To determine the effect of state level abortion restrictions on mifepristone use from 2010 to 2020.
Methods: We conducted a state-level repeated cross-sectional study to estimate the effect of abortion hostility on mifepristone claims from 2010 to 2020. We used MarketScan commercial insurance claims data to create a cohort of reproductive-aged female enrollees aged 15-49 in each year. We created an annual state-level abortion hostility index adapted from the Guttmacher Institute and other sources. The resultant index is a 13 point score based on presence of policies protecting or restricting abortion access. The outcome was mifepristone for abortion identified using HCPCs codes. We used a linear mixed effects model with mifepristone claims as the outcome, predictors of baseline abortion hostility score and change in hostility score from baseline, and potential confounders of urbanicity and age group.
Results: In 2010, the median baseline abortion hostility index was 2 (IQR -1 to 3). 21 states had no change in score over the 10-year study period. For the other 29 states the median change in hostility index was 1 (IQR -1 to 2). The median rate of mifepristone claims ranged from 33.5 to 61.6 claims per 100,000 reproductive-aged female enrollees from 2010 to 2020. A one unit increase in baseline hostility index was associated with a change of -17.0 in the rate of mifepristone claims (95% CI: -24.1 to -9.8). A one unit increase in the change in hostility index over follow-up was associated with change of -2.8 in the rate of mifepristone claims (95% CI: -8.8 to 3.1). After adjusting for urbanicity and age group, baseline hostility index and change in hostility index were associated with changes of -16.1 (95% CI: -23.2 to -9.0) and -7.4 (95% CI: -13.4 to -1.4) in the rate of mifepristone claims, respectively.
Conclusions: Higher baseline levels of state abortion hostility and greater increases in state abortion hostility from 2010 to 2020 were associated with fewer mifepristone claims per 1000 female MarketScan enrollees age 15-49. The strong association between baseline hostility and mifepristone use indicates important structural determinants of mifepristone access that predated the new restrictions on abortion enacted between 2010 and 2020. Intervention is needed at multiple levels to improve access.
