publications

Current Version

revisions requested - Journal of Human Resources

In 2011, Pennsylvania passed regulations requiring abortion-providing facilities to meet ambulatory surgical facility standards, which ultimately caused the closure of almost half of the state’s abortion facilities. All closing facilities were geographically near facilities that remained open, meaning distance to the nearest clinic was unchanged while local clinic capacity fell. I use a difference-in-differences design supplemented with a synthetic control method and find that reduced clinic capacity caused 20-30 percent fewer abortions in the first 8 weeks of gestation and more abortions at later gestational ages. While evidence suggests births may have increased slightly, the main impact closures had on local women was a delay in abortions.

Despite leading the world in medical spending and innovation, the United States continues to experience a crisis in birth-related health outcomes. Since high levels of spending and medical technology innovation have failed to improve outcomes like infant and maternal mortality, other approaches must be examined. In this paper, I study the effect of Medicaid coverage of doula services in the pregnancy, delivery, and perinatal process. Using variation in the timing of coverage start dates and comparing Medicaid births without doula coverage to Medicaid births with the option of doula services, I find that Medicaid support for doula services increases the likelihood of a doula being present at delivery by up to 25%. Due to the small share of births ever involving a doula, this effect is small in absolute terms (.1% of all Medicaid births) but represents a meaningful percentage change. Medicaid births with doulas present are generally healthier than those without a doula present: these births are less likely to be pre-term, low birthweight, or to result in a Cesarean section after a failed labor attempt.

Current Version

Until recently, women across the United States had to visit a physician (through private practice or federally funded family planning centers) to obtain a prescription for the pill. With this barrier to access to contraception under the standard, physician-prescribing approach, allowing women to access the pill through pharmacist prescriptions could provide women with better control over their family formation decisions. Using pharmacists zip codes at the time of licensure and county-level birth and abortion data in Oregon, I find that pharmacist prescription of birth control reduces the county-level abortion rate by an average of 4-7% (approximately 0.04 fewer abortions per 1000 women of childbearing age) across all post-period quarters – though the effects are not statistically significant. Births, on the other hand, seem to stay the same or increase – suggesting that pharmacists' prescriptions may cause false confidence and user error among patients or that people of childbearing age are better able to plan the timing of their pregnancies. While these results lack statistical significance, future steps on this project will include geographically disaggregated birth data, which may reduce some of the treatment measurement error currently in place in the county-level analyses.

 (with Slawa Rokicki) 

Papers with a * indicate work with undergraduate students at Grinnell College.