Roberts studies the ways that policies and the health care system punish, rather than support, vulnerable pregnant women, including pregnant women using alcohol and drugs and pregnant women considering abortion. Her research falls into three categories. First, she conducts research about the relationship between alcohol and drug use and abortion, including alcohol and drug use as reasons for abortion, how alcohol and drug use affect women's ability to obtain an abortion, and how having an abortion affects women's alcohol and drug use. Second, she researches women's experiences with state-level abortion restrictions, such as waiting periods, targeted regulation of abortion provider (TRAP) laws, and "20-week" abortions bans, and women's experiences with state-level policies targeting alcohol and drug use during pregnancy. Third, she conducts research to inform the development of genuine public health approaches to abortion, including generating the evidence-base to inform facility standards for abortion facilities and research to inform how health departments should engage with abortion.
As part of her work related to pregnant women's alcohol and drug use, Roberts has served as an advisor to multiple local health departments, and served on advisory committees for the March of Dimes and CA ACOG, the Association of Reproductive Health Professionals, as well as SAMHSA. As part of her research about abortion, she has served on advisory committees for projects focusing on insurance coverage for abortion and on later abortion.
Using qualitative data collected from pregnant and parenting women using alcohol and/or drugs in a California county, finds that women using drugs attended and avoided prenatal care for reasons not connected to their drug use: concern for the health of their baby, social support, and extrinsic barriers such as health insurance and transportation. Also finds that drug use itself was a barrier for a few women. In addition to drug use, women experienced multiple simultaneous risk factors. Both the drug use and the multiple simultaneous risk factors made resolving extrinsic barriers more difficult. Women also feared the effects of drug use on their baby's health and feared being reported to Child Protective Services, each of which influenced prenatal care use.
Using data from the UCSF Turnaway Study (n=956 women seeking abortion in the United States), we found that nearly 5% reported alcohol, tobacco, or drug use as a reason for deciding to have an abortion. Women reporting alcohol as a reason drank at levels exceeding a low threshold and did not appear to be terminating otherwise wanted pregnancies. None of the women reporting drug use as a reason reported fear of punishment as a reason for abortion. Ninety-eight percent of women reporting alcohol, tobacco, or drugs as a reason had unintended pregnancies. Findings are inconsistent with hypotheses that recommendations for abstinence from alcohol during pregnancy and punitive policies relating to drug use lead women using low levels of alcohol or using drugs to terminate otherwise wanted pregnancies.
Estimates how Louisiana's Hospital Admitting Privileges law for abortion providers would affect the distance women need to travel for an abortion. Finds that if all of Louisiana's abortion facilities close (one of the possible outcomes of the law), the distance women would need to travel for an abortion would more than triple from 58 miles each way to 208 miles each way. Also finds that if all of the facilities close, the proportion of Louisiana women of reproductive age who would need to travel more than 150 miles for an abortion would increase from 1% to 72%.
Examines the relationship between receiving versus being denied an abortion and subsequent alcohol use over two and a half years, using data from UCSF’s longitudinal Turnaway Study (n=956). While women who had an abortion were more likely to report any and binge alcohol use than women denied an abortion and gave birth, this difference was due to a reduction in alcohol use among women giving birth rather than an increase in use among women having abortions. Thus, assertions that having an abortion leads women to increase alcohol use to cope with having had an abortion are not supported.
Examines women’s experiences with Utah’s 72-hour waiting period and two-visit requirement for abortion, which was the first 72-hour waiting period law to go into effect in the U.S. Overall, Utah’s 72-hour waiting period and two-visit requirement did not prevent women who presented for information visits from having abortions, but did burden women with financial costs, logistical hassles and extended periods of dwelling on decisions they had already made. They also led some women to worry that they may not be able to have the type of abortion they preferred and pushed at least one women beyond her facility’s gestational limit for providing abortion care. The two-visit requirement increased the cost of the abortion by 10% and the 72-hour wait was actually an 8-day wait in practice.