Defense Date


Document Type


Degree Name

Doctor of Philosophy



First Advisor

Dr. Kristina B. Hood


Black women and HIV-positive women have increased maternal mortality rates and other negative pregnancy outcomes, in part due to disparate prenatal care. Although women who seek abortions do not have the same negative outcomes, abortion stigma exists and is normalized in healthcare. Limited work has examined prenatal care provision for women in these groups, and even less work has explored the prenatal care provision by healthcare trainees (i.e., medical, nurse practitioner, and physician assistant students). Examining the role of bias on the prioritization of prenatal care items by healthcare trainees is imperative. Healthcare education sets the stage for future practice, and as such it is important to examine students’ biases and assumptions before they become full-time providers. One hundred twenty-six participants were recruited from various healthcare training programs to complete an online experiment using a 2 (patient race: Black, White) x 2 (patient HIV status: HIV-, HIV+) x 2 (patient abortion status: has never had an abortion, has had an abortion). Participants were randomly assigned to read one vignette for a patient, then were asked to prioritize two sets of prenatal care items for their patients in two minutes or less. They also completed a series of measures to be included as potential covariates in our analyses.

Through a combination of univariate Kruskal-Wallis tests, ordinal logistic regressions, and binary logistic regressions we assessed seven hypotheses. For H1 through H3, we predicted that each individual condition (patient race, HIV status, and abortion status) would interact with respondent implicit biases to predict care item prioritization. We found some significant effects for H2, the interactions of HIV status and HIV-related bias, such that, generally, individuals with higher HIV-related stigma were more likely to screen patients for drug use. We also found significant effects for H3, the interactions of abortions status and abortion stigma, such that students, generally, were more likely to screen patients who have had abortions for tobacco use. For H4 through H7, we looked at the unique interactions of our conditions, controlling for implicit biases. Though none of the overall models were significant, we did find several significant pairwise comparisons across these hypotheses.

There are many potential explanations for our findings, including a small sample size yielding lower power than anticipated, a need for more complex or ambiguous patient vignettes, and the possibility that the respondents’ biases do not impact their perceptions of prenatal care. Potential limitations include the lack of statistical power across many analyses due to an error in the implicit bias task software, a sample of primarily medical students, and ongoing issues such as our nation’s grappling with systemic racism and the COVID-19 pandemic. Despite a lack of confirmation of many of our hypotheses, this study lays the foundation on which to build subsequent studies related to implicit bias and prenatal care, along with important information for amending healthcare training programs, reducing bias among healthcare trainees, and improving pregnancy outcomes for all.


© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission