• Kathryn Wouk, Isabel Morgan, Jasmine Johnson, Christine Tucker, Rebecca Carlson, Diane C Berry, and Alison M Stuebe.
• Department of Maternal and Child Health, Carolina Global Breastfeeding Institute, Gillings School of Global Public Health, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA.
• J Womens Health (Larchmt). 2021 Aug 1; 30 (8): 112711591127-1159.

AbstractIntroduction: People of color and low-income and uninsured populations in the United States have elevated risks of adverse maternal health outcomes alongside low levels of postpartum visit attendance. The postpartum period is a critical window for delivering health care services to reduce health inequities and their transgenerational effects. Evidence is needed to identify predictors of postpartum visit attendance in marginalized populations. Methods: We conducted a systematic review of the peer-reviewed literature to identify studies that quantified patient-, provider-, and health system-level predictors of postpartum health care use by people of color and low-income and uninsured populations. We extracted study design, sample, measures, and outcome data from studies meeting our eligibility criteria, and used a modified Cochrane Risk of Bias tool to evaluate risk of bias. Results: Out of 2,757 studies, 36 met our criteria for inclusion in this review. Patient-level factors consistently associated with postpartum care included higher socioeconomic status, rural residence, fewer children, older age, medical complications, and previous health care use. Perceived discrimination during intrapartum care and trouble understanding the health care provider were associated with lower postpartum visit use, while satisfaction with the provider and having a provider familiar with one's health history were associated with higher use. Health system predictors included public facilities, group prenatal care, and services such as patient navigators and appointment reminders. Discussion: Postpartum health service research in marginalized populations has predominantly focused on patient-level factors; however, the multilevel predictors identified in this review reflect underlying inequities and should be used to inform the design of structural changes.

17 keywords...

hide…