American Journal of Obstetrics and Gynecology | 2019
“Safe Start”: A community health worker program that improves perinatal outcomes in high risk women: 295
Abstract
295 “Safe Start”: A community health worker program that improves perinatal outcomes in high risk women Sindhu K. Srinivas, Celeste Durnwald, Laura Line, Melissa Patti, Melissa Bucher, Shayna Cunningham, Valerie Riis Maternal and Child Health Research Center, Perelman SOM, University of Pennsylvania, Philadelphia, PA, Maternity Care Coalition, Philadelphia, PA, Maternal and Child Health Research Center, University of Pennsylvania, Philadelphia, PA, Yale School of Public Health, New Haven, CT OBJECTIVE: Prevalence of chronic conditions (e.g., obesity, hypertension, diabetes, depression, substance use disorder) in pregnant women is increasing, with higher rates among low-income and minority populations. We assessed the effectiveness of a community health worker (CHW) program to improve perinatal outcomes in lowincome pregnant women with chronic conditions. STUDY DESIGN: Between March 2015 and February 2018, we implemented and evaluated “Safe Start”, a CHW program representing a partnership between an inner city, hospital-based prenatal clinic, a community-based organization, and a Medicaid managed care insurer. Safe Start CHWs engage in patient navigation and systematic case review with health care providers and insurers. Patient eligibility criteria included: pregnant <34 weeks; presence of obesity, hypertension, diabetes, depression and/or a substance use disorder; and evidence of >1⁄4 1 missed appointment. Multivariable logistic and poisson regression were used to compare women enrolled in the program (n1⁄4208) with women who were eligible to participate but did not enroll (n1⁄4226), controlling for potential confounders. Outcomes assessed included: prenatal care adequacy, preterm birth, cesarean delivery, NICU admission, neonatal abstinence syndrome, postpartum visit attendance, breast-feeding and contraception use. RESULTS: Safe Start program participants were significantly more likely to be African American and report substance use than women in the comparison group (Table 1). Relative to the comparison