Associations Between Hospital Maternal Service Level and Delivery Outcomes Jennifer Vanderlaan, PhD, MPH, RN, CNM a, * , Roger Rochat, MD b , Bryan Williams, PhD a , Anne Dunlop, MD, MPH a , Susan E. Shapiro, PhD, RN, FAAN a a Emory University Nell Hodgson Woodruff School of Nursing, Atlanta, Georgia b Emory University Rollins School of Public Health, Atlanta, Georgia Article history: Received 19 July 2018; Received in revised form 4 February 2019; Accepted 22 February 2019 abstract Objective: This study explored the associations between delivery hospital self-reported level of maternal service, as defined by the American Hospital Association, and both maternal and neonatal outcomes among women at high maternal risk, as defined by the Obstetric Comorbidity Index. Methods: This was a secondary analysis of linked delivery hospitalization discharge and vital records data for women experiencing singleton births in Georgia from 2008 to 2012. The need for maternal transfer was defined using a sample- specific cut-off of the risk score calculated using the Obstetric Comorbidity Index. Outcomes included poor maternal outcome (severe maternal morbidity or death), maternal length of stay, preterm delivery, low birth weight, and peri- natal death. The analysis was completed using hierarchical logistic regression with a two-level model considering hospital level of maternal service and controlling for maternal race and transfer status. Results: In these data, there was no difference in the odds of a poor maternal or neonatal outcome according to delivery hospital level of maternal care; however, delivery at a hospital with maternal service level III was associated with a higher odds of an extended length of stay. Conclusions: For this group of pregnant women in need of maternal transfer, delivery hospital self-reported level of maternal care was not associated with the odds of poor maternal or neonatal outcomes. This study supports the need for improved definitions of hospital level of maternal services. Ó 2019 Jacobs Institute of Women's Health. Published by Elsevier Inc. The American College of Obstetricians and Gynecologists has proposed a structure for defining hospital maternal service level that, it is believed, will help to reduce severe morbidity and mortality by better stratifying hospital services into categories that match the needs of women at high maternal risk (American College of Obstetricians and Gynecologists and Society for Maternal–Fetal Medicine et al., 2015). The leveling of hospitals, based on the availability of designated capabilities, is one component of a regionalization system that allows women to be directed to the nearest hospital with the level of delivery care commensurate with the complexity of the case (Berns, 2010). According to the three delays framework, delays in directing women at high maternal risk to appropriate delivery hospitals is the second delay that increases maternal morbidity and death (Thaddeus & Maine, 1994). Preventing this delay involves the early identification of a woman’s elevated risk and directing her to a higher acuity antepartum and intrapartum care. Regional systems speed the process by specifying which delivery hospitals are appropriate for women at high risk. Although the 1975 proposal called for integrated maternal and fetal care networks, evaluation and development focused on the fetal component (Hankins et al., 2012). Before improvements in defining maternal service level can be identified, baseline associations between maternal service level and delivery outcomes for current defi- nitions of maternal service level must be measured. However, evaluating associations between delivery outcomes and maternal service levels has several challenges. The first challenge is identifying datasets that allow for the analysis of both maternal and neonatal outcomes. Decisions regarding maternal care, such as the provision of steroids, delay Supported by grant number R36HS024655 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality. * Correspondence to: Jennifer Vanderlaan, PhD, MPH, RN, CNM, Emory Uni- versity Nell Hodgson Woodruff School of Nursing,1520 Clifton Road, Atlanta, GA 30322. Phone: þ518-209-2229; fax: 404-727-8514. E-mail address: javande@emory.edu (J. Vanderlaan). www.whijournal.com 1049-3867/$ - see front matter Ó 2019 Jacobs Institute of Women's Health. Published by Elsevier Inc. https://doi.org/10.1016/j.whi.2019.02.004 Women's Health Issues xxx-xx (2019) 1–7