P0272 - Risk factors for stillbirth and neonatal mortality among Kenyan women in Mobile WACh NEO, a pilot two-way SMS communication program.

Background.
Globally, 2.6 million neonates die each year with an equal number of stillbirths. The majority of these deaths are preventable with improved access to skilled delivery attendance and early neonatal care. Two-way SMS communication with health care workers is an efficient and low-tech intervention to connect women and their babies to this care. Messaging can promote evidence-based interventions to support facility delivery, identify neonatal danger signs and engage women in appropriate care seeking. Understanding the populations that will benefit from this type of intervention is critical to its scalable success. We analyzed data from a pilot demonstration project to determine the maternal and neonatal characteristics associated with perinatal mortality and describe the use of SMS by these women to inform the design of a larger trial.

Methods.
Mobile WACh NEO was a longitudinal demonstration project implemented at 2 public maternal child health clinics in Kenya- one urban and one rural. Women in their third trimester of pregnancy were enrolled and excluded if they did not have access to any mobile phone. All participants received weekly and then daily peripartum personalized, action-oriented SMS with a question about birth preparation, neonatal warning signs and newborn care and were encouraged to send questions to a study nurse. Primary outcomes for this analysis were perinatal death (deaths from 28 weeks gestation through 28 days of age) and subcategories of stillbirth and neonatal death. These were determined by maternal report through SMS, phone call and/or during a 14 week postpartum interview. Correlates were determined using binomial regression adjusted for clinic site.

Results.
798 pregnant women were enrolled at a median gestation of 32 weeks (IQR 30, 34). Follow-up was available on 797. Median age was 24 years (IQR 21, 29) and 82% (657) owned their own phone. Demographics differed between rural and urban sites. 92% (737) sent at least one message. Thirty women (3.8%) suffered a perinatal loss: 13 stillbirths and 17 neonatal deaths. Each year of maternal age was associated with elevated risk of stillbirth [aRR 1.13 (CI 1.02 - 1.24)]. No other risk factors for perinatal death were identified. Among mothers who went home after birth and then experienced an infant death, the majority of mothers (10/11) did not message the study nurse with concerns about their baby in the days preceding death.

Conclusion.
Increasing maternal age was a risk factor for stillbirth, however no other specific maternal risk factors for perinatal mortality were identified. This suggests targeting specific high risk populations may not be necessary. Low levels of messaging from mothers whose infant died highlights the need for future interventions to focus on anticipatory guidance and engage women in early identification of neonatal illness.