AbstractsMedical & Health Science

Background: Newborn deaths contribute 44% of all under-five deaths. Community health worker (CHW) during home-visits may identify and refer newborns to health facilities for postnatal care and treatment of danger signs. However, little is known on the care seeking practices and health system capacity to care for healthy and sick newborns in sub Saharan Africa. Objective: The overall objective of the studies was to assess newborn referral care seeking practices, compliance, and associated community and health systems factors in order to inform scale up of newborn care programs in Uganda and other low income countries with high newborn mortality. Methods: Four studies (I-IV) nested within a cluster randomized trial were conducted between 2011 and 2013 at the Iganga-Mayuge Health Demographic Surveillance Site in eastern Uganda. In Study I, focus group discussions (n=12) with men and women and in-depth interviews (n=11) with mothers and traditional birth attendants were used to obtain a deeper understanding of the social and cultural factors that affect caretakers’ compliance with community newborn referrals. Case vignettes, observations through role plays and record reviews were used in a cross sectional study to assess the ability of 57 trained community health workers to identify and refer sick newborns to health facilities (Study II). Study (III) was retrospective cohort of all referred newborns, during which interviews were held with 700 caretakers to determine compliance rate to seek health facility based care within 24-hours of a referral. In a cross sectional study, capacity to provide newborn care was assessed in all the 20 health facilities within the cluster randomized trial, using observations and interviews with of health workers (Study IV). Results: Community members understood the newborn period differently from health workers. A seclusion period observed immediately after birth restricted movement of the mother and newborn until the umbilical cord dropped off, but was not binding in case of illness (Study I). Of the 57 CHWs assessed, 68% were considered knowledgeable with a median knowledge score of 100% (IQR 94%-100%), and 36 (63%) considered skilled in identifying sick newborns (Study II). A total of 724 newborns were referred, of which 700 were successfully traced. Fifty three percent (373/700) were referred for postnatal care/immunization and 47% because they had at least one danger sign (Study III). Overall, 63% of the caretakers of referred newborns complied within less than24 hours, but more caretakers of sick newborns (243/327, 74%) complied, compared with 196/373 (53%) of those referred for immunization and postnatal care (p<0.001). A majority, (493, 77%) sought care from lower level health facilities. The determinants of compliance were: referred for danger signs Adjusted Odds Ratio (AOR) = 2.3, (95% CI: 1.6-3.5); CHW making a reminder visit to the referred newborn shortly after referral (AOR =1.7; 95% CI: 1.2 –2.7); and age of mother being 25-29 or 30-34 years, (AOR =0.4; 95% CI: 0.2 -…