A low resource simulation-based training package leads to increased knowledge and skill retention post Basic Emergency Obstetric and Neonatal Care in rural Tanzania; A stepped- wedge research design Short Title: BEmONC simulation-based training package leads to skill retention in Tanzania

Abstract Background: The Basic Emergency Obstetric and Neonatal Care (BEmONC) training program merges core modules from Helping Mothers Survive and Helping Babies Survive training programs to improve obstetrical and newborn outcomes. The duration of BEmONC training varies between facilities and countries. Longer term low dose high frequency practice has been shown to prevent skill decay in both Helping Mothers Survive and Helping Babies Survive literature. A recent program evaluation of BEmONC training with extensive changes to infrastructure in rural Tanzania links training with improve obstetric and neonatal outcomes. This study sought to document skill and knowledge outcomes after disseminating a low resource focused simulation package. Methods: A stepped-wedge research design was used to test the intervention: the simulation package. Participants were midwives/nurses, clinical officers, and medical officers from two hospitals and four Health Centres in Mwanza Region, Tanzania. Facilities were purposely assigned to one of the two clusters. The intervention had three components: (1) 5-day BEmONC workshop, (2) peer to peer facility-based simulation practice using peer cards and (3) clinical mentorship. Cluster A and Cluster B attended the 5-day BEmONC workshop. Cluster A was supported with the component two and three post workshop. Cluster B received these components after the 6-month assessment. Knowledge and skill were analyzed at baseline, post workshop, at 6 months and at 12 months using the r core statistics; p-values < 0.05 were considered statistically significant. Results: Cluster A (n=7) and Cluster B (n=8) participants demonstrated significant knowledge and skill improvements post-initial workshop. At 6 months, Cluster A aggregate skill scores were significantly higher than Cluster B, who showed skill decay. At 12 months, aggregate skill scores between Cluster A and Cluster B were similar. Conclusion: The simulation package was feasible and effective for improving baseline skill, and retaining skill sets at 6 and 12 months. This is the first study to pilot the newly developed essential care for labor and birth module. This approach provides a low resource education option for improving healthcare provider abilities to provide quality care. More research is needed to link low resource initiatives with clinical outcomes.