RT Journal Article SR Electronic T1 Cohort study of neonatal resuscitation skill retention in frontline healthcare facilities in Bihar, India, after PRONTO simulation training JF BMJ Paediatrics Open JO BMJ Paediatrics Open FD BMJ Publishing Group Ltd SP e000628 DO 10.1136/bmjpo-2019-000628 VO 4 IS 1 A1 Brennan V Higgins A1 Melissa M Medvedev A1 Hilary Spindler A1 Rakesh Ghosh A1 Ojungsangla Longkumer A1 Susanna R Cohen A1 Aritra Das A1 Aboli Gore A1 Tanmay Mahapatra A1 Dilys M Walker YR 2020 UL http://bmjpaedsopen.bmj.com/content/4/1/e000628.abstract AB Background Use of simulation in neonatal resuscitation (NR) training programmes has increased throughout low-income and middle-income countries. Many of such programmes have demonstrated a positive impact on NR knowledge and skill acquisition along with reduction of early neonatal mortality and fresh stillbirth rates. However, NR skill retention after simulation programmes remains a challenge.Methods This study assessed facility level NR skill retention after PRONTO International’s simulation training in Bihar, India. Training was conducted within CARE India’s statewide in-job, on-site Apatkaleen Matritva evam Navjat Tatparta mentoring programme as part of a larger quality improvement and health systems strengthening initiative. Public sector facilities were initially offered training, facilitated by trained nursing graduates, during 8-month phases between September 2015 and January 2017. Repeat training began in February 2018 and was facilitated by peers. NR skills in simulated resuscitations were assessed at the facility level at the midpoint and endpoint of initial training and prior to and at the midpoint of repeat training.Results Facilities administering effective positive pressure ventilation and assessing infant heart rate increased (31.1% and 13.1%, respectively, both p=0.03) from midinitial to postinitial training (n=64 primary health centres (PHCs) and 192 simulations). This was followed by a 26.2% and 20.9% decline in these skills respectively over the training gap (p≤0.01). A significant increase (16.1%, p=0.04) in heart rate assessment was observed by the midpoint of repeat training with peer facilitators (n=45 PHCs and 90 simulations). No significant change was observed in other skills assessed.Conclusions Despite initial improvement in select NR skills, deterioration was observed at a facility-level post-training. Given the technical nature of NR skills and the departure these skills represent from traditional practices in Bihar, refresher trainings at shorter intervals are likely necessary. Very limited evidence suggests peer simulation facilitators may enable such increased training frequency, but further study is required.