Globally, 2.9 million newborn infants die within the first 28 day of life and 2.6 million babies are born dead, 1.3 million of these being alive at the onset of labor. Newborn health is part of the "unfinished agenda" and requires greater visibility in the post-2015 agenda and is a key priority, and a direct indicator of progress of the SDG's which sets out a vision of a world in which there are no preventable deaths of newborns or stillbirths, where every birth is celebrated and babies not only survives, but thrives to reach their full potential. There is an urgent need for research strengthening neonatal care in low recourse settings, which the NEO-study aims to contribute to.

The aim of the NEO-study is to understand the direct and underlying determinants of sub-standard care and improve the quality of care using innovative technologies such as video recordings and animated clinical videos to strengthen decision making and management of emergencies in newborns.

Study design This is the study protocol for a 14-month quality improvement study involving all district level hospital and cottage level hospitals in Pemba, Zanzibar.

Methodology The first part of the study is a 10-week observational baseline where all district level hospital deliveries are included, and data collected about the pregnancy, delivery and delivery outcome.

The intervention is a 9-month period where we will facilitate a quality improvement cycle using Low Dose High Frequency training with the Safe Delivery App as an anchor point and facilitate the integration into clinical practice through the Safe Delivery Focal Points at each hospital. All staff in maternity, pediatric wards and outpatient clinics will receive training and be encouraged to use the Safe Delivery App on a weekly basis.

After the end of the intervention period the investigators will repeat the 10-week observational study in the same months the following year as the baseline study and the findings will be used to measure adherence to guidelines, quality of care and the impact on perinatal and neonatal morbidity and mortality.

The study population for the primary endpoint are all newborns and their mothers who will be delivered in one of the district or cottage hospitals and all newborns admitted to either the maternity or pediatric departments.

Time frame From September 2019 to October 2020.

Expected outcomes The NEO-study is anticipated to improve quality of care and significantly decrease perinatal and neonatal mortality.

The aim of the NEO study is to understand the direct and underlying determinants of sub-standard care and improve the quality of care using innovative technologies to strengthen decision making and management of newborn emergencies. The NEO-study builds on and add to established mHealth initiatives for maternal and child health in Zanzibar, Tanzania and Denmark.

This is the study protocol for a 14-month quality improvement study involving all district level hospitals and one cottage hospital (Chake Chake, Mkoani, Wete, Micheweni) in Pemba, Zanzibar. The protocol is developed in equal partnership between the research team from Public Health Laboratory, Ivo de Carneri and the research team from University Hospital Copenhagen (Rigshospitalet). The partnership between the two institutes have been long-term and long-standing for more than a decade. The first exploratory visits for the NEO-study was conducted in February 2016 and October 2017. On both visits the investigators held fruitful meetings with managements at all district level hospitals and the relevant departments; maternity, paediatric and outpatient clinics to ensure participatory commitment and involvement in the process and development of the project. The principals of the Paris Declaration on Aid Effectiveness are applied to the presented research protocol and the investigators have included the needs expressed by the local health system and changed the protocol in accordance with requests and necessities expressed [1]. The investigators have raised funds between the first exploratory meetings and the study has the necessary financial support from national and international donors. The investigators are continuing to raise funds for future collaborations and extension and expansion of the NEO-study.

The final version of the protocol has been developed by the international and local research teams during a three-month research stay at Public Health Laboratory, Ivo de Carneri, Pemba from February to May 2019. In this period, the investigators conducted meetings, hearing, direct and indirect observations and pilot-testing of tools and feasibility testing of method of choice to ensure participatory involvement and continuous support from hospital managements at all district level hospitals. Furthermore, the investigators have sought to involve clinical staff at relevant maternity, paediatric and outpatient departments by informal conversations during observatory days and the investigators will make continuous efforts of staff involvement; at all levels to create local ownership. The investigators have identified a local focal person at each department who can support the implementation and be part of the change. The investigators are exploring new ground with novel methods to capture and improve the quality of care, and therefore the investigators have pre-tested the feasibility, acceptability, technicality and limited-efficacy testing in the "Feasibility Study - The efficacy of Video Analysis as a Supplementary objective tool to Criterion Based Clinical Audit (CBCA) in Newborn Emergencies" (NO.ZAREC.02/APR/2019/20)

The investigators hope this research proposal can add to the unfinished agenda of neonatal mortality and morbidity and support health workers who are assisting women and their babies during labour, delivery and in the first month of life.

Inclusion Criteria:

• All newborns and their mothers delivered at Wete District Hospital, Chake Chake District hospital and Mkoani District Hospital or Micheweni Cottage Hospital.

All newborns <1 month admitted to either the maternity or paediatric department

Exclusion Criteria:

• None

De-identified individual participant data for all primary and secondary outcomes will be made available.

The data will be available for the research team from the end of the baseline until two years after the end of the study,