'The development sector is a graveyard of pilot projects!' Six critical actions for externally funded implementers to foster scale-up of maternal and newborn health innovations in low and middle-income countries

Background: Donors often fund projects that develop innovative practices in low and middle-income countries, hoping recipient governments will adopt and scale them within existing systems and programmes. Such innovations frequently end when project funding ends, limiting longer term potential in countries with weak health systems and pressing health needs. This paper aims to identify critical actions for externally funded project implementers to enable scale-up of maternal and newborn child health innovations originally funded by the Bill & Melinda Gates Foundation ('the foundation'), or influenced by innovations that were originally funded by the foundation in three low-income settings: Ethiopia, the state of Uttar Pradesh in India and northeast Nigeria. We define scale-up as the adoption of donor-funded innovations beyond their original project settings and time periods.

Methods: We conducted 71 in-depth, semi-structured interviews with representatives from government, donors and other development partner agencies, donor-funded implementers including frontline providers, research organisations and professional associations. We explored three case study maternal and newborn innovations. Selection criteria were: a) innovations originally funded by the Bill & Melinda Gates Foundation ('the foundation'), or influenced by innovations that were originally funded by the foundation; b) innovations for which a decision to scale-up had been made, allowing us to reflect on the factors influencing those decisions; c) innovations with increased geographical reach, benefitting a greater number of people, beyond districts where foundation-funded implementers were active. Our data were analysed based on a common analytic framework to aid cross-country comparisons.

Results: Based on study respondents' accounts, we identified six critical steps that donor-funded implementers had taken to enable the adoption of maternal and newborn health innovations at scale: designing innovations for scale; generating evidence to influence and inform scale-up; harnessing the support of powerful individuals; being prepared for scale-up and responsive to change; ensuring continuity by being part of the transition to scale; and embracing the aid effectiveness principles of country ownership, alignment and harmonisation.

Conclusions: Six critical actions identified in this study were associated with adopting and scaling maternal and newborn health innovations. However, scale-up is unpredictable and depends on factors outside implementers' control.