Menu
Menu
Menu
Menu
Background: Donors commonly fund innovative interventions to improve health in the hope that governments of low and middle-income countries will scale-up those that are shown to be effective. Yet innovations can be slow to be adopted by country governments and implemented at scale. Our study explores this problem by identifying key contextual factors influencing scale-up of maternal and newborn health innovations in three low-income settings: Ethiopia, the six states of northeast Nigeria and Uttar Pradesh state in India. Methods: We conducted 150 semi-structured interviews in 2012/13 with stakeholders from government, development partner agencies, externally funded implementers including civil society organisations, academic institutions and professional associations to understand scale-up of innovations to improve the health of mothers and newborns these study settings. We analysed interview data with the aid of a common analytic framework to enable cross-country comparison, with Nvivo to code themes. Results: We found that multiple contextual factors enabled and undermined attempts to catalyse scale-up of donor-funded maternal and newborn health innovations. Factors influencing government decisions to accept innovations at scale included: how health policy decisions are made; prioritising and funding maternal and newborn health; and development partner harmonisation. Factors influencing the implementation of innovations at scale included: health systems capacity in the three settings; and security in northeast Nigeria. Contextual factors influencing beneficiary communities’ uptake of innovations at scale included: sociocultural contexts; and access to healthcare. Conclusions: We conclude that context is critical: externally funded implementers need to assess and adapt for contexts if they are to successfully position an innovation for scale-up.
We adopted a health policy analysis approach informed by the stages heuristic framework [13] that identifies sequential stages in the policy process: agenda setting, policy formulation and policy implementation, and on the literature on scale-up and context to frame different contextual domains. From this we developed a framework to guide our study consisting of three distinct stages that are critical to scale-up: Having used these categories to develop a topic guide, researchers from Nigeria, Ethiopia, India and the UK piloted it at a workshop in Addis Ababa leading to minor adaptations being made to reflect different country contexts. Researchers used the guide to conduct semi-structured interviews with purposively selected stakeholders working in MNH, or having substantial experience and/or knowledge of issues relating to scale-up of MNH innovations including policy, financing and health systems issues. The interviewees were drawn from different sectors: government, development partners, civil society organisations (CSOs) including implementers of donor- funded MNH programmes, academic institutions and professional associations. Interviewees were managers and directors, programme officers and research and evaluation and technical officers. Fifty interviews were conducted in each of the three settings between July 2012 and April 2013. Our sample of interviewees represent the majority of implementer and development partner organisations working on MNH in each of our three settings. We have deliberately not named specific organisations in our paper because of our commitment to maintaining respondent confidentiality. The MNH implementers we sampled are characterised as follows: the majority were large international nongovernmental organisations or large local nongovernmental organisations, together with a smaller number of US-based universities and for-profit consultancy companies implementing MNH programmes. Most of these implementers had in the past received large grants from different donors to maintain particular interventions and some were receiving multiple grants for separate pieces of work at the time of the interviews. Many implementers also worked with smaller local CSOs to implement work packages in particular locations. While a substantial amount of externally funded MNH-related work in the three settings took the form of projects to develop innovative interventions, some implementers also received donor funding for direct technical support to government agencies as well as advocacy work. The development partners we sampled included a mix of donors – bilateral agencies and philanthropic foundations – and UN agencies, some of which also funded MNH innovations. In addition to funding projects some development partners also contributed to larger health programmes, provided technical support for government, and in Ethiopia contributed to a pooled fund for work corresponding to the Millennium Development Goals. The MNH projects we explored in our interviews generally lasted up to five years and more commonly three to four years. The scale varied from a small handful of districts to several districts across multiple states or regions, and some were part of large multi-country grants. Some projects involved single innovations, while others involved a package of connected innovations. The interviewers included NS, RDTG, DBh and ATW, and other researchers with training in qualitative methods. The interviews were conducted in private spaces to preserve confidentiality and all respondents gave informed consent before the interview. Where it was agreed with the respondent, a sound recorder was used for data capture. Interviewers wrote ‘expanded field notes’ [6] shortly after the interview comprising detailed notes arranged under thematic headings, with direct quotes to illustrate respondents’ voices. Through simultaneously capturing and analysing data, interviewers identified emerging interpretations and hypotheses to explore in ensuing interviews. We adopted several steps to maximise the validity of our findings. We adopted an investigator triangulation approach to compare and agree researchers’ interpretations; this helped reinforce the validity of the results reported because each set of expanded field notes was the work of multiple researchers. Moreover, an analysis workshop enabled us to reach consensus on interpretations among researchers involved in the study and cross-country comparisons. Our relatively large qualitative sample, with interviewees from a variety of organisations, helped balance the views we present, and cross-checks of interviewees’ views enabled us to triangulate findings. We also conducted member checks: we presented emerging findings to interviewees and other relevant country stakeholders in Addis Ababa, Abuja and Lucknow who were invited to comment on the accuracy of our messages. The analysis of the interview data was undertaken in five stages: 1) an analysis workshop in London at which NS, DW ATW, RD and DBh reviewed and agreed emerging findings and developed an analytic framework to enable us to directly compare our three study settings; 2) using Nvivo Version 10, NS and DW analysed the expanded field notes, using a framework approach to code a priori and emerging themes; 3) the analytic framework was used to organise the emerging themes ; 4) NS drafted the paper, which was then reviewed by all authors to ensure that the findings are represented coherently and accurately. In order to maintain anonymity of our interviewees it is not appropriate to make the qualitative dataset supporting the conclusions of this article publically available.
N/A
