RESUMO
BACKGROUND: There is a global drive to promote facility deliveries but unless coupled with concurrent improvement in care quality, it might not translate into mortality reduction for mothers and babies. The World Health Organization published the new "Standards for improving quality of care for mothers and newborns in health facilities" but these have not been tested in low- and middle-income settings. UNICEF and its partners are taking the advantage provided by the Mother and Baby Friendly Hospital Initiative in Bangladesh, Ghana and Tanzania to test these standards to inform country adaptation. This manuscript presents a framework used for assessment of facility quality of care to inform the effect of quality improvement interventions. METHODS: This assessment employed a quasi-experimental design with pre-post assessments in "implementation" and "comparison" facilities-the latter will have no quality improvement interventions implemented. UNICEF and assessment partners developed an assessment framework, developed uniform data collection tools and manuals for harmonised training and implementation across countries. The framework involves six modules assessing: facility structures, equipment, drugs and supplies; policies and guidelines supporting care-giving, staff recruitment and training; care-providers competencies; previous medical records; provider-client interactions (direct observation); and client perspectives on care quality; using semi-structured questionnaires and data collectors with requisite training. In Bangladesh, the assessment was conducted in 3 districts. In one "intervention" district, the district hospital and five upazilla health complexes were assessed. similar number of facilities were assessed each two adjoining comparison districts. In Ghana it was in three hospitals and five health centres and in Tanzania, two hospitals and four health centres. In the latter countries, same number of facilities were selected in the same number of districts to serve for comparison. Outcomes were structured to examine whether facilities currently provide services commensurate with their designation (basic or comprehensive emergency obstetric and newborn care). These outcomes were stratified so that they inform intervention implementation in the short-, medium- and long-term. CONCLUSION: This strategy and framework provides a very useful model for supporting country implementation of the new WHO standards. It will serve as a template around which countries can build quality of care assessment strategies and metrics to inform their health systems on the effect of QI interventions on care processes and outcomes.