Implementation fidelity of a multisite maternity waiting homes programme in rural Zambia: application of the conceptual framework for implementation fidelity to a complex, hybrid-design study.

Background: Implementation fidelity measures are critical for understanding complex interventions. Though maternity waiting homes (MWHs) are recommended by the WHO and have been used to help pregnant women access health facilities for decades, a gap exists regarding fidelity studies on MWHs. To better understand intervention outcomes results, we assessed the fidelity of implementation of an improved Core MWH Model in 10 facilities in rural Zambia. Methods: We analysed indicators for fidelity employing a widely used conceptual framework. We compared performance from October 2016 to July 2018 to goals set out during intervention design. The Core MWH Model consists of three pillars-infrastructure, policies and linkages to care-each designed to be culturally appropriate and responsive to community standards for safety and comfort. Results: 70% of sites exhibited high adherence to the Core MWH Model components. User experience corroborated poorer performing elements: insufficient lighting, small cooking spaces, non-locking cabinets and few educational classes. Mission statements and governing documents were not always visible or available. The proportion of 3206 users who came from>10 km away was higher than the proportion of the surrounding population living at that distance except in two sites with low populations of remote-living women. Women stayed for just below the target of 14 nights. MWH occupancy rate overall was lower (52%) than the target (80%). MWH users reported high quality and satisfaction. Only three MWHs reached 50% female membership on their governance committees but met other key indicators for community ownership and engagement. Conclusions: This fidelity evaluation of an MWH model in rural Zambia helps bridge the evidence-practice gap for the WHO's recommendation on implementing MWHs and adds to the expanding body of literature on implementation fidelity studies in global health. Trial registration number: NCT02620436.