Background Bangladesh has adopted the WHO framework on Maternal and Newborn care. USAID’s MaMoni MNCSP works to improve maternal and newborn care across the public and private health systems. Project baseline study found about poor health facility readiness (3%) for quality antenatal care, 9% of facilities had partograph available, and about 20% of providers were trained on ANC and delivery care. In 2019, Model for Improvement (MFI), a method for structuring an improvement project was introduced. Altogether, facilities implemented and tested Maternal Newborn Health (MNH) clinical bundles comprising 73 improvement projects on quality ANC, correct partograph use, (CPU) and Essential Newborn Care (ENC) using the PDCA approach. A visual display board reflect facility performance and periodic review.

Objectives To describe the effectiveness of ‘Model for Improvement’ in improving MNH clinical bundles under learning network model.

Methods Manikganj district was selected as project intervention district. A total of 31 diversified facilities has joined this network to learn about improvement methodologies and its implication. The ‘MFI’’ framework were used for continuous improvement. This model helped defined their aim, ideas, and measurement process. Also the PDCA (Plan Do Check Act) cycle outlined the steps for the actual testing of the change ideas. A comparative analysis of selected MNH indicators was done between baseline survey (2019) and service statistics (2020).

Results About 200 health managers and service providers were trained on different clinical themes and leadership. Supportive supervision was deployed. Thus resulted improvement in quality ANC from 20% to 85%, CPU from 26% to 95% and quality ENC from 9% to 86%.

Conclusions Such initiative in Bangladesh has facilitated innovation both within and across facilities. This network of care can rapidly achieve significant improvements in quality health care. With results from Manikganj, project is scaling up MNH clinical bundles in other districts.