Non-governmental organisations (NGOs) are key stakeholders in public health programmes in low resource settings. Despite known evidence-based interventions for improving maternal and child health (MNCH) and HIV outcomes, there remains an inequitable distribution of preventable and treatable maternal and infant morbidity and mortality within and between countries and regions globally. Through a scoping review and two published case studies, the aim of this research paper style thesis was to explore and critically reflect on how the use of theories, models and frameworks (TMFs) and mixed-method research approaches within NGO-implemented programmes can yield improved estimates of service uptake, clinical outcomes, and understanding of reasons for and consequences of failure to uptake evidence-based prevention of mother to child transmission (PMTCT) of HIV and MNCH services in Zimbabwe. Scoping review findings demonstrate that NGOs are central actors in the HIV response as catalysts of change through advocacy and activism, direct service delivery providers and technical partners in low resource settings. Limited evidence of application of scientific methods by NGOs through transdisciplinary research collaborations indicate missed opportunities for strengthening evidence-based practice in context. In the first case study, an observational, multisite, cohort study used sampling-based methods to actively trace a random sample of 371 HIV positive mothers in Mashonaland East Province Zimbabwe with no documented uptake of recommended Early Infant Diagnosis (EID) HIV testing for their HIV-exposed infants. Corrected estimates of EID uptake following tracing almost doubled, from 31.2% of infants with documented EID in routine records to a cumulative incidence of EID with death as a competing risk of 60.0% (95% CI: 58.7% to 61.3%) after contact tracing. Findings highlight significant underreporting of both infant HIV testing and mortality in routine data sources. The second case study explored the lived experience of ‘non-adherers’ to recommended HIV/MNCH services. A qualitative study among women with a recent non-institutional birth was conducted with 71 women (30 in-depth interviews; 41 in focus group discussions) who had delivered at home in the previous 6 months in Mashonaland Central Province, Zimbabwe. Narratives of women’s birth and postpartum experiences following a home delivery highlighted a coalescing of structural inequality and vulnerability at the time of delivery which prevented access to institutional delivery. Social norms for ‘facility delivery for all’ produced through health promotion initiatives to increase skilled attendance at birth, created unintended social consequences (shaming, blaming) and punishments (fines, reduced quality of healthcare) for home delivery which introduced additional risk to women at the time of delivery and in the postnatal period. Women’s narratives underscore the importance of understanding social norms, contextual vulnerabilities and potential unintended consequences of health policy translation on the most vulnerable who are unable to comply to recommended public health practice. Embedding mixed-method research within routine NGO programmes in this thesis yielded improved estimates of service uptake and outcomes; provided additional insights into the reasons for suboptimal uptake of PMTCT/MNCH services among rural women in Zimbabwe; and, led to the expansion of existing TMFs. The scoping review and case study findings provided critical insights to inform adaptation of NGO programme strategies that have improved programme performance and been adopted into national policy and programmes at scale in Zimbabwe. Transdisciplinary programme science embedded within NGO 7 programmes has the potential to optimise coverage and effectiveness of evidence-based interventions that are equitable and contextually relevant.