Seasonality in the access and availability of nutritious diets throughout the year is a major challenge that undesirably affects low-income consumers in developing countries. However, it is known that properly functioning physical infrastructure such as roads, electricity, water supply systems and marketplaces can mitigate the adverse effects of seasonality on the availability, accessibility, and affordability of nutritious foods. This ensures that low-income consumers have access to nutritious diets (ND) throughout the year. Moreover, gender-responsive infrastructural development could mitigate gender-based barriers that hamper women and girls’ access to basic services. This can help to alleviate socio-cultural and systemic inequalities against women and girls, thereby contributing to women’s economic empowerment (WEE), and gender equality (GE). Consequently, the International Center for Evaluation and Development (ICED) has developed an evidence and gap map (EGM) on physical infrastructure’s impact on nutritious diets and gender outcomes. The EGM included studies from published and grey literature sources which used various methods for data analysis. Hence, it is critical to identify the best practice methods for generating useful evidence on the impact of physical infrastructure in low-and-middle income countries. This study, therefore, presents results from the synthesis of the estimation strategies an approach employed by the 285 primary studies included in the EGM. Additionally, the best-practiced methods used in evaluating the impact of physical infrastructure on the outcomes of interest were identified through experts’ recommendation. A coding form was developed using Eppi-reviewer software to capture information on methods of data collection and analysis of the included studies in the EGM. The findings show that the majority (67%) of the primary studies employed only quantitative methods, 11% used qualitative study design, and 22% used mixed methods. Regression-based methods were most used (48%), followed by descriptive studies (38%) and quasi-experimental methods (14%). Within regression-based methods, binary models (32%) and ordinary least squares regression (23%) were frequently employed. Thematic analysis (18%) and narrative/exploratory analysis (22%) were frequently employed for qualitative data analysis. Primary data sources were used (69%), with surveys being the most common data collection method. Findings and experts’ recommendation on best practice methods allow researchers, policymakers, funders, and implementers of infrastructure intervention to consistently generate high quality evidence for decisionmaking on infrastructural interventions.
