Measuring feasibility: complex questions need complex tools.

We sought to assess the feasibility of mainstreaming balanced energy protein supplementation, a maternal nutrition intervention, into Bangladesh's routine antenatal care system in tandem with an ongoing effectiveness trial in northwestern Bangladesh. Feasibility is an implementation science outcome defined as the extent to which a new intervention can be implemented successfully in a given context. We found feasibility difficult to measure using existing Anglophone quantitative tools translated into Bangla and identified particular challenges with using Likert scales. We also found it challenging to measure feasibility early in implementation, as recommended in the implementation science literature, due to potential respondents' unfamiliarity with the intervention and what implementation would look like. To address these issues, we explored alternative assessment methods, such as focus groups and workshops. These formats presented additional difficulties, including identifying the correct participants and moderating participant power dynamics. In conclusion, we question whether existing feasibility assessment tools, developed in English, are suitable for use in non-Anglophone contexts as well as whether Likert scales are appropriate for use in low-income and middle-income countries more broadly. We also question current recommendations on the timing of feasibility assessments. We feel that-particularly with new, difficult-to-conceptualise interventions-feasibility must be assessed later in implementation or only after providing detailed explanations of the intervention to respondents.