Cultural adaptation and psychometric assessment of acceptability, appropriateness, and feasibility measures for family planning interventions among midwives and nurses in Ghana.

Abstract

Background: Implementation research relies on accurate measurement of early implementation outcomes such as acceptability, appropriateness, and feasibility. However, instruments developed in high income settings may demonstrate limited cross-cultural validity when applied in different health systems. In postpartum family planning, differences in counselling routines, workload, and service organisation mean that unadapted measures can overlook important aspects of provider experience. This study aimed to culturally adapt and psychometrically evaluate the Acceptability of Intervention Measure, the Intervention Appropriateness Measure, and the Feasibility of Intervention Measure when applied to postpartum family planning counselling and use of the World Health Organization Medical Eligibility Criteria mobile application among midwives and family planning nurses in Ghana.

Methods: A sequential exploratory mixed methods design was employed. Three focus group discussions with 18 midwives and family planning nurses were conducted to culturally adapt the original measures developed by Weiner and colleagues. Insights from the qualitative phase informed revisions to the survey instrument. The adapted tool was subsequently administered to 150 midwives who had used the Medical Eligibility Criteria mobile application during one-on-one postpartum family planning counselling. Psychometric evaluation included assessment of internal consistency reliability, exploratory and confirmatory factor analysis, predictive validity, and tests of convergent and discriminant validity.

Results: The adapted implementation outcome measures demonstrated strong internal consistency, with Cronbach alpha coefficients ranging from 0.82 to 0.93 across the three constructs for both the counselling intervention and the mobile application. Confirmatory factor analysis indicated acceptable model fit following minor modifications, providing supportive evidence for the proposed factor structure of the acceptability, appropriateness, and feasibility constructs. Predictive validity was demonstrated through significant associations between implementation outcome scores and providers' preferences regarding counselling practices and use of digital tools. Although the expanded adapted scales showed high reliability, only the abridged versions consisting of four acceptability items and three items each for appropriateness and feasibility demonstrated full convergent and discriminant validity.

Conclusion: Culturally adapted versions of the Acceptability of Intervention Measure, Intervention Appropriateness Measure, and Feasibility of Intervention Measure can be applied with acceptable reliability to evaluate postpartum family planning interventions among midwives and family planning nurses in Ghana. The findings suggest that the abridged versions of these measures retain stronger construct validity than expanded adapted versions, highlighting the importance of balancing contextual adaptation with preservation of the underlying measurement structure.