Measuring the psychosocial wellbeing of Rohingya mothers in Cox’s Bazar, Bangladesh: Psychometric properties of an MHPSS assessment battery

Abstract

Background: Psychosocial research in humanitarian settings has been historically dominated by a focus on distress and disorder. As such, there is a need to establish the validity of instruments for a broad range of psychosocial outcomes, particularly among highly affected and under-represented populations. The current study describes the adaptation and testing of multiple psychosocial instruments among displaced Rohingya mothers in Bangladesh.  Methods: We used baseline data from 600 mothers of malnourished children aged 2 and under enrolled in an intervention study in Cox’s Bazar, Bangladesh. Instruments assessed distress (International Depression Symptom Scale [IDSS] and Kessler-6 [K-6]); functional impairment (World Health Organization Disability Assessment Schedule [WHODAS]); subjective wellbeing (global Satisfaction With Life [SWL] and Personal Wellbeing Index [PWI];  and, coping (Brief COPE and locally developed coping items). Instruments were piloted and refined, then used for data collection by Bangladeshi-Rohingya interviewer pairs. We conducted exploratory factor analysis, evaluated internal consistency, examined construct validity through correlation with other scales, and used regression models to explore demographic factors associated with psychosocial health.  Results: Both the WHODAS and coping items fit 2-factor models; other scales were unidimensional. Cronbach’s alphas ranged from .76 to .90 for the refined scales. With the exception of coping, scale correlations supported construct validity; separate measures of the same construct were highly correlated, distress and impairment were moderately correlated, and both were inversely correlated with wellbeing. Correlates of poorer psychosocial health included relative socioeconomic disadvantage, current pregnancy, and being unmarried.   Conclusions: Most of the standard psychosocial assessment tools performed adequately, but they did not appear to fully capture local experiences and included items of little local relevance. Findings highlight the need for further mixed methods research to develop a rich battery of instruments with cross-cultural validity, particularly for positive outcomes such as coping which was particularly challenging to assess.