Measuring Psychosocial Adversity as a Neurodevelopmental Risk Factor: Development & Validation of the Childhood Psychosocial Adversity Questionnaire–Bangladesh (CPAQ-B)

Abstract

BACKGROUND: Research examining effects of psychosocial adversities on child neurocognitive development has occurred largely in high-income countries. A key barrier to global extension of this work has been a lack of culturally appropriate measurement tools. This thesis reports on the development, validation, and initial application of the Childhood Psychosocial Adversity Questionnaire–Bangladesh (CPAQ-B), the first context-tailored tool to assess early childhood psychosocial adversity comprehensively as a neurodevelopmental risk factor in a low-resource setting. PARTICIPANTS: 355 mother-child dyads from an urban slum of Dhaka, Bangladesh took part in qualitative work (N=45), item pretesting (N=25), questionnaire piloting (N=53), and administration of the CPAQ-B (N=232) at child ages 18, 24, 48, and/or 60 months. Community-based clinicians (N=35) also participated in qualitative sessions. METHODS: A conceptual model of early psychosocial adversity was developed to inform item selection. Initial items underwent expert review, pretesting, piloting, and item reduction. The CPAQ-B was then administered for psychometric analysis and validation including assessment of internal consistency, test-retest and inter-rater reliability, and criterion and predictive validity via correlation with parallel measures and child IQ scores. Finally, as an initial use of the tool, multivariate ordinary least squares regression modeling was used to assess whether 48-month CPAQ-B scores predict 60-month child intelligent quotient (IQ) scores when controlling for concurrent risks. RESULTS: 180 initial items were generated and piloted, and reduced to 64 items in the final scale. Exploratory factor analysis supported 11 subscales assessing for child maltreatment, various maternal and family stressors, and neighborhood-level exposures. Evidence supported strong subscale internal consistency and full-scale test-retest and inter-rater reliability. Subscale and full-scale scores correlated significantly with comparator measures. 48-month CPAQ-B scores significantly predicted 60-month WPPSI-IV IQ in multivariate regression models controlling for socioeconomic variables and malnutrition. CONCLUSIONS: The CPAQ-B represents a novel research tool measuring childhood psychosocial adversities, with good initial validity evidence for use among low-SES children ages 18-60 months in Dhaka, Bangladesh. Future work may adapt the instrument for use in other countries in the region, while methods may inform similar efforts to develop context-tailored assessments across other socio-cultural settings.