Psychometric properties of the REACh developmental screening tool for preschool children in Sri Lanka.

AIM: To design a culturally appropriate and cost-effective developmental screening tool with better psychometric properties to assess preschool children aged 3 to 5 years.

METHOD: A cross-sectional study was conducted using the second iteration of the Ragama Early Assessment of Children (REACh) tool. Item analysis and revisions were confirmed with consensus agreement. It was piloted and later administered by 28 teachers. The accuracy of findings by teachers and the tool's internal consistency were evaluated. Concurrent validity was confirmed with criterion-standard assessments by a developmental paediatrician. Psychometric properties were assessed using the Kuder-Richardson 20 (KR-20) and kappa statistics. Cut-offs for fine motor, gross motor, receptive language, expressive language, social-emotional, and pre-academic were derived using receiver operating characteristic curve analysis.

RESULTS

The tool was administered on 597 children. The accuracy of findings by teachers was 100% for all domains. All KR-20 values for 3- to 4-year-olds were greater than 0.7 except for fine motor (0.629). All KR-20 values for 4- to 5-year-olds were greater than 0.7 except fine motor (0.683) and social-emotional (0.694). The sensitivities for all domains were greater than 0.7 except for receptive language and social-emotional in the 3- to 4-year-olds and for expressive language and pre-academic in the 4- to 5-year-olds.

INTERPRETATION: Acceptable sensitivity for most items in this iteration of the REACh will enable screening and referral to services. The findings suggest that the REACh can be adapted for use in South Asian settings and is suitable for implementation in other low-resource settings.