Validating self reported home safety practices in a culturally diverse non-inner city population

Objective: To determine the validity of face to face, self reported responses to questions about the presence of safety devices and use of safety practices in the home aimed at preventing unintended injuries to preschool aged children. Methods: The authors invited families with children enrolling in a medium sized Midwestern US community Head Start program to participate in a randomized study of home safety practices. Participation involved consenting parents (n = 452) completing an initial questionnaire during Head Start enrollment or in their home. Project staff conducted home inspections to confirm parental responses to 16 questions. Only inspections conducted within 34 days of questionnaire completion (n = 259) were included in the validation study. Parents were told that the home visit would assess the need for safety devices, but were not informed of the validation aspect of the study. Results: Sensitivities were generally high for all 16 safety practices, whereas negative predictive value and specificity varied considerably. Positive predictive value was also high for most practices, and did not vary by ethnicity. Answers provided by parents in their home were different and more reliable than those provided by parents interviewed at school (p = 0.001). Conclusions: Use of safety devices and practices by parents of preschool aged children reported in a face to face interview are generally reliable. Reliability increases if the interview is conducted in the home. Parents may also be more willing to report potential problems if they perceive they may receive corrective assistance.