Lessons Learned from Evaluation of the Convergent Validity the SPARK36 in Preventive Youth Health Care: Implications for Further Research and Practice.

objectiveSPARK36 is a structured interview tool for nurse-led consultations with parents of 3-year-olds in Youth Health Care. Previous studies demonstrated its feasibility, reliability and known-groups validity. This study examined its convergent validity.

methodsIn a cross-sectional study during the COVID-19 period, associations between scores on 12 SPARK36 domains and conceptually related subscales of three parent-reported instruments were explored. Correlations were assessed using Kendall's Tau-C.

resultsOf the 599 parents of children attending a SPARK36 consultation, 286 (41.7%) did not return a questionnaire. These parents more often had lower educational levels, non-Belgian nationalities, less often spoke Dutch at home, and more often showed increased or high risk based on the SPARK36 (19.9% vs. 13.7% respectively; p = 0.04), indicating potential selection bias. None of 32 tested correlation coefficients (e.g., SPARK36 'child behavior' with SDQ scales) exceeded 0.20. CONCLUSION FOR PRACTICE: Convergent validity could not be demonstrated possibly due to the overall response rate, selective non-reporting, and the fact that most children showed no developmental or parenting problems. Three lessons emerged: (1) equity challenges occur, when parent-report tools are used among higher-risk families; (2) ongoing support may be needed to ensure consistent data collection by the nurses, given the large variation in data collection across them; (3) crisis situations disturb data collection, though a 50% questionnaires response rate may still be achievable during adverse contexts. The lessons also suggest that interview formats like SPARK36 may help reduce inequities that arise when preventive care relies on parent-report questionnaires.