Measuring the quality of preventive and developmental services for young children: national estimates and patterns of clinicians' performance.

OBJECTIVE: To generate a national picture of performance in the area of preventive and developmental services for children aged 4 to 35 months using 4 composite quality measures in the areas of 1) anticipatory guidance and parental education, 2) screening for family psychosocial risks, 3) screening for smoking and drug and alcohol use in the home, and 4) provision of family-centered care.
METHODS: Data from the National Survey on Early Childhood Health (N = 2068) were used to calculate the 4 composite performance measures, which, taken together, represent 23 topics included in the American Academy of Pediatrics health supervision guidelines. The reliability and degree of redundancy within and across these 4 measures were evaluated. Four methods for scoring these measures were used. Quality scores for subgroups of children were calculated, and logistic regression analysis was performed to examine the association of demographic, health, and health system variables with receiving recommended care.
RESULTS: Regardless of the scoring method used, performance is highest in areas of family-centered care and screening for smoking and drug and alcohol use in the home. Performance is lowest in the areas of anticipatory guidance and education and assessment for family psychosocial risks. Using a scoring method that takes into account parent preferences for guidance and beliefs about discussing psychosocial topics, composite quality measure scores ranged from 13.5% to 59.6% of parents of young children receiving recommended care. Overall, 94.0% of parents reported 1 or more unmet needs for parenting guidance, education, and screening by pediatric clinician(s) in 1 or more of the content of care areas evaluated. Uninsured children and children aged 18 to 35 months are disproportionately represented among the 15.3% of children whose parents indicated an unmet need in each of the 4 areas of care. Although the reliability of each composite measure was high, no single item in any composite was highly correlated with the remaining combined items. Performance on any 1 composite measure for a child was only somewhat predictive of performance for the other measures. There are significant variations in performance on the basis of child age, race, insurance status, maternal education, marital status, and parent language as well as other factors.
CONCLUSIONS: National results using 4 complementary composite quality measures confirm the need for improving the quality of preventive and developmental services for young children in the United States. The 4 measures identify areas of care and subgroups of children for whom improvements in quality are most needed. The measures provide a parsimonious yet comprehensive assessment across distinct health supervision topics and 4 essential aspects of preventive and developmental services. Until valid measures of outcomes of preventive and developmental services are identified or 1 single process of care measure is shown to be highly predictive of these outcomes, assessing multiple aspects of recommended care will be necessary to assess performance of health care providers or systems of care.