Evaluation of Healthy Term Infants to Define Impairment Thresholds and Promote Unbiased Assessments of High Risk Infants: A Quality Assurance measure for Network Follow-Up Assessments (Short Name: Term Reference Study)

Methods Study design: Cohort study of extremely preterm and healthy term infants with a similar eDC. Population: as assessed from the nursery record, eligibility criteria for the healthy term comparison group must be met and no exclusion criteria included (see Protocol for inclusion and exclusion criteria. enrollment ratios and procedures. While the appropriate ratio may change over time, we would currently project assessing one healthy term infant for every five high-risk generic follow-up infants in the first year and for ten high risk generic follow-up infants in succeeding years. Blinding: Standards and feasibility. Prior to the assessment and recording of findings, examiners should not look at the medical records and should not have seen the infant for the preceding six (preferably twelve) months. Data collection. all term reference evaluated in the first two years of the proposed study will require identification and data-gathering for the neonatal period prior to contacting the care-giver 1-2 months before the corresponding preterm (index) infants are to be seen for the 22-26 follow up visit). analyses. Descriptive analyses characterizing point and interval estimates of the distribution of Bayley-iii scores will be performed. inferential analyses for continuous and dichotomous outcomes will utilize generalized linear mixed models to account for correlations induced by center. The expected confidence intervals for the mean Bayley iii scores are noted above for the proposed sample size of patients in the initial and subsequent years. 1. Hypothesis 1 (among high-risk infants the ascertainment of adverse outcomes based on the standard deviations estimated from a concurrent, healthy term sample will yield a higher proportion of infants with neurodevelopmental delay than adverse outcomes ascertained from Bayley-iii, manual-based norms). 2. Hypothesis 2 (Comparison of the initial sample [n [?] 180] with subsequent samples [n [?] 90] will identify drift greater than or equal to 7.5 points (0.5 standard deviations) on the Bayley-iii Composite Scales). Safety monitoring: n/a SeConDaRY STuDY For infants enrolled into the main study, the secondary study is asking for the The Child Behavior Checklist 1.5-5 years( CBCL) to be administered at the 22-26mo visit. The CBCL is a questionnaire, widely used method of identifying problem behavior in children. Problems are identified by a respondent who knows the child well, usually a parent or other care giver. Baseline group comparisons of child and maternal characteristics will be performed using t-tests or Wilcoxon tests for continuous variables and chi-square tests for categorical variables. For these analyses, behavior scores in the borderline and in the clinical range will be combined into one category. Variables that demonstrate a relationship to behavioral scores with be included, along with an indicator for preterm group, in logistic regression models to examine their effects on behavior scores. The effect of borderline/clinical behavior scores on Bayley scores will be estimated using linear regression, with an indicator ([Quote]dummy[Quote]) variable for behavior score in the borderline/clinical range, and adjusting for group differences/possible confounding child and maternal characteristics. Because of possible/likely collinearity between behavioral scores, each behavioral score used as a predictor will be tested in a separate model to estimate its effect without collinearity The sample size that has been approved is 160 term infants compared to 160 preterm infants. in addition study term CBCL data will be compared to the published aCeBa CBCL normative data. Means and SD of scores will be compared with t tests. Borderline and clinical cut-points for the study cohort will be compared categorically with the aCeBa normative population data cut-points.