Attention problems in children born very preterm: evidence from a performance-based measure.

Background: Children born very preterm (VPT) are at high risk for attention problems. This study's purpose was to describe the Conners Kiddie Continuous Performance Test (K-CPT) assessment in children born VPT, including rates of clinically elevated scores, change over time, and associations between K-CPT scores and parent reported attention problems.

Methods: We studied 305 children from a multi-site study of children born VPT who completed at least one K-CPT assessment at age 5, 6, and/or 7 years. Parent-reported ADHD symptoms and diagnosis were also collected. We calculated K-CPT completion rates, mean scores, and rates of clinically elevated scores at each timepoint. Linear mixed models examined change over time in K-CPT scores. Correlations and generalized linear models investigated associations between K-CPT scores and ADHD symptoms and diagnoses.

Results: K-CPT scores showed expected age-related improvements from age 5-7, with significant intra- and inter-individual variability. Up to 1/3 of children had clinically elevated attention problems and another 1/3 had subclinical elevations. K-CPT scores were modestly correlated with parent-rated ADHD symptoms and children with a parent-reported ADHD diagnosis performed worse on nearly all K-CPT metrics.

Conclusion: Performance-based measures like the K-CPT can be useful for research and clinical practice in VPT populations.

Impact: Attention problems are a specific area of weakness for children born very preterm. Performance-based tests of attention have benefits and drawbacks compared to parent report measures yet are understudied in this population. We examined one performance-based measure (the Conners Kiddie Continuous Performance Test [K-CPT]) in 305 children born very preterm. We observed improving task scores from age 5-7 years with significant intra- and inter-individual variability, a sizable proportion of children with clinically and subclinically elevated scores, and modest associations between K-CPT scores and parent reported attention problems. The K-CPT could be a useful clinical and research tool in this population.