Development and validation of the PROMIS Pediatric Sleep Disturbance and Sleep-Related Impairment item banks.

Abstract

To develop and evaluate the measurement properties of child-report and parent-proxy versions of the Patient-Reported Outcomes Measurement Information System (PROMIS) Pediatric Sleep Disturbance and Sleep-Related Impairment item banks. A national sample of 1104 children (8-17 years old) and 1477 parents of children 5-17 years old was recruited from an internet panel to evaluate the psychometric properties of 43 sleep health items. A convenience sample of children and parents recruited from a pediatric sleep clinic was obtained to provide evidence of the measures' validity; polysomnography data were collected from a subgroup of these children. Factor analyses suggested two dimensions: sleep disturbance and daytime sleep-related impairment. The final item banks included 15 items for Sleep Disturbance and 13 for Sleep-Related Impairment. Items were calibrated using the graded response model from item-response theory. Of the 28 items, 16 are included in the parallel PROMIS adult sleep health measures. Reliability of the measures exceeded 0.90. Validity was supported by correlations with existing measures of pediatric sleep health and higher sleep disturbance and sleep-related impairment scores for children with sleep problems and those with chronic and neurodevelopmental disorders. The sleep health measures were not correlated with results from polysomnography. The PROMIS Pediatric Sleep Disturbance and Sleep-Related Impairment item banks provide subjective assessments of child's difficulty falling and staying asleep as well as daytime sleepiness and its impact on functioning. They may prove useful in the future for clinical research and practice. Future research should evaluate their responsiveness to clinical change in diverse patient populations.