Exploring the Congruence of actigraphy and the Pediatric Autism Insomnia rating Scale

Abstract

Objective/Background

Insomnia is common in children with autism spectrum disorder (ASD). We recently developed and validated the 21-item Pediatric Autism Insomnia Rating Scale (PAIRS). This report explores the associations and agreements between actigraphy and PAIRS.

Participants Methods

Children with ASD, with and without sleep problems, were assessed with a battery of parent-rated and clinician measures (N = 134). In a subset (n = 70), a wrist-worn actigraph measured sleep for five consecutive nights. Parents completed logs for scoring sleep intervals. Spearman correlations evaluated associations with the PAIRS and actigraphy indices (sleep onset latency = SOL, wake after sleep onset = WASO, total sleep time = TST, sleep efficiency = SE%). Agreements on “poor sleepers” based on PAIRS total score (≥33) and conventional thresholds for TST and SE% were evaluated with Cohen's Kappa and McNemar's test.

Results

Actigraphy data were averaged over 4.64 ± 0.68 nights in 70 children (mean age = 7.3 ± 2.9, 74.3 % male). There were no significant correlations between PAIRS and any actigraphy indices. On TST, 48.6 % (n = 34) and on SE% 52.9 % (n = 37) were classified as “poor sleepers” compared to 32.9 % (n = 23) on PAIRS (kappa = 0.11 for TST and 0.27 for SE%). P-values on McNemar's Chi square test for PAIRS with TST and with SE% were 0.072 and 0.011, respectfully.

Conclusions

These results suggest that actigraphy and PAIRS do not agree. Actigraphy TST captures movement and an estimate of specific sleep parameters. PAIRS is a broader measure that incorporates sleep disturbance and sleep-related impairment.