Introduction: Using weighted blankets to address sleep problems among children with Attention-Deficit/Hyperactivity Disorder (ADHD) could complement pharmacological treatment in clinical practice. A cross-over RCT showed an effect of weighted blankets (WB) vs. control blankets (CB) on objectively measured sleep and parent-reported sleep problems (Lönn et al., 2023). However, evidence of long-term benefits is scarce, and additional research with longer follow-up time is warranted. This study aimed to evaluate changes in sleep when using weighted blankets among children with ADHD and sleep problems.

Materials and Methods: A longitudinal intervention design was chosen to evaluate children’s sleep during a 16-week sleep intervention with WB. Data was collected during baseline, a cross-over phase (4+4 weeks), and an open-label phase (8 weeks). Children (n= 71) choosing to continue the trial with the WB when entering the open-label phase were included. Outcomes were measured with the parent-reported Children's Sleep Habits Questionnaire (CSHQ total score), the child-reported Insomnia Severity Index (ISI), and actigraphy (Motionware 1.2.47 Camntech) including Sleep Onset Latency (SOL), Wake After Sleep Onset (WASO), Total Sleep Time (TST), and Sleep Efficiency (SE). Bedtime, wake-up time, and age were added as exploratory variables to explain changes in TST. Data was analyzed with mixed effect models, Spearman correlation and linear regression.

Results: Mixed effect models, including children (n=71) mean age 9.49 (sd 2.25, range 6-14) entering the open-label phase, showed that CSHQ decreased by -5.07 (p=0.000), and ISI decreased by -3.40 (p=0.000) from baseline to 16-week follow-up.Mixed effect models showed decreased TST by -11.2 min from baseline to 16-week follow-up (p=0.006). SOL, WASO, and SE did not change during the intervention (p>0.05).Exploratory analysis using mixed effect models on bedtime showed that bedtime was delayed with +13 min from baseline to 16-week follow-up (p=0.000). The wake-up time was not changed (p=0.874).Correlation analyses showed weak non-statistical significant relationships between TST and sleep problems (CSHQ) and TST and insomnia severity (ISI) at baseline (CSHQ: r=-0.15; ISI: r=0.15) and 16-week follow-up (CSHQ: r=-0.20; ISI: r=-0.06). TST and bedtime showed a strong correlation at baseline (r=-0.61, p=0.000) and 16-week follow-up (r=-0.65, p=0.000). TST and wake-up time showed a weak non-statistical significant relationship at baseline (r=-0.15) and 16-week follow-up (r=-0.16).Linear regression on baseline measurements showed that TST decreased with -13.9 min (p=0.000) and bedtime was delayed with 21 min (p=0.000) by increasing age.

Conclusions: Parent-reported sleep problems and child-reported insomnia severity decreased during the 16-week sleep intervention with weighted blankets for children with ADHD. The objective measures with actigraphy showed that total sleep time decreased during the 16-week intervention. Total sleep time was not found to be associated with parent- or child-reported outcomes but was associated with bedtime and age. These findings need to be further investigated to understand essential factors in sleep changes for implementing weighted blankets in clinical practice.