Sleep disruption and delirium in critically ill children: Study protocol feasibility.

Only 10 children were analysed, and they were a convenience sample, so these feasibility numbers are small and may not apply to other PICUs. The study changed its endpoint from 48 hours to 10 hours because many children left the PICU early, so it did not capture the planned amount of sleep and environment data. Most children were young and not very ill: 90% were 2 years or less, median PRISM III score was 2.50, and none needed mechanical ventilation, so it says little about sicker or older children. Children with expected stays under 48 hours, deep sedation, restraints, sensory impairment, chronic critical illness, or anesthesia exposure were excluded, limiting generalisability. Video-based delirium screening had lower agreement than expected, and some screens were missed when children were asleep or video quality was poor. The pilot did not analyse whether sleep, light, sound or caregiving were associated with delirium; those data were not reported here. Family and staff feedback was not formally collected, and all four English-as-a-second-language families approached refused, partly because of video recording. The protocol was costly and time-intensive, which limits its use as routine clinical care.

One author, Tondi M. Harrison, is an Executive Editor at Research in nursing & health. The authors declared no other conflicts of interest.

Do not read the 60% delirium rate as proof that sleep disruption caused delirium. This was a 10-child feasibility pilot, not a test of cause or treatment, and the sleep and delirium data analysis was not reported here.