Does a modest loss of sleep affect neurocognitive functioning of children?

Objective(s): To (a) assess effects of modest sleep restriction and extension on children's neurobehavioral functioning (NBF); and (b) test the hypothesis that sleep restriction improves sleep efficiency compared to sleep extension. Study population: Children (39 boys, 38 girls) from grades 4 (n = 42, mean age 9.8 years) and 6 (n = 35, mean age = 11.6 years). Children were excluded if they had acute or chronic physical illness, used medication or reported developmental or psychiatric disorder. Study design: Between-groups comparison using random assignment to groups for analyses of instructions on sleep status and an added criteria of success in meeting randomly assigned sleep-behavior goals defining the groups for evaluation of effects of sleep changes on neurobehavioral functioning. Methods: All subjects slept at home, kept sleep - wake diaries and wore wrist activity monitors during their sleep period for 5 consecutive days. On days 1 and 2 they slept following their regular patterns. After the second night, on day 1 or 2 they were tested for neurobehavioral functioning (NBF) using the neuropsychological evaluation system (NES) developed by one of the authors of this paper. On day 3 the child's parents were called with instructions on whether to have the child go to bed an hour earlier or an hour later for the next 3 nights. A repeat NBF assessment was obtained on the morning after night 5. All NBF testing occurred between 08:00 and 10:00 h. The neuropsychological evaluation included six tests yielding nine different measures: finger taping speed, simple reaction time, continuous performance (CPT) (omission errors, commission errors and reaction time), symbol-digit substitution, visual digit span (forward and backward), and serial digit learning. Sleep-wake diaries were completed for the child's subjective report of sleep characteristics. An activity meter worn on the non-dominant wrist during the sleep times documented the decreased activity usually associated with sleep. The activity was used as a surrogate measure for sleep state using a scoring routine developed by one of the authors. This surrogate measure becomes important for defining the groups used in the evaluation of the effects of sleep changes on neurobehavioral functioning (NBF). Three hypotheses were tested: (1) most children would restrict or expand sleep at home on demand; (2) restricting sleep for children improved 'quality' compared to extending sleep; and (3) sleep restriction compared to extension increased subjective daytime fatigue and compromised NBF. The test for the first two hypotheses used the groups as defined by random assignment to restricted or extended sleep. The last used the same groups, but were further refined by the success criterion for the restricted or extended goal. For this analysis, a third group of those failing to reach the criterion for desired sleep change, was included as a 'no change' group. Results: All subjected measures of sleep and objective measures of activity showed the expected differences with strong statistical significance (usually < 0.005) between the restricted and extended sleep groups. There were two exceptions to this. Morning rise time did not vary; rather, as was expected, the sleep times were changed by earlier or lated bed times. The change in percentage of sleep, defined as quiet or inactive by the activity meter, showed only a small and unimpressive difference favoring the restricted sleep group; about the same as an initial difference between the randomly assigned groups. The sleep time, defined by the activity meter as a percent of time in bed, showed a more significant increase for the restricted than extended sleep groups (increase by 3.5% vs. decrease by 0.8%, P < 0.005), but again, the change in the restricted group only increased the percentage of sleep to the value observed at baseline for the extended group. The changes in these measures, while statistically significant, were in the range of the variation observed between the groups. The sleep time changes were more impressive, amounting to over an hour difference in time in bed (8 h for restricted sleep group compared to 9.2 for the extended) and activity measures indicating periods of inactivity associated with sleep similarly differing by about an hour (7.6 vs. 8.6 h). Similarly, subjective reports indicated shorter sleep latency but increased fatigue for restricted compared to extended sleep. The arbitrary criterion set for successful compliance was an average change of 30 min sleep time in the requested direction over the 3 days. Sleep time was presumably defined by the surrogate measure based on decreased wrist activity. About 63% of the children (65% for extension and 62% for restriction) made this goal. Since nearly half of the children in each experimental group failed to achieve the criterion for successful sleep change, they were analyzed as a 'no-change' group for the evaluation of the effects of sleep changes on NBF. The NBF effects on the nine measures showed significant differences (P < 0.05) for simple reaction time (getting worse for both the restricted and 'no change' group, but not changing for the extension group) and digit span forward (increasing for the extension group by 0.28, decreasing for the restriction group by 0.36, with little change (0.05 increase) for the 'no change' group). The continuous performance test reaction time (CPT-RT) showed highly significant (P < 0.005) differences, with an average decrease of 0.28 ms for the sleep-extension group, compared to 0.17 increase for the 'no change' and 1.0 decrease for sleep-restriction groups. Conclusion: The authors reached three basic conclusions. First, that most children can increase or decrease sleep times on demand over a short period of time, and can do this by changing their bed times. Second, that decreasing sleep time improves sleep efficiency and quality, but also increases daytime fatigue.