This narrative review asked why many people with insomnia report sleeping worse than sleep tests show. They often think they took longer to fall asleep and slept less than overnight brain recordings (polysomnography) or wrist movement monitors (actigraphy) suggest. The paper has no sample of its own. It pulls together studies of people with insomnia, good sleepers, other patient groups and rats. It cites adult samples and also mentions this mismatch in youth with ADHD and depressed youth. The authors give no age range.
The authors weighed 13 possible explanations. Only three met their bar for good quality evidence: sleep being felt as being awake, worry, and brief awakenings in the night. A poor sense of time was not supported, since all three studies on it found no difference from good sleepers. Watching a clock while falling asleep was linked to a bigger overestimate of how long it took. In one study, insomnia patients woken after a placebo said they had been asleep only 30% of the time. After a prescription sleeping pill, that rose to 97%.
The authors stress that misjudging sleep does not rule out a real sleep shortage. Agreement among the raters on evidence quality was only moderate, and the studies that tried to correct the misjudgment were small with no follow-up. The authors suggest clinicians should not assume people with insomnia are exaggerating their sleep problem.
Talk to a clinician about your own situation.
The numbers Stats and what the terms mean
The statistics behind the plain-language claims above.
- Harvey, 2011: agreement among six reviewers when rating evidence quality, 73.8% (Fleiss's kappa .60, Krippendorff's alpha .61).
- Harvey, 2011: agreement among six reviewers when rating the direction of the evidence, 92.5% (kappa .80, alpha .80).
- Mendelson (cited in Harvey, 2011): share of insomnia patients woken from sleep who felt they had been asleep, 30% on placebo versus 97%, 66% and 80% on the three sleeping pills tested. What the terms mean:
- Fleiss's kappa A score for how much several raters agree beyond what chance alone would give, where 1 is perfect agreement; here it measured how closely the six reviewers' ratings matched.
- Krippendorff's alpha Another score for agreement among raters beyond chance, also running up to 1 for perfect agreement.
- Placebo A look-alike treatment with nothing active in it, given to a comparison group so people can't tell which group they're in; here it was compared with three sleeping pills.
Source: Allison G. Harvey, Nicole K. Y. Tang. (Mis)perception of sleep in insomnia: A puzzle and a resolution.. Psychological Bulletin (2011). doi:10.1037/a0025730