Insomnia and sleep disturbance are common symptoms of depression in adults, but this is not always the case in adolescents. In fact, a group of researchers at the University of Pittsburgh who studied insomnia in depressed teens found that, compared to adults, insomnia in adolescents is different—very different.

A 2005 study examined sleep differences between 51 depressed teens and 42 of their non-depressed peers. All of the depressed teens reported insomnia, many complaining that they hardly slept at all most nights. Subjects were interviewed before and after spending three nights in the lab where their sleep was recorded with polysomnogram and video recorder.

Participants completed standard measures of depressive symptoms and rated their sleep according to onset, total sleep time, number of awakenings, and ease of waking up the next day. The main finding of the study was that none of the depressed teens who complained of poor sleep actually slept poorly.

As a group and as individuals, their sleep records were entirely normal (Bertocci MA et al, J Am Acad Child Adolesc Psychiatry 2005;44(11):1158–1166). This finding was not unexpected. Indeed members of this group (see Dahl R et al, Psychiatr Res 1991; Vol 38(2):201–214) and others (Armitage R et al, J Affect Dis 2001;63(1–3):139–148; Emslie et al, Arch Gen Psychiatry 1990;47(2):119–1124) have previously found an absence of EEG correlation with subjective sleep complaints in teens.

Yet perhaps a separate analysis of the most depressed adolescents might find a weak correlation? What they found instead was the opposite: the teens who rated their depression the highest—and their sleep the worst—actually had the most normal polysomnograms.

In 2008, the group followed up this study with the addition of a comparison group of anxious teens. Again they found that the depressed adolescents complained of insomnia but slept normally. However, the anxiety group reported better sleep than they actually had. Anxious adolescents apparently underreport sleep difficulties almost as much as depressed teens overreport them. This finding was unexpected. As the authors note, “Sleep problems and anxiety disorders have been linked at many points in the life span.”

The anxious kids actually had more awakenings and more minutes awake than either the MDD sample or the controls, but they reported fewer. They did report slightly longer sleep latencies that were verified by the EEGs. But interestingly, their estimations of how long it took for them to fall asleep were less accurate than the other two groups (Forbes EE et al, J Am Acad Child Adolesc Psychiatry 2008;47(2):148–155).

Taken together, what do these studies tell us about insomnia in teen depression? First, they suggest that in this group reduced sleep is fundamentally a misperception similar to other somatic complaints. More severe depression is related to more complaints of sleep problems (and other depressive symptoms), but not to measurably impaired sleep. Like physical symptoms in somatization disorder, they are distressing but not real. The second implication of this work is trickier.

How do we approach such a universally distressing symptom, knowing that it is actually a false perception? Two researchers at Oxford University have shown that distorted sleep perceptions can be changed. Drs. Nicole Tang and Allison Harvey studied 40 adults, all of whom presented with a complaint of primary insomnia. Each subject kept a sleep log and wore an actigraph (an activity monitor) for three nights. They were asked to estimate their sleep onset latency and their total sleep time. Then half the subjects were shown the discrepancy between their perceptions and their actual sleep, while the other half were not. Both groups were tested for another three nights. The subjects who were shown evidence of their misperception did much better at follow up. Their predictions of their sleep onset latency and total sleep time improved, as did their anxiety around sleeping. And interestingly, their true sleep latency got shorter—consistent with less worry (Tang NKY and Harvey AG, Behav Res Ther 2004;42(1):27–39).

So maybe what our depressed teens need is not a medication but an actigraph. It just so happens there’s an app for that. For 99 cents, your tech savvy teens can download Sleep Cycle, a smartphone app that uses the accelerometer in the phone to measure and graph the user’s sleep cycle. They have to sleep with the phone, though, which can lead to texting all night with friends.

Finally, let’s give an assist to Mozart in the struggle to improve our kids’ sleep. A study from Taiwan showed that fifth graders given a 45 minute music CD to listen to at naptime and bedtime had improved sleep at the end of six weeks (Tan LP, J Music Ther 2004;41(2):128–150). One word of caution: we are not sure that fifth graders in America have “naptime,” so your results may vary.

An even more impressive study of 94 youth (ages 19 to 28) from Hungary demonstrated a specific benefit on sleep quality and mood from listening to classical music (Harmat L et al, J Adv Nur 2008;62(3):327–335). No follow up on whether it improved their dancing.