Original article
Adolescence Sleep Disturbances as Predictors of Adulthood
Sleep Disturbances—A Cohort Study
Alex Dregan, M.Sc.
a,
*
, and David Armstrong, M.D.
b
a
Sociology of Sleep Group, Department of Sociology, University of Surrey, Guildford, Surrey
b
Department of General Practice and Primary Care, King’s College, London
Manuscript received April 28, 2009; manuscript accepted November 9, 2009
Abstract Purpose: The present study aimed to estimate whether sleep disturbances in adolescence predicted
sleep disturbances in later years.
Method: Our sample included 7,781 cohort members from the United Kingdom’s National Child
Development Study. Sleep disturbances at ages 16, 23, 33, and 42 were measured by asking whether
cohort members had difficulties in falling/maintaining sleep or waking unnecessarily early in the
morning.
Results: Multivariate regression analyses indicated that sleep disturbance at age 16 was a significant
predictor of sleep disturbances at ages 23, 33, and 42. Continuity of a number of risk factors, especially
depression, accounted for some of the persistence of sleep disturbances over time but did not explain
a significant part of ongoing sleep disturbance.
Conclusions: Our findings suggest that many sleep disturbances start in adolescence and continue
into later years. Ó 2010 Society for Adolescent Medicine. All rights reserved.
Keywords: Adolescence; Sleep disturbance; Depressive disorder; Longitudinal study; Cohort study; Life course; Comorbidity
Sleep disturbance has high prevalence being reported by
about a third of the adult population [1,2]. In cross-sectional
surveys of sleeping disturbances it is also common to find
that the problem is a chronic one, with sleep disturbances
being reported as having lasted for years or even decades
[3]. Evidence from longitudinal studies to support the chro-
nicity of sleep disturbance, however, is more limited. A
number of studies have reported some persistence over
shorter periods of time in children, adolescents, adults, and
the elderly [4–8]. An 8-year UK longitudinal study [9] found
that a third of those who reported insomnia symptoms in
1985 were still reporting these 4 years later (1989), and
a further 10% of the 1989 survey continued to report
insomnia symptoms in 1993. Studies that have followed up
the same group of people over longer periods to support
the common assumption that a sleep disturbance is a chronic
condition are less available. Indeed, review articles in the
field have emphasized the need for studies of the longitudinal
course of sleep disturbances [2,10].
If sleep disturbances do persist over time, then this might
be explained in two different ways [11]. First, sleep distur-
bances might be ‘primary’ in the sense that it is the initial
sleep disturbance that somehow engenders a later sleep
disturbance, perhaps through an underlying physiological
cause or the creation of a behavioral pattern. Alternatively,
ongoing sleep disturbances might be secondary to another
quite different factor that persists over time and of which
sleep disturbance is simply a symptom. Chronic depression,
for example, might produce long-term sleep disturbances but
it is the continuity of the depression that is the primary
problem and the associated chronic sleep disturbances
secondary to it. Another way of expressing this distinction
is to ask whether a past sleep disturbance is a better predictor
of a current one than other risk or comorbid factors. An
earlier study [12], for example, surveyed an adult population
who had documented complaints of sleep disturbances 10 to
12 years previously and found that a poor sleep history was
*Address correspondence to: Alex Dregan, Sociology of Sleep Group,
Department of Sociology, University of Surrey, Guildford GU2 XH, Surrey,
UK.
E-mail address: a.dregan@surrey.ac.uk
1054-139X/10/$ – see front matter Ó 2010 Society for Adolescent Medicine. All rights reserved.
doi:10.1016/j.jadohealth.2009.11.197
Journal of Adolescent Health 46 (2010) 482–487