322 Insomnia
1985; Ohayon & Caulet, 1996). Epidemiological data indi-
cate that between 7% (Mellinger et al., 1985) and 10%
(Ohayon & Caulet, 1996) of the population use a sleep-
promoting drug. Other estimates indicate that 20% of indi-
viduals with insomnia have used a sleep medication in the
past and that 40% have used alcohol as a sleep aid (Gallup,
1991). Studies of insomnia complaints in general medical
practice reveal even higher prevalence rates. For example,
one survey found that more than 30% of medical patients had
either moderate or severe insomnia, and that almost one quar-
ter of those regularly used prescribed hypnotics (Hohagen
et al., 1993). The prevalence of hypnotic use is systematically
higher among older adults, women, and individuals with
chronic health problems (Ohayon & Caulet, 1996). More
than 40% of hypnotic drugs are prescribed for older adults,
although this segment of the population represents only about
13% of the population.
Correlates and Risk Factors
Several demographic, psychosocial, and health variables have
been associated with insomnia complaints. Surveys have con-
sistently found higher rates of insomnia complaints among
women, older adults, and individuals who are unemployed,
separated or widowed, living alone and/or homemakers
(Ford & Kamerow, 1989; Mellinger et al., 1985). Women are
twice as likely as men to report insomnia; however, it is un-
clear whether this higher rate is accurate or re”ects gender dif-
ferences in reporting or sleep perception. In addition, between
25% and 40% of individuals over the age of 60 complain of
sleep dif“culties, with about half of these individuals report-
ing serious insomnia (Foley, Monjan, Izmirlian, Hays, &
Blazer, 1999; Mellinger et al., 1985; National Institutes of
Health, 1994). These “gures remain fairly stable even after
controlling for the presence of comorbid medical problems
(Bliwise, King, Harris, & Haskell, 1992). Some evidence also
indicates that insomnia episodes are predictive of future in-
somnia episodes (Breslau, Roth, Rosenthal, & Andreski,
1996; Klink, Quan, Kaltenborn, & Lebowitz, 1992) and that a
positive family history of insomnia may also increase the risk
for future insomnia (Bastien & Morin, 2000).
There has been no longitudinal study of psychosocial risk
factors for insomnia. However, several studies have provided
indirect evidence that stress may increase the vulnerability to
develop insomnia. In a retrospective study, 74% of poor
sleepers recalled speci“c stressful life events associated with
the onset of their insomnia, and the frequency of such events
was greater during the year the sleep problem began than in
either the previous or subsequent years (Healy et al., 1981).
Signi“cant losses through separation, divorce, or the death of
a loved one were the most common precipitants. In another
study, individuals with insomnia reported a greater frequency
of negative life events (mostly related to interpersonal rela-
tionships) and diminished coping skills relative to normal
controls during the year preceding the onset of their insomnia
(Vollrath, Wicki, & Angst, 1989). The rate of reported sleep
disturbances among residents of Israel was also higher during
rather than before or after the Gulf War (Askenasy & Lewin,
1996). Another study (Morin, Rodrigue, & Ivers, under re-
view) found that it was the daily hassles, rather than major
live events, that placed individuals at greater risk for
sleep disturbances. Research about personality factors and
cognitive styles has repeatedly found that individuals with in-
somnia are more likely, relative to good sleepers, to display
anxious pro“les and engage in excessive worrying, obsessive
ruminations, and internalization of psychological con”icts
(Edinger, Stout, & Hoelscher, 1988; A. Kales, Caldwell,
Soldatos, Bixler, & Kales, 1983).Sleep and PsychopathologyEpidemiological, cross-sectional, and longitudinal data indi-
cate a high rate of comorbidity between sleep disturbances
and psychopathology (for a review, see Morin & Ware,
1996). This is no surprise given that sleep disturbance is a
diagnostic criterion or a clinical feature in several psychiatric
disorders, particularly anxiety (e.g., generalized anxiety dis-
order) and affective disorders (e.g., major depression). The
“rst line of evidence supporting a link between insomnia and
psychopathology comes from epidemiological surveys.
About 40% of randomly selected community residents with
insomnia complaints also experience signi“cant psychologi-
cal symptoms, relative to base rates of about 15% among
respondents without sleep complaints (Ford & Kamerow,
1989; Mellinger et al., 1985). Surveys of psychiatric outpa-
tients indicate that 50% to 80% have sleep complaints and
over 75% present signi“cant sleep disturbances during the
acute phase of their illness (Sweetwood, Grant, Kripke,
Gerst, & Yager, 1980).
Several cross-sectional studies have found a higher preva-
lence of psychiatric disorders among poor sleepers than
among good sleepers. Although speci“c estimates vary
greatly depending on the criteria and the samples selected,
estimates from clinical case series of patients consulting for
insomnia at sleep disorder clinics indicate that about 35% to
40% of those patients have at least one comorbid psychiatric
disorder (Buysse et al., 1994; Morin, Stone, McDonald, &
Jones, 1994). The most prevalent Axis I conditions in-
clude depression (major depression and dysthymia), anxiety
(e.g., generalized anxiety disorder), and substance abuse dis-
orders. In major depression, sleep disturbances often persist
even after the depression has lifted, while in older adults,