Psychological Factors Associated with Asthma 107reported to be 24% (Yellowlees, Alpers, Bowden, Bryant, &
Ruf“n, 1987), 12% (Yellowlees, Haynes, Potts, & Ruf“n,
1988), 10% (Carr, Lehrer, Rausch, & Hochron, 1994), 9%
(van Peski-Oosterbaan, Spinhoven, van der Does, Willems,
& Sterk, 1996), and 10% (Afari, Schmaling, Barnhart, &
Buchwald, 2001). The reasons for this increased co-
occurrence are not known, but there are several possible con-
tributing factors: Patients with each condition experience
similar symptoms (Schmaling & Bell, 1997), and conse-
quences, such as avoidance of situations where previous at-
tacks have occurred, or situations that bear similarities to the
venues of previous attacks. Anxiety-related hyperventilation
may exacerbate asthma through cooling of the airways, and
asthma may increase susceptibility to panic through hyper-
capnia, and the anxiogenic side effects of beta-agonist
medications. See Carr (1999) for a recent review on other
potential associations between asthma and panic.
Mood Disorders
As with other chronic medical conditions, asthma also is
associated with a greater prevalence of mood disorders than in
the general population. Among samples of patients with
asthma, lifetime diagnoses of major depression or dysthymia
were found among 16% (Yellowlees et al., 1987) and 40%
(Afari et al., 2001) of persons, which can be compared to 17%
in the National Comorbidity Study (Kessler et al., 1994). The
cross-sectional association of several conditions including
asthma and major depression among young adults in a
population-based study was done to investigate the hypoth-
esis that in”ammation-associated activation of the HPA axis
results from dysfunctional responses to stress (Hurwitz &
Morgenstern, 1999). This study found that persons with
asthma were approximately twice as likely as persons without
asthma to have had an episode of major depression in the
past year. A study of twins found evidence for a genetic
contribution to the association of allergies and depression
(M. Wamboldt et al., 2000). As noted previously, allergies
frequently co-occur with asthma, leading to evidence for a
genetic contribution to the association between asthma and
depression. Hypothesized links between the mechanisms in-
volved in depression and allergies (and by association,
asthma) have been suggested previously (Marshall, 1993).
Previously, we noted that respiratory drive and the ability
to perceive dyspnea were impaired among patients with a
near-fatal asthma attack. Depression has been postulated
to be associated with these impairments (Allen, Hickie,
Gandevia, & McKenzie, 1994); and others have suggested
that depression is an important risk factor for fatal asthma
(B. Miller, 1987).
In contrast with studies that have found an increased inci-
dence of psychiatric disorders among patients with asthma,
other studies have found anxiety and depressive symptoms to
be strongly related to respiratory symptoms, but not a diag-
nosis of asthma (Janson, Bjornsson, Hetta, & Boman, 1994).
There is signi“cant variability in the sampling and mea-
surement methodology used by the studies in this area, and
the effects of such methodological variations should be con-
sidered when attempting to draw conclusions about the state
of knowledge in this area. For example, Janson et al. (1994)
accrued a random population sample, then assessed those re-
spondents who endorsed breathing symptoms more fully to
con“rm a diagnosis of asthma. But most other studies used
nonrandom samples of convenience, such as hospitalized
patients with asthma, or patients from a respiratory clinic.
The source of subjects and sampling methods likely result in
differences that may affect rates of psychiatric symptoms and
disorders. For example, a population-based sample will
probably have a smaller proportion of patients with moderate
and severe asthma than does a sample of hospitalized pa-
tients, based on the distribution of asthma severity in the
population.
Measurement variability also can result in different
estimates of psychiatric disorders. While measures of symp-
tom severity convey dimensional information more readily
amenable to robust parametric statistical techniques than the
current psychiatric diagnostic nomenclature, the classi“cation-
based diagnostic systems are the gold standard, and symptom
measures only estimate the presence or absence of a given di-
agnosis. Unfortunately, the relative contributions of sampling
and measurement variability to divergent results are dif“cult to
untangle as self-reported symptom measures are more practi-
cal and therefore often used in population-based studies
whereas more labor-intensive diagnostic interviews are more
likely to be utilized in smaller clinical samples.Functional StatusPopulation-based studies such as the Medical Outcomes
Study demonstrate that individually, psychiatric conditions
and chronic medical conditions are associated with poorer
functional status (e.g., Hays, Wells, Sherbourne, Rogers, &
Spritzer, 1995). Asthma is associated with a lower quality of
life (Quirk & Jones, 1990), is the third leading cause of lost
time from work, behind two categories of back problems
(Blanc, Jones, Besson, Katz, & Yelin, 1993); and generally,
has a negative effect on functional status (Bousquet
et al., 1994; Ried, Nau, & Grainger-Rousseau, 1999). In par-
ticular, patients with asthma anddepression or anxiety had sig-
ni“cantly worse physical functioning and health perceptions