106 Asthma
within-subjects variability in PEFR was associated with more
active and aroused mood states. This study sought to differ-
entiate mood states• arousal levels from their hedonic va-
lence, and generally found that arousal levels were more
strongly associated with PEFR than hedonic valence. This
approach represents an important step toward reducing error
in the measurement of mood.
In a sample of 32 adults who rated moods and stressors
and measured PEFR for an average of 140 days, once a day,
50% of participants had one or more signi“cant mood or
stress predictors of PEFR (Schmaling, McKnight, & Afari,
in press). Patients• characteristics were examined in an at-
tempt to identify variables that were related to an association
between mood and PEFR. Lower global relationship satis-
faction tended to characterize participants for whom PEFR
was associated with relationship stress on a day-to-day
basis, and the presence of an anxiety disorder tended to be
related to a greater day-to-day covariation of anxiety and
PEFR.
Another study demonstrated strong associations between
moods and stressors and PEFR among 20 adults who moni-
tored psychosocial variables and peak ”ow “ve times a day
for 10 days (Smyth, Soefer, Hurewitz, Kliment, & Stone,
1999). Moods and stressors accounted for 17% of the vari-
ance in peak ”ow; positive moods were associated with
increased peak ”ow, and negative moods and stressors were
associated with peak ”ow decrements.
Steptoe and Holmes (1985) asked 14 men (half of whom
had asthma) to monitor mood and PEFR four times a day for
24 days. Six of the seven participants with asthma, but
only three of the seven participants without asthma, demon-
strated signi“cant within-subject associations between mood
and PEFR. Fatigue was the sole mood state that demonstrated
an association for all three participants without asthma. For
participants with asthma, which moods were associated with
asthma varied by person.
Summary
In their review, Rietveld, Everaerd, and Creer (2000) state, •it
remains unclear whether stress-induced airways obstruction
really exists.Ž Their critiques of the methodological issues
with the studies in this area are worthy of note. For example,
PEFR and spirometry are dependent on effort, and spurious
associations between mood and asthma could result if the
measure of asthma is dependent on effort and in”uenced by
mood. Their conclusions are based on classic research as-
sumptions that for an effect to be observed reliably, it should
be repeatedly demonstrable in studies with suf“cient sample
sizes utilizing between-subjects designs. Thus, one possible
interpretation of the research to date is that the association
between stress and asthma is weak. Another interpretation is
that the possible mechanisms involved in an association be-
tween stress and asthma are multifaceted and complex, prob-
ably involving endocrine, immune, and autonomic pathways.
It is a thorny issue to reveal simultaneously the mechanism
and the effect, since signi“cant error may be introduced by
inadvertently covarying factors that may affect the observed
association between stress and asthma. With the likelihood
that multiple, complex paths are involved, let us suggest that
the association between stress and asthma isidiographic,that
is, to be determined on an individual basis and in”uenced by
an as-yet not fully characterized set of variables that convey or
protect against the risk for an association between stress and
asthma. An idiographic approach also is consistent with the
results of existing studies that have examined the association
between stress and asthma on an individual-by-individual
basis and found that some but not all individuals demonstrate
an association, and that different emotions may be associated
with change in pulmonary function for different individuals.
Future research in this area should systematically examine
data resulting from laboratory versus naturalistic sources of
emotions; consider a parsimonious set of interaction effects
such as the interaction of emotion induction and suggestion,
or emotion induction in various environments (e.g., with or
without the presence of a supportive signi“cant other); exam-
ine the effects of emotions with positive versus negative va-
lences; consider the role of emotional intensity; and examine
these relationships in well characterized subgroups of pa-
tients for differences by gender, age group (adults versus ado-
lescents), asthma severity, and so forth. In addition, the type
of stressor and the potential coping methods available for
the participants• use is worthy of further study. Potentially,
the trend toward different results obtained in laboratory situ-
ations using contrived tasks and in naturalistic studies may be
attributable to differences in the ability to engage in active or
passive coping, as the former has been associated with bron-
chodilation and the latter with bronchoconstriction (Lehrer,
1998; Lehrer et al., 1996).Comorbid Psychiatric Disorders: Prevalence and EffectsPanic DisorderPanic disorder occurs among patients with asthma at a rate
several times greater than has been reported in the popula-
tion. For example, the National Comorbidity Survey (Kessler
et al., 1994) reported the lifetime prevalence of panic dis-
order to be 3.5%. By comparison, the prevalence of panic
disorder among samples of patients with asthma has been