Psychological Factors Associated with Asthma 105signi“cantly more strongly among subjects with asthma com-
pared to those without. However, participants with asthma
had signi“cantly more shortness of breath, minute ventilation,
arousal, and depression during the medical photographs than
during the mental arithmetic, as compared to participants
without asthma. These results are similar to those of Rietveld
and colleagues, suggesting that stressful tasks (especially
tasks during which only passive coping was likely) increased
asthma and emotional symptoms among adults with asthma.
Although objective changes in pulmonary function (i.e., res-
piratory resistance) were observed, these changes were not
speci“c to, nor more pronounced among, participants with
asthma.
Finally, two additional studies were performed in the
laboratory that used a personally relevant stressor paradigm
to enhance ecological validity. In this paradigm, the patient
with asthma and their intimate partner discussed two topics:
a problem in their relationship and an asthma attack that
occurred when the partner was present. Mood and pulmonary
function (peak expiratory ”ow rate, PEFR) were recorded
before, in the midst of, and after the discussions, which
were videotaped and behaviorally coded. The “rst study in-
volved six individuals with severe asthma and their partners
(Schmaling et al., 1996). PEFR improved for two patients
and deteriorated for four patients over 30 minutes of interac-
tion with an average magnitude of about one standard devia-
tion. Across patients, more hostile and depressed moods
were associated with decrements in PEFR. The second study
involved 50 patients with mild-to-moderate asthma and their
partners (Schmaling, Afari, Hops, Barnhart, & Buchwald,
submitted). On average, pulmonary function (PEFR) de-
creased one-third of a standard deviation, and self-reported
anxiety was related to decrements in pulmonary function.
Variability in pulmonary function was associated with more
aversive behavior, less problem-solving behavior, and less
self-reported anxiety. Interactions with a signi“cant other
appeared to result in more change in pulmonary function
among participants in the “rst study with more severe asthma
(average baseline FEV 1 was 56% of predicted), than among
participants in the second study with asthma of mild-to-
moderate severity (average baseline PEFR was 78% of pre-
dicted), and could not be explained by differences in global
relationship satisfaction between the two samples, which
were comparable. Decrements in pulmonary function were
associated with depression and hostility in the “rst study, and
anxiety in the second study. Participants in the two studies
had similar average levels of observed behavior and self-
reported depression and hostility, but among the sample with
severe asthma, self-reported anxiety was nearly three times
that of the sample with asthma of mild-to-moderate severity.
More marked anxiety may be related to the larger magnitude
of change in pulmonary function among participants with se-
vere asthma, even though anxiety was not related to changes
in pulmonary function in this group, potentially due to the
small sample size, and ceiling effects and limited variability
in the anxiety measure in the sample with severe asthma.
Taken together, stressful laboratory tasks are associated
with changes in subjective asthma symptoms. Studies of
adults suggest that stressful tasks also are associated with
changes in objective measures of pulmonary function, but
since not all studies included a healthy control group (Ritz
et al., 2000, being an exception), we cannot say that stress-
induced changes in pulmonary function are speci“c to or
more pronounced among persons with asthma.Studies in the Natural EnvironmentSeveral studies have measured the covariation of asthma
symptoms and/or measures of air”ow with mood states,
using daily or more frequent monitoring of patients with
asthma. Hyland and colleagues have published several case
reports and multiple case series that examine the association
between mood states and peak ”ow. Among 10 adults with
asthma who completed measures of mood state (positive and
negative affect) and peak ”ow in the morning and evening,
three showed signi“cant correlations of mood state with
PEFR over 15 days (rs .50): more positive mood
states were associated with higher PEFR (Hyland, 1990).
These relationships were more robust in the evening than the
morning. One study (Browne & Hyland, 1992) examined
the association between mood states and peak ”ow in a single
case before and during the initiation of medical treatment.
They reported robust correlations averaging .55 before treat-
ment, and .77 during the initiation of treatment, with more
positive mood states being related to higher PEFR.
Apter and colleagues (1997) had 21 adults with asthma
rate mood states and measure PEFR for 21 days, three times
a day. The data were pooled across subjects and observations
and revealed multiple associations between mood states and
PEFR, after controlling for symptom ratings. Positive mood
states (pleasant, active) were concurrently associated with
greater PEFR; unpleasant and passive mood states were asso-
ciated with lower PEFR. An examination of lagged associa-
tions showed that pleasant mood states predicted subsequent
higher PEFR, but PEFR did not predict subsequent mood
states. In another study by this group (Af”eck et al., 2000),
48 adults with moderate to severe asthma collected mood
states and PEFR data thrice daily for 21 days. Between-
subjects associations of PEFR and mood state revealed
no statistically signi“cant associations. Four percent of