112 Asthma
analyzed. We have recently combined components of asthma
education and CBT for panic disorder to develop a treatment
protocol appropriate for adults with asthma and panic disorder
(Feldman, Giardino, & Lehrer, 2000). The treatment includes
components on education about asthma and panic, asthma
self-management, anxiety management, instruction on distin-
guishing asthma attacks from panic attacks, exposure and
problem-solving therapy, and “nally, relapse prevention. This
treatment is currently being empirically tested. The NHLBI
guidelines for asthma treatment recommend referral to mental
health professionals when stress appears to interfere with
medical management of asthma (NHLBI, 1997).
Written Emotional Expression Exercises
Persons with asthma, and especially children with asthma,
are more likely to experience negative emotions than are
healthy individuals, but may be less likely to express them
(Hollaender & Florin, 1983; Lehrer et al., 1993; Silverglade,
Tosi, Wise, & D•Costa, 1994). However, empirical data as to
whether and how negative emotions precipitate or exacerbate
asthma attacks are mixed (Lehrer, 1998).
Smyth, Stone, Hurewitz, and Kaell (1999) asked participat-
ing subjects to write an essay expressing their thoughts and
feelings about a traumatic experience. They demonstrated
generally improved health outcomes. Among participants with
asthma, the authors reported a clinically signi“cant improve-
ment in FEV 1 after a four-month follow-up, with no improve-
ment noted in a control group who wrote on innocuous topics.
Other Psychosocial Interventions
Castés et al. (1999) provided children with asthma a six-month
program that included cognitive stress-management therapy,
a self-esteem workshop, and relaxation/guided imagery.
Improvement occurred both in clinical measures of asthma
and in asthma-related immune-system measures. The treat-
ment group, but not the control group, signi“cantly decreased
their use of beta-2 agonist medications, showed improvements
in FEV 1 , and, at the end of treatment, no longer showed a re-
sponse to bronchodilators (consistent with improvement in
asthma). Basal FEV 1 improved to normal levels in the treat-
ment group after six months of treatment. Children in the treat-
ment group showed increased natural killer cell activity and a
signi“cantly augmented expression of the T-cell receptor for
IL-2, along with a signi“cantly decreased count of leukocytes
with low af“nity receptors for IgE. The results suggest that,
over the long-term, stress management methods may have
important preventive effects on asthma, and may affect the
basic in”ammatory mechanisms that underlie this disease.
Direct Effects of Psychological Treatments
on the Pathophysiology of AsthmaRelaxation TrainingIn an earlier review (Lehrer, Sargunaraj, & Hochron, 1992),
we concluded that relaxation training often has statistically
signi“cant but small and inconsistent ef fects on asthma.
More recent studies have yielded a similar pattern (Henry,
de Rivera, Gonzales-Martin, Abreu, 1993; Lehrer et al., 1994;
Lehrer, Hochron, et al., 1997; Loew, Siegfried, Martus,
Trill, & Hahn, 1996; Smyth, Stone, et al., 1999; Vazquez &
Buceta, 1993a, 1993b, 1993c). Outcome measures, popula-
tions, and relaxation procedures differ across studies, and
may explain some of the inconsistencies. Although clinically
signi“cant relaxation-induced changes in pulmonary func-
tion have been noted in asthma, they do not occur consis-
tently. It is possible that relaxation training may have an
important effect only among people with emotional asthma
triggers, or that the pre-existing effects of asthma medication
attenuated the effects of relaxation training in these studies.
Data from our laboratory suggest that the immediate ef-
fects of relaxation on asthma may differ from the longer term
effects (Lehrer et al., 1994; Lehrer, Hochron, et al., 1997).
We found that pulmonary functiondecreasedbetween the
beginning and end of speci“c relaxation sessions, and that
these decreases were correlated with evidence of increased
parasympathetic tone. Such •parasympathetic reboundŽ ef-
fects are commonly seen during the practice of relaxation. A
smallimprovement in pulmonary function was observed
over six weeks of treatment, showing that the immediate ef-
fects of relaxation may differ from the longer term effects.
We have hypothesized that this improvement results from
a general decrease in autonomic reactivity. Gellhorn (1958)
hypothesized that this is a general effect of relaxation
methods, mediated by decreased sympathetic arousal, and
consequent downregulation of homeostatic parasympathetic
re”exes. More recent literature con“rms this hypothesis
(Lehrer, 1978, 1996). Therefore, when assessing the impact
of interventions that are directed at reducing autonomic
arousal or reactivity, it may be important not only to measure
physiological changes over the course of multiple sessions,
but also to be aware that measures taken immediately fol-
lowing the termination of a session of relaxation training
may re”ect to observe any therapeutic bene“ts that may be
produced.
Vazquez and Buceta (1993a, 1993b, 1993c) studied the
effects of an asthma education program, both alone and
combined with progressive relaxation instruction. They
found evidence for relaxation-induced therapeutic effects
on duration of asthma attacks only among children with a