Psychosomatic Medicine Tips for Better Browsing
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by De Peuter, S.
Right arrow Articles by Van den Bergh, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by De Peuter, S.
Right arrow Articles by Van den Bergh, O.
Related Collections
Right arrow Anxiety
Right arrow Pulmonary
Psychosomatic Medicine 67:454-461 (2005)
© 2005 American Psychosomatic Society

ORIGINAL ARTICLES

Can Subjective Asthma Symptoms Be Learned?

Steven De Peuter, MA, Ilse Van Diest, PhD, Valentine Lemaigre, MA, Wan Li, MD, Geert Verleden, PhD, Maurits Demedts, PhD and Omer Van den Bergh, PhD

From the Departments of Psychology (S.D.P., I.V.D., W.L., O.V.d.B.), and Pneumology (V.L., G.V., M.D.), University of Leuven, Leuven, Belgium.

Address correspondence and reprint requests to Omer Van den Bergh, PhD, Department of Psychology, Tiensestraat 102, B-3000 Leuven, Belgium. E-mail: omer.vandenbergh{at}psy.kuleuven.ac.be


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Objective: We investigated whether perception of subjective asthma symptoms can be brought under control of biomedically irrelevant cues in the environment, i.e., whether subjective asthma symptoms can be learned in response to harmless stimuli.

Methods: Twenty patients with asthma and 20 healthy participants were presented with two placebo-inhalers presented as new chemicals for diagnosing asthma. One inhaler was coupled three times with rebreathing 5% CO2 in oxygen, the other inhaler was coupled three times with rebreathing oxygen. In the subsequent test phase, both inhalers were coupled once with oxygen. We assessed airway resistance and subjective symptoms throughout the study.

Results: Both groups expected and reported more symptoms with the inhaler that was previously associated with the CO2 trials compared with trials with the inhaler that was used on trials without CO2 without concomitant effects on respiratory resistance. The learning effects were most pronounced in a subgroup of patients reporting symptoms of hyperventilation during asthma exacerbations in daily life.

Conclusions: Subjective respiratory symptoms can be learned in response to harmless stimuli and a substantial proportion of patients with asthma might be especially vulnerable to this phenomenon. Because asthma patients rely mainly on perceived symptoms for their medication use, it is likely that they will take reliever medication based on expected symptoms instead of real exacerbations of respiratory dysfunction.

Key Words: asthma • symptom perception • conditioning • CO2 inhalation • learning • rebreathing

Abbreviations: ASC = Asthma Symptom Checklist; ANX = ASC anxiety subscale; DYS = ASC dyspnea subscale; FAT = ASC fatigue subscale; FOT = forced oscillation technique; HYP = ASC hyperventilation subscale; IRR = ASC irritability subscale; NA = negative affectivity; OBS = ASC obstruction subscale; VAS = visual analog scale.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
A large proportion of patients with asthma—numbers up to 60% have been reported—show poor perception of the status of their respiratory system (1–3). Some patients report symptoms in excess of physiological abnormalities; others hardly notice significant changes in respiratory functioning. Especially at the onset of a rapidly developing attack of airway obstruction, impaired awareness of physical changes can be fatal because appropriate medication is not administered timely. Conversely, overperception of symptoms is associated with excessive medication intake, unwarranted illness behavior, and hospitalization (4,5).

A variety of factors has been proposed to explain the discrepancy between pulmonary status and subjective symptom reports, including gender, age, asthma severity (6,7), recurrent exacerbations, increased sputum eosinophils (7), slow onset of attacks (8), and recent experience of breathlessness (9). Psychologic explanations have focused mainly on rather permanent conditions such as psychiatric comorbidity (i.e., panic, anxiety, depression) and personality variables such as defensiveness and negative affectivity (NA; (4,5); see reference (10) for a review). However, several other psychologic factors such as suggestion, attentional and interpretative processes, mood, and illness schemata also play an important role in the perception of pulmonary symptoms (11,12). Because asthma exacerbations often occur in association with particular environmental cues, they may also fit nicely into the framework of associative learning. The potential role of associative learning (or conditioning) for asthma has been recognized for decades (13). Little systematic research has been devoted to it, however, and most evidence is anecdotal (14–17). Some studies, however, provide solid evidence for the role of learning in asthma. First, in a convincing demonstration with rat subjects, mast cell secretion was conditioned to an audiovisual stimulus (18). Second, bronchial smooth muscle contraction in human participants was conditioned to colored slides (19).

The latter studies demonstrated the impact of conditioning processes on physiological processes but paid less attention to subjective complaints (19). However, conditioning may also play an important role through psychologic processes (20). Recent theorizing about symptom perception in asthma (21) argued that patients may learn an association between asthma exacerbations and cues that are unrelated to airway responsivity. Subsequent encounters with such a cue may induce worry over a possible attack, directing attention inward to search for signs of respiratory distress. As a result, chances are high that harmless bodily sensations will be interpreted as signs of airway narrowing (22,23). A set of studies by Van den Bergh and colleagues is particularly relevant. They showed that a few respiratory challenges inducing symptoms (through CO2-enriched air inhalation) in association with a specific cue (an odor added to the CO2-enriched air or fearful imagery during CO2 inhalation) were sufficient to elicit similar respiratory symptoms on presenting the cue alone. In addition, learning of respiratory symptoms was more likely in persons with high levels of NA (24–27).

In the present study, we used a similar paradigm to test a group of patients with mild to moderate asthma and a control group of asthma-free participants. Rebreathing 5% CO2-enriched air was used to induce symptoms because 1) previous findings had shown that inhaling 5% CO2 through the rebreathing technique (see Methods) did not elicit an airway response; 2) CO2 at that concentration has no smell or taste and can only be noticed by the physiological changes it produces (e.g., faster or deeper breathing); and 3) self-reported symptoms in response to CO2 inhalation were to a large part overlapping with the symptoms reported by patients with asthma in response to a typical asthma challenge such as histamine-induced bronchoconstriction. This technique enabled us to induce a respiratory symptom experience with resemblance to the subjective asthmatic experience in association with a biologically irrelevant cue without compromising the airways of asthmatics.

We assumed that experiencing a respiratory challenge in a diagnostic context would trigger schematic memory representations of previous symptom episodes in patients with asthma. Because nonasthmatics have no knowledge of such episodes, we hypothesized that patients with asthma, compared with normals, would 1) report more symptoms during CO2 inhalations on an asthma symptom checklist in the acquisition phase, and 2) would learn to report more of those symptoms. In addition, 3) we expected these effects to be strongest in high-NA asthma patients.

To help interpreting the source of the expected effects on symptom reports, a detailed analysis of several breathing parameters was carried out.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Participants
Twenty patients with asthma without other respiratory diseases (10 men; mean age 24 ± 7.8 years) participated. Severity of asthma was evaluated according to the GINA guidelines (28), after assessment of symptoms, of lung function parameters at rest and of prescribed medication. Eight patients qualified for intermittent asthma, six for mild persistent asthma and six for moderate persistent asthma. None of them had experienced an asthma attack for 6 weeks before participation. Twenty participants without asthma or any other respiratory disease (11 men; mean age 23.7 ± 7.6 years) served as the control group.

Because of technical difficulties and equipment failure, data from one patient (intermittent asthma) and one control subject were excluded from the analyses.

Measurements and Apparatus
Lung Function Testing
We measured respiratory resistance by means of the forced oscillation technique (FOT; (29)).1 Both at baseline and at the end of the experiment, forced expiratory volume in 1 second (FEV1), forced vital capacity (FVC), and peak expiratory flow (PEF) were calculated from a forced expiratory maneuver following FOT measurements.

Administration of CO2 and Measurement of Breathing Behavior
CO2 was administered from an airtight plastic bag (maximum capacity 10 L) containing 6 L of a decompressed air mixture of 5% CO2 in 95% oxygen (Air Liquide, Belgium; we used 95% O2 to avoid progressive hypoxia, which is known to suppress symptom perception, (31)). The bag was mounted onto a bacterial filter that was fitted on a heated pneumotachograph (Fleisch No. 2; Switzerland), which was further connected to a mouthpiece. The pneumotachograph was calibrated daily with a 1-L syringe. A small tube sampled air at the mouthpiece and sent it to an infrared CO2 monitor (POET II; Criticare), monitoring fractional end-tidal CO2 concentration (FetCO2) continuously. It was calibrated daily using calibration gas containing 5% CO2 in oxygen. Both devices were connected to a Labmaster card and a PC.

Participants wore a noseclip and (re)breathed from the bag through the mouthpiece (rebreathing technique; (32)). Volume and CO2 waveforms were sampled at 20 Hz. All waveforms were visually inspected offline to eliminate technical abnormalities. Specifically designed software was used to extract pauses and irregularities and to calculate inspiratory time (Ti, seconds), expiratory time (Te, seconds), inspiratory volume (Vi, mL), drive (inspiratory flow, mL/s), and FetCO2 (%) per breathing cycle.

Subjective Measures
We used the Asthma Symptom Checklist (ASC) to measure subjective symptoms (33,34). The ASC is a 36-item checklist used to assess the subjective symptomatology in asthma. We used the validated Dutch translation (35) consisting of six subscales: symptoms of airway obstruction (OBS), dyspnea (DYS), fatigue (FAT), symptoms of hyperventilation (HYP), anxiety (ANX), and irritation (IRR). Internal consistency for five of the six subscales is high (Cronbach’s {alpha}: 0.93, 0.88, 0.86, 0.87, 0.92) and acceptable for HYP (Cronbach’s {alpha}: 0.76). Three versions were used in this experiment: 1) the version at baseline assessed how frequently each of the symptoms is experienced during an asthma exacerbation. Ratings ranged from 1 (never) to 5 (always); 2) after the context-exposure trial and at test (see Procedure), patients rated how intensely they had experienced each of the symptoms during the preceding trial on an 11-point scale, end points being 0 (no symptoms) and 10 (symptoms as bad as possible); and 3) during the acquisition phase (see Procedure) a short version of the ASC was used, containing two items of each subscale to be rated on the same 11-point scale as version (2).

Previously (12), we showed that the ASC can be used to sensitively measure changes in symptomatology. Moreover, the ASC is a sensitive measure of the effects of CO2 inhalation.

The NA scale of the Positive and Negative Affect Schedule (PANAS) was used to assess negative affectivity as a personality trait (36). Participants have to rate the degree to which 10 negative adjectives are applicable to themselves. End points are very little or not at all (scored 1) to very much (scored 5), resulting in a total score between 10 and 50.

All questionnaires were presented in the paper-and-pencil version.

Procedure
All participants were recruited through advertisement in a local newspaper or in the online newsletter from the University of Leuven and contacted the experimenter (S.D.P.). Participants were informed over the phone that they were to "inhale two substances that could have an effect on their airways" and that minor complaints could occur, which would disappear soon after they stopped inhaling the substances.

On entering the laboratory, participants signed an informed consent form. Date of birth, medication intake, height, and weight were recorded. Next, baseline FOT measurement and lung function (FEV1, FVC, PEF) testing was done in that order, followed by the administration of the questionnaires (ASC and PANAS for asthma patients, only PANAS for healthy subjects). Subsequently, participants breathed through the mouthpiece for approximately 1 minute providing a baseline FetCO2 level. Next, participants breathed for 2 minutes from a bag filled with approximately 6 L of oxygen to get used to the procedure (context exposure trial). Thereafter, participants completed the ASC with the instruction to rate the intensity with which they had experienced each symptom during the rebreathing trial.

Acquisition Phase
The design of the study is depicted in Figure 1. Six 2-minute rebreathing trials were administered, each followed by the short version of the ASC. Three trials were 5% CO2 trials; the other trials used 100% O2. After the gas mixture was led into the bag, a puff from a placebo inhaler (containing propellant only; Allen & Hanburys, U.K.) was released into the bag. There were two inhalers, one plain white, the other with a black mark, and both were presented as a new chemical developed to improve the diagnosis of asthma. The current experiment, participants were told, was part of the procedure to measure the amount of complaints that patients with asthma and healthy participants experienced for each of the inhalers to be able to set the "threshold for asthma." One inhaler was consistently coupled with 5% CO2 (in oxygen) inhalation; the other with 100% O2. For half of the participants, the black inhaler was used on CO2 trials and the white on O2 trials. The reverse was true for the other half. The experimenter explicitly mentioned the color of the inhaler that was used.



View larger version (37K):
[in this window]
[in a new window]
  		Figure 1. Schematic presentation of the study design.

 

The O2 and CO2 trials were administered in a semirandomized order in such a way that no more than two trials with the same mixture followed each other and the last two trials were different. FOT measurements were repeated after the third and last trial of the acquisition phase. Immediately before each trial, participants indicated on a visual analog scale (VAS) to what extent they expected to experience symptoms.

Test Phase
After the last FOT measurement of the acquisition phase, we told participants that we would test both substances one last time after a short break (10 minutes). After the break, each inhaler was administered once more with 100% O2.

Half of the participants received the CO2-associated inhaler first and thereafter the O2-associated inhaler; the other half received the inhalers in the reversed order. Before each test trial, we assessed participants’ expectations of symptoms, after each trial, we assessed experienced symptoms with the full ASC. After the second test trial, FOT measurement was repeated and posttest spirometry was performed. Participants were fully debriefed and received financial compensation (15{OV0556}).

The medical ethical board of the hospital approved of the study.

Data Reduction and Statistical Analysis
Pre- and postlung function data were compared using paired t tests for dependent samples. We calculated mean end-tidal CO2 concentration (FetCO2), inspiratory time (Ti), expiratory time (Te), inspiratory volume (Vi), and drive for the first 45 seconds and the last 45 seconds of each trial. For the baseline trial, these values were entered in a 2 (time: first 45 or last 45 seconds of trial) x 2 (group: asthma versus control) analysis of variance (ANOVA) with repeated measures on the first variable. Data from the acquisition and test phase were analyzed using a 2 (time) x 2 (type of trial: CO2 versus O2) x 2 (group) design with repeated measures on the first two variables. Greenhouse-Geisser corrections were used when appropriate. For the sake of brevity, results of the respiratory measures will only be mentioned when important.

We used a median split to divide the patients with asthma into low versus high NA asthma patients and conducted a second series of ANOVAs on the patients with asthma with NA as a between-group variable.

For the ASC, item scores were added per scale. For the context exposure trial, the symptom scores were analyzed using a one-way ANOVA with group (patients with asthma versus control participants) as between-subject variable; for the acquisition phase, we performed a 2 (type of trial) x 3 (trial) x 2 (group) ANOVA with repeated measures on the first two variables; and for the test phase, we performed a 2 (type of trial) x 2 (group) ANOVA with repeated measures on the first variable. Expectancy ratings collected during the acquisition and test phase were analyzed similarly.

Whenever the assumption of homogeneity was violated for one of the ANOVAs, we used the Box-Cox transformation (37) to transform the data.2 If variances remained significantly different after transformation, we used a multivariate approach for repeated measures (Wilks’ test).

Finally, we correlated the differences in symptom scores between trials in the test phase across participants with the questionnaire data. Additionally, we calculated four difference scores: For the last two trials of the acquisition phase and for the test phase separately, we subtracted expectancy ratings on O2 trials from expectancy ratings on CO2 trials. Next, we subtracted the total symptom scores on O2 trials from the total symptom scores on CO2 trials. Finally, we calculated correlations between difference scores for expectancy ratings and difference scores for symptoms.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 NOTES
 REFERENCES
 
Baseline
As can be seen in Table 1, the asthma group had higher PEF values than the control group at baseline (t[36] = –3.03, p < .01), but on average, a higher airway resistance at 6 Hz (t[36] = –2.13, p < .05).3 These differences were no longer significant at posttest. The other lung function parameters did not differ between groups and did not change significantly during the experiment (Table 1). There was a significant effect of measure moment for resonant frequency (F[3,102] = 6.72, p < .001). Tukey HSD test showed that this was the result of an overall fall in resonant frequency from baseline to the third and final measure moment (p < .001 and .01, respectively). Baseline FetCO2 levels were 5.16% (standard deviation [SD] = 0.47) for the control group and 4.88% (SD = 0.42) for the asthma group (difference not significant). The mean NA score for the asthma and control group was 20.32 (SD = 7.39) and 17.95 (SD = 4.35), respectively, a difference that was not significant (F[1,36] = 1.45, not significant).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Means and (Standard Deviations) for Lung Function and Forced Oscillation Technique (FOT) Values per Group Throughout the Study

 

Context Exposure
Means and standard deviations for expectancy ratings and symptom scores during the context exposure trial and acquisition phase are shown in Table 2. There were no significant differences on any of the symptom (sub)scales or in respiratory behavior between the asthma and control group for the context exposure trial.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Means and (Standard Deviations) for Expectancy Ratings and Symptom Scores for the Context Exposure and Acquisition Trials

 

Acquisition Phase
Subjective Ratings
Overall, participants expected more symptoms on trials with CO2 (mean VAS rating = 27, SD = 21.12) than on trials without CO2 (mean = 20.88, SD = 20.25, d = 0.29) (F[1,34] = 11.53, p < .01). This effect was specified by a significant interaction with trial; expectancies for the first 100% O2 and 5% CO2 trial were similar, whereas from trial 2 on, the expectancies increased for the trials with CO2.

On average, trials with CO2 elicited a total symptom score of 28.85 (SD = 6.42), whereas trials without CO2 elicited a total score of 16.15 (SD = 4.51, d = 0.56; F[1,36] = 32.34, p < .001; Table 2). There were main effects of type of trial for the obstruction (OBS), dyspnea (DYS), anxiety (ANX), irritability (IRR), and hyperventilation (HYP) subscales; trials with CO2 elicited more symptoms compared with trials without CO2 (F[1,36] = 12.23, 44.58, 10.06, 21.75, and 10.44, respectively; all p’s< .01). For ANX, there was an additional main effect of trial (F[2,72] = 3.58, p < .05); compared with the first trial, anxiety scores were higher for the second and third trial. Finally, we observed a significant interaction between type of trial and trial for OBS; participants lowered their OBS ratings over trials for trials without CO2, whereas there was a slight increase in OBS ratings for trials with CO2 (F[2,72] = 4.52, p < .05). There were no significant effects for fatigue subscores (Wilks’ test).

The main effect of group for total symptom score did not reach significance, nor did any interaction effect involving group as a variable. There was, however, a significant effect of group for the HYP subscale; the asthma group reported overall more hyperventilation symptoms than the control group (F[1,36] = 6.18, p < .05). Finally, the analysis with NA as a between-variable did not produce significant effects.

Respiratory Measures4
The results of the CO2 manipulation were similar to those observed in previous research with the CO2 paradigm (25,26). Toward the end of the trials, both inspiratory and expiratory time decreased, participants breathed with a larger volume, and drive and FetCO2 increased. These effects were in general much stronger for trials with CO2 compared with trials with 100% O2.

Test Phase
Subjective Ratings
Means and standard deviations for expectancy ratings and symptom scores during the test phase for the study groups are shown in Table 3. We observed a significant learning effect on expectancy ratings: Participants expected to experience more symptoms with the inhaler that had been paired with CO2 during acquisition (mean = 29.42, SD = 4) than with the inhaler that had been paired with 100% O2 (mean = 14.95, SD = 2.90, d = 0.64; F[1,36] = 17.64, p < .001). There was a trend toward an interaction with group; both groups expected more symptoms on trials with the inhaler that had been paired with CO2, but this difference tended to be larger in the asthmatic group than in the control group (F[1,36] = 3.25, p = .08).


View this table:
[in this window]
[in a new window]
  		TABLE 3. Means (and Standard Deviations) for Expectancy Ratings and Symptom Scores for the Test Trials

 

Participants’ total symptom scores were in line with their expectations; more symptoms were reported with the inhaler previously associated with CO2 (F[1,36] = 7.96, p < .01). This learning effect was also significant for the DYS subscale (F[1,36] = 12.18, p < .01). In addition, in line with the results from the acquisition phase, patients with asthma reported significantly more hyperventilation symptoms than the control group (Masthma = 5.89, SD = 1.22; Mcontrol = 2.03, SD = 1.22, d = 0.67; F[1,36] = 5.01, p < .05). The overall group-effect on ASC total scores did not reach significance, nor did any interaction effect with group.

Separate ANOVAs on the asthma group with NA as between-subject variable (median split for negative affectivity: median = 18; Mlow group = 15, SD = 2.71; Mhigh group = 26.2, SD = 6.26) revealed only a significant NA effect for IRR in the test phase; high NA asthma patients reported more IRR symptoms than low NA patients (F[1,17] = 4.99, p < .05). No other effect of NA was observed.

Because of the large variance within the asthma group and unequal variances between groups, we looked in detail at the nontransformed data. Although 13 of 19 participants in each group had a positive difference score between the two test trials (more symptoms on the trial with the "CO2" cue than on the trial with the "O2" cue), the mean difference was larger in the asthmatic compared with the control group (mean = 16.9 versus 4.8, respectively). Taking the median (5) of the difference in the asthmatic group as a criterion, 10 patients with asthma versus three controls showed a learning effect ({chi}2 = 5.9, p < .05). Compared with the remainder of the group, those 10 patients with asthma reported significantly more hyperventilation symptoms during an asthma exacerbation in daily life (F[1,16] = 4.40, p = .05).

Respiratory Measures
Participants’ drive was significantly higher when the inhaler previously associated with CO2 was used (F[1,35] = 5.4; p < .05). The main effect of time for drive was significant (F[1,35] = 40.85, p < .001), but this effect was specified by a significant time x group interaction effect (F[1,35] = 6.21, p < .05). Drive increased for both groups from the first 45 seconds of the trials (mean = 474, SD = 45) to the last 45 seconds of the trials (mean = 557, SD = 36, d = 0.46). However, during the first 45 seconds of the trial, the asthma group’s drive (mean = 532, SD = 61) was higher than the control group’s (mean = 413, SD = 63, d=0.65; Fig. 2).



View larger version (17K):
[in this window]
[in a new window]
  		Figure 2. Respiratory drive during the test phase as a function of time and experimental group.

 

Lung function parameters did not systematically differ in relation to group or type of trial.

Correlations
For the acquisition phase, the correlation between the difference in expectancy ratings between trials with and without CO2 and the corresponding difference in symptom scores was 0.34 (p < .05); for the test phase, it reached 0.52 (p = .001). However, the correlations dropped to 0.08 for the acquisition phase and to 0.14 for the test phase when only the control group was considered (neither correlation significant). In contrast, for the asthma group, these correlations were 0.62 (p < .01) for the acquisition phase and 0.69 (p = .001) for the test phase).

Overall, there was a significant correlation between the difference in symptom level between the two test trials on the one hand and scores on the ANX and HYP subscales during the context exposure trial on the other hand (r = 0.64 and 0.50, respectively, p < .01). There were no significant correlations between the symptom difference scores and any of the difference scores in breathing responses during the test phase.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
NOTES
 REFERENCES
 
We demonstrated respiratory symptom learning in patients with asthma and a group of healthy participants. In a first phase, participants experienced three respiratory challenges inducing symptoms that were purportedly produced by "a new chemical for diagnosing asthma" from a (placebo) inhaler. When subsequently administered that same inhaler, but without the respiratory challenge, participants still expected and reported more symptoms, particularly of dyspnea and hyperventilation, compared with breathing from an inhaler that had not been associated with a respiratory challenge. Moreover, the participants’ drive was higher when the inhaler was used that had been accompanied by CO2 rebreathing, and this effect appeared sooner for the asthmatic group. Changes in pulmonary status were not responsible for the observed learning effects.

These findings suggest that learning may be involved in asthma patients’ symptom reports. A variety of internal and external stimuli such as exercise, cigarette smoke, cold air, and atmospheric pollutants may cause airways obstruction and subjective symptoms. Subsequent encounters with the same stimuli may loosen the relationship between physical changes and subjective symptoms and underlie overperception. Interestingly, whereas previous studies with this paradigm have only used odors or fearful imagery (25), the present study shows that also a visual cue—in this case an inhaler—can serve as conditioned stimulus for respiratory symptoms.

Contrary to our expectations, little difference emerged between the study groups in an overall comparison. Because participants from the control group had no asthma, it is surprising that they also learned to report (asthma) symptoms. This is probably the result of the instructions; in accordance with ethical standards, we told both healthy participants and patients with asthma that they would experience complaints in response to the "chemicals." Consistent with this, there was only a nonsignificant tendency in the test phase for patients with asthma to expect more symptoms than healthy participants in response to the inhaler that had been associated with CO2 inhalation.

Despite the absence of an overall group effect, closer inspection of the data showed large individual differences for symptom learning, with more patients with asthma than control participants (10 versus 3) showing a difference in symptom scores of five or more during test and the mean difference score being larger in the former compared with the latter group. Because the correlations between expectancy ratings and symptom scores were also higher in patients with asthma compared with control participants, at least a substantial proportion of patients with asthma seems to rely more on what they expect to happen than on actual sensations when they report symptoms. Those patients reported to have more hyperventilation symptoms during a typical asthma exacerbation in daily life.

Overall, the asthma group reported more hyperventilation symptoms than the control group during the acquisition and test phase across trial types. Because patients with asthma breathed more rapidly than the control participants during acquisition (shorter expiratory time) and their drive was also higher during the test phase, the symptoms possibly reflected a real increase in ventilation. However, this does not explain the learning effect or the difference in symptoms between the two test trials. There were no significant correlations between the difference in symptom scores and differences in breathing behavior for both test trials, whereas these symptom differences were related to the level of anxiety and hyperventilation symptoms during the (harmless) context exposure trial. In addition, the symptom difference scores correlated substantially with the difference in expectancy ratings. This suggests that symptom learning relies on anxiety-related perceptual–cognitive mechanisms rather than on accurate perception of (learned) respiratory behavior, which confirms earlier findings of our group (27). In general practice, up to 30% of adults with asthma report symptoms related to a dysfunctional breathing pattern and hyperventilation (38–40). Interestingly, our data show that especially those patients tend to associate these symptoms with their condition and report them as symptoms of asthma. It is likely, then, that a symptom-learning paradigm using real asthmatic challenges instead of CO2-enriched air may produce subjective symptoms that are even closer to real asthma symptoms.

The only effect of NA was observed on irritability, which is a negative emotion itself. No other effects of NA reached significance in contrast to results from former studies in our laboratory. Previous studies have shown that high NA individuals have a stronger attentional bias toward bodily sensations and are more inclined to contemplate possible negative health effects of the induced symptoms (4,5,41,42). However, the effect of NA on symptoms seemed to depend on the attentional demands of the ongoing events. A new and/or strong respiratory challenge wiped out the NA effects on symptoms, because in those circumstances, both high and low NA participants were attentionally involved with processing the events (41,43). The mildly threatening context of the current experiment might have increased attention in both high and low NA participants, consequently erasing habitual attentional differences between the groups.

Regarding the ecologic validity of the present manipulation, the possibility should be noted that, because CO2 inhalation has the same physiological effects in patients with stable, mild asthma as in normal controls, the patients with asthma did not consider the subjective effects of CO2 inhalation comparable to asthma symptoms. Consequently, no group differences should show up because our manipulation was arbitrary. Indeed, different symptom clusters are related to different ways in which breathing is compromised (11), and the sensations or the breathing discomfort induced by CO2 can be discriminated from the sensations evoked by resistive breathing and asthmatic airway reactions (44–46). However, although patients can make the distinction between the effects of CO2 inhalation and an asthma exacerbation, this is not always what they do.

In a comparison between the effects of CO2 inhalation (n = 20, increase of 2% above baseline using a similar rebreathing technique) and a histamine challenge (n = 50; in a standard diagnostic protocol inducing real bronchoconstriction), basically the same symptom profile emerged on the Asthma Symptom Checklist for the two types of challenges. Only differences in intensity showed up; our CO2 inhalation manipulation was much less intensive than the histamine challenge (Fig. 3).



View larger version (16K):
[in this window]
[in a new window]
  		Figure 3. Symptom profile of a histamine provocation and CO2 rebreathing. OBS = obstruction; DYS = dyspnea; ANX = anxiety; IRR = irritability; FAT = fatigue; HYP = symptoms of hyperventilation.

 

The present results therefore show that patients learn to report asthma symptoms on the basis of what could be called an "arbitrary" respiratory challenge, suggesting that this may occur even more likely when experiencing a true asthma exacerbation as a respiratory challenge. These findings are particularly relevant, because patients have been found to rely more on their subjective perception of discomfort than they do on objective findings to guide medication consumption (47) and tend not to regard the sensation associated with low readings of peak flow as severe constriction if they expect to improve soon (3).

In conclusion, subjective respiratory symptoms can be learned in response to harmless stimuli. Our findings suggest that learning may play an important role in overperception: If a few minor, short-lived, controlled, and supervised challenges with CO2 inhalation can elicit increased symptom reporting in a research setting, then the effect of a full-blown, unpredictable asthma exacerbation in daily life (which typically lasts longer than 2 minutes and does not occur in a hospital setting) supposedly is much larger. Because patients with asthma rely mainly on perceived symptoms for their medication use, chances are high that they will take reliever medication based on expected symptoms instead of on real exacerbations of respiratory dysfunction. The fact that it happens mainly in a subset of patients who report symptoms of hyperventilation during asthma exacerbations in daily life indicates that further research to characterize these symptom learners may shed more light on overperception. The present study may offer an interesting laboratory tool to investigate such processes; ethical constraints advise against repeated induction of true asthma exacerbations—or even bronchoconstriction—for research purposes only, and true exacerbations require a much longer recovery time and/or medication to return to baseline level.

We thank Prof. Dr. K. P. Van de Woestijne for critical technical advice and valuable comments on earlier drafts of the paper and F. Rochette for his respiratory DIY knowledge.


   NOTES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
REFERENCES
 
1A signal containing all harmonics of 2 Hz up to 32 Hz (2, 4, 6, ... Hz) was applied at the mouth. The impedance of the respiratory system was obtained by measuring the relationship between pressure and flow at the mouth. A Fourier analysis uses the pressure and flow signals to calculate an impedance value for each of the investigated frequencies. The impedance data obtained by the oscillation technique are immediately partitioned into a real (resistance) and an imaginary (reactance: capacitance and inductance) part of the impedance. The resonant frequency is the frequency at which the influence of capacitance and inertia cancels out (29). For analysis, we selected the resistance at 6 Hz because this frequency is considered highly specific for airway changes occurring in asthma. Additionally, we selected resonant frequency because of the high specificity and sensitivity of this parameter (30). Back

2All tables and figures use nontransformed means. Although these are sometimes not the values entered in the analysis, they provide a much better view of the data. Back

3Most likely, this is the result of unfamiliarity with lung function tests, equipment, and procedure; PEF measurements are well known to be vulnerable to motivation and practice. The control group performed spirometry for the first time, whereas all patients with asthma were familiar with the procedure. Back

4During analysis, we noticed that a substantial amount of respiratory data from trials with CO2 during the acquisition phase was lost as a result of software failure. For 11 participants (four from the asthma group), there were no data available. For the other participants, we averaged the data over the three acquisition trials per type of trial (with or without 5% CO2) and analyzed them using a 2 (time) x 2 (type of trial: CO2 versus O2) x 2 (group) design with repeated measures on the first two variables. Back

Dr. De Peuter is supported by grant G.0270.01 (FWO-Flanders) and BIL01/06.

DOI:10.1097/01.psy.0000160470.43167.e2


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 

  1. Rubinfeld AF, Pain MCF. Perception of asthma. Lancet 1976;1:882.[Medline]
  2. Nguyen B-P, Wilson SR, German DF. Patients’ perceptions compared with objective ratings of asthma severity. Ann Allergy Asthma Immunol 1996;77:209–15.[Medline]
  3. Kendrick AH, Higgs CMB, Whitfield MJ, Laszlo G. Accuracy of perception of severity of asthma: patients treated in general practice. BMJ 1993;307:422–4.
  4. Put C, Demedts M, Van den Bergh O, Demyttenare K, Verleden G. Asthma symptoms: influence of personality versus clinical status. Eur Respir J 1999;13:751–6.[Abstract]
  5. Put C, Van den Bergh O, Demedts M, Verleden G. A study of the relationship among self-reported noncompliance, symptomatology, and psychological variables in patients with asthma. J Asthma 2000;37:503–10.[Medline]
  6. Chetta A, Gerra G, Foresi A, Zaimovic A, Del Donno M, Chittolini B, Malorgio R, Castagnaro A, Olivieri D. Personality profiles and breathlessness perception in outpatients with different gradings of asthma. Am J Respir Crit Care Med 1998;157:116–22.
  7. in’t Veen JCC, Smits HH, Ravensberg AJJ, Hiemstra PS, Sterk PJ, Bel EH. Impaired perception of dyspnea in patients with severe asthma: relation to sputum eosinophils. Am J Respir Crit Care Med 1998;158:1134–41.[Abstract/Free Full Text]
  8. Turcotte H, Boulet L-P. Perception of breathlessness during early and late asthmatic responses. Am Rev Respir Dis 1993;148:514–8.[Medline]
  9. Wilson RC, Jones PW. Influence of prior ventilatory experience on the estimation of breathlessness during exercise. Clin Sci 1990;78:149–53.[Medline]
  10. Lehrer P, Feldman J, Giardino N, Song H-S, Schmaling K. Psychological aspects of asthma. J Consult Clin Psychol 2002;70:691–711.[CrossRef][Medline]
  11. De Peuter S, Van Diest I, Lemaigre V, Verleden G, Demedts M, Van den Bergh O. Dyspnea: the role of psychological processes. Clin Psychol Rev 2004;24:557–81.[CrossRef][Medline]
  12. Put C, Van den Bergh O, Van Ongeval E, De Peuter S, Demedts M, Verleden G. Negative affectivity and the influence of suggestion on asthma symptoms. J Psychosom Res 2004;57:249–55.[CrossRef][Medline]
  13. Turnbull JW. Asthma conceived as a learned response. J Psychosom Res 1962;6:59–70.[CrossRef][Medline]
  14. Danker PS, Miklich DR, Pratt C, Creer TL. An unsuccessful attempt to instrumentally condition peak expiratory flow rates in asthmatic children. J Psychosom Res 1975;19:209–13.[CrossRef][Medline]
  15. Dekker E, Groen J. Reproducible psychogenic attacks of asthma; a laboratory study. J Psychosom Res 1956;1:58–67.[CrossRef][Medline]
  16. Dekker E, Pelser HE, Groen J. Conditioning as a cause of asthmatic attacks: a laboratory study. J Psychosom Res 1957;2:97–108.[CrossRef][Medline]
  17. Franks CM, Leigh D. The theoretical and experimental application of a conditioning model to a consideration of bronchial asthma in man. J Psychosom Res 1959;4:88–98.[CrossRef][Medline]
  18. MacQueen G, Marshall J, Perdue M, Siegel S, Bienenstock J. Pavlovian conditioning of rat mucosal mast cells to secrete rat mast cell protease II. Science 1989;243:83–5.[Abstract/Free Full Text]
  19. Miller DJ, Kotses H. Classical conditioning of total respiratory resistance in humans. Psychosom Med 1995;57:148–53.[Abstract/Free Full Text]
  20. Sarafino EP, Gates M, DePaulo D. The role of age at asthma diagnosis in the development of triggers of asthma episodes. J Psychosom Res 2001;51:623–8.[CrossRef][Medline]
  21. Rietveld S, Everaerd W. Symptom perception in asthma. In: Kaptein AA, Creer TL, eds. Respiratory Disorders and Behavioral Medicine. London: Martin Dunitz; 2002:117–38.
  22. Rietveld S, Kolk AM, Prins PJM. The influence of lung function information on self-reports of dyspnea by children with asthma. J Pediatr Psychol 1996;21:367–77.[Abstract/Free Full Text]
  23. Rietveld S, Kolk AM, Prins PJM, Colland VT. The influence of respiratory sounds on breathlessness in children with asthma: a symptom-perception approach. Health Psychol 1997;16:547–53.[CrossRef][Medline]
  24. Devriese S, Winters W, Stegen K, Van Diest I, Veulemans H, Nemery B, Eelen P, Van de Woestijne KP, Van den Bergh O. Generalization of acquired somatic symptoms in response to odors: a Pavlovian perspective on multiple chemical sensitivity. Psychosom Med 2000;62:751–9.[Abstract/Free Full Text]
  25. Stegen K, De Bruyne K, Rasschaert W, Van de Woestijne KP, Van den Bergh O. Fear-relevant images as conditioned stimuli for somatic complaints, respiratory behavior and reduced end-tidal pCO2. J Abnorm Psychol 1999;108:143–52.[CrossRef][Medline]
  26. Van den Bergh O, Stegen K, Van de Woestijne KP. Learning to have psychosomatic complaints: Conditioning of respiratory behavior and complaints in psychosomatic patients. Psychosom Med 1997;59:13–23.[Abstract/Free Full Text]
  27. Van den Bergh O, Stegen K, Van de Woestijne KP. Memory effects on symptom reporting in a respiratory learning paradigm. Health Psychol 1998;17:241–8.[CrossRef][Medline]
  28. Global Initiative for Asthma. Asthma Management and Prevention. A Pocket Guide for Physicians and Nurses. National Institutes of Health, National Heart, Lung and Blood Institute. Available at: http://www.ginasthma.com/gina_pocket.pdf. Accessed July 7, 2004.
  29. Ritz T, Dahme B, DuBios AB, Folgering H, Fritz GK, Harver A, Kotses H, Lehrer PM, Ring C, Steptoe A, Van de Woestijne KP. Guidelines for mechanical lung function measurements in psychophysiology. Psychophysiology 2002;39:546–67.[CrossRef][Medline]
  30. Mazurek HK, Marchal F, Derelle J, Hatahet R, Moneret-Vautrin D, Monin P. Specificity and sensitivity of respiratory impedance in assessing reversibility of airway obstruction in children. Chest 1995;107:996–1002.[Abstract/Free Full Text]
  31. Eckert DJ, Catcheside PG, Smith JH, Frith PA, McEvoy RD. Hypoxia suppresses symptom perception in asthma. Am J Respir Crit Care Med 2004;169:1224–30.[Abstract/Free Full Text]
  32. Read DJ. A clinical method for assessing the ventilatory response to carbon dioxide. Australas Ann Med 1966;16:20–32.
  33. Kinsman RA, Luparello TJ, O’Banion R, Spector S. Multidimensional analysis of the subjective symptomatology of asthma. Psychosom Med 1973;35:250–67.[Abstract/Free Full Text]
  34. Ritz T, Bobb C, Edwards M, Steptoe A. The structure of symptom report in asthma. A reevaluation. J Psychosom Res 2001;51:639–45.[CrossRef][Medline]
  35. Put C. Symptom perception in asthma. Medicine meets psychology. [Doctoral dissertation]. Leuven: University of Leuven; 2001.
  36. Watson D, Clark LA, Tellegen A. Development and validation of brief measures of positive and negative affect: the PANAS scales. J Pers Soc Psychol 1988;54:1063–70.[CrossRef][Medline]
  37. Box GEP, Cox DR. An analysis of transformations. J R Stat Soc 1964;26:211–53.
  38. Thomas M. Breathing exercises and asthma. Thorax 2003;58:649–50.[Free Full Text]
  39. Thomas M, McKinley RK, Freeman E, Foy C. Prevalence of dysfunctional breathing in patients treated for asthma in primary care: cross sectional survey. BMJ 2001;322:1098–100.[Abstract/Free Full Text]
  40. Thomas M, McKinley RK, Freeman E, Foy C, Prodger C, Price D. Breathing retraining for dysfunctional breathing in asthma: a randomised control trial. Thorax 2003;58:110–5.[Abstract/Free Full Text]
  41. Stegen K, Van Diest I, Van de Woestijne KP, Van den Bergh O. Negative affectivity and bodily sensations induced by 5.5% CO2-enriched air inhalation: is there a bias to interpret bodily sensations negatively in persons with negative affect? Psychol Health 2000;15:513–25.[CrossRef]
  42. Stegen K, Van Diest I, Van de Woestijne KP, Van den Bergh O. Is there an attentional bias to internal stimuli in persons with negative affect? Cognition Emotion 2001;15:813–29.[CrossRef]
  43. Stegen K, Neujens A, Crombez G, Hermans D, Van de Woestijne KP, Van den Bergh O. Negative affect, respiratory activity and somatic complaints in a CO2 inhalation paradigm. Biol Psychol 1998;49:109–22.[CrossRef][Medline]
  44. Binks AP, Moosavi SH, Banzett RB, Schwartzstein RM. ‘Tightness’ sensation of asthma does not arise from the work of breathing. Am J Respir Crit Care Med 2002;165:78–82.[Abstract/Free Full Text]
  45. Lansing RW, Im BSH, Thwing JI, Legedza ATR, Banzett RB. The perception of respiratory work and effort can be independent of the perception of air hunger. Am J Respir Crit Care Med 2000;162:1690–6.[Abstract/Free Full Text]
  46. Mahler DA, Harver A. Do you speak the language of dyspnea? Chest 2000;117:928–9.[Free Full Text]
  47. Apter AJ, Affleck G, Reisine ST, Tennen HA, Barrows E, Wells M, Willard A, ZuWallack RL. Perception of airway obstruction in asthma: sequential daily analysis of symptoms, peak expiratory flow rate, and mood. J Allergy Clin Immunol 1997;99:605–12.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Psychosom. Med.Home page
T. Ritz, A. Steptoe, C. Bobb, A. H. S. Harris, and M. Edwards
The Asthma Trigger Inventory: Validation of a Questionnaire for Perceived Triggers of Asthma
Psychosom Med, November 1, 2006; 68(6): 956 - 965.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by De Peuter, S.
Right arrow Articles by Van den Bergh, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by De Peuter, S.
Right arrow Articles by Van den Bergh, O.
Related Collections
Right arrow Anxiety
Right arrow Pulmonary

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS