METHODS

Study population

^NACMAAS is a prospective cohort study, described in detail elsewhere.¹³ Briefly, more than 1000 subjects were recruited in utero (8–10th week of pregnancy) by screening parents in the antenatal clinic using skin testing and a questionnaire regarding allergic diseases.¹⁴ We report here on 525 children at higher risk of allergic disease (one atopic parent) who formed the recruitment basis for the nested randomised study of the secondary prevention of asthma by the use of early treatment (as described in detail previously).¹⁴ The study had the approval of the local research ethics committee.

Study protocol

During the first three years of life, the study team contacted parents by telephone every three months to check on the health of the child. In addition, parents were asked to contact the study team immediately if their child wheezed. If parents reported that their child had wheezed, they were asked to visit their general practitioner (GP) immediately for verification. If they were unable to attend the GP’s surgery, a member of the study team would visit the child in their home on the same day to confirm that the child was wheezing.

All subjects were invited to a review clinic at 3 years of age (within four weeks of birthday) to perform sR_aw measurements. The children were classified as follows:

• Never wheezed (no parental or physician reported wheeze)

• Unconfirmed wheeze (parentally reported wheeze not confirmed by the GP or study physician on examination)

• Confirmed wheeze (parentally reported wheeze confirmed by physician examination).

Lung function

All children were asymptomatic at the time of the lung function assessment and were asked to refrain from β₂ agonist use for at least four hours before their appointment. sR_aw was measured by a single step procedure from the simultaneously measured changes of respiratory flow and changes of plethysmographic pressure, omitting the measurement of thoracic gas volume.¹⁵ Measurements were carried out during tidal breathing using a modified facemask (Astratech no. 2; Astra Denmark).⁸ The procedure was explained to the parent and the use of the facemask was demonstrated to the child. It was explained to the child that a video would start when they breathed through the mask. Children were encouraged to sit in the plethysmograph cabinet alone, but if they refused, the parent was asked to sit in the plethysmograph with the child on their knee.⁸ The door of the plethysmograph was closed and the subject asked to breathe through the facemask. The spirometry trace was observed and the child was encouraged to breathe at a respiratory rate of 30–45 breaths/min. If the child was accompanied the parent was asked to inhale and hold their breath for approximately 20 seconds. When the specific resistance loops returned to a normal shape after this manoeuvre and the spirometry trace was stable, sR_aw was calculated from the median of five technically acceptable loops. Three measurements of sR_aw were performed with each child, and the mean of these was used in the analysis.

Statistical analysis

Statistical analysis was carried out using SPSS for Windows, version 9.0. sR_aw measurements followed a log normal distribution, hence were subject to a log e-transformation prior to analysis. Results are thus reported as geometric means (GM). One way analysis of variance (ANOVA) was used to compare differences between the three symptom groups (never wheezed, unconfirmed wheeze, confirmed wheeze).

RESULTS

Five hundred and twenty five children were followed from birth. Four hundred and seventy five children were invited to a review clinic at 3 years of age (one died, 10 moved out of the area, two were lost to follow up, and 37 were withdrawn by parents). Data were collected on 454 children (fig 1).

One hundred and eighty six (41.0%) parents had reported that their child had wheezed, and 130 (28.6%) of these were confirmed by the primary care physician or a member of the study team. Of the 428 children who attended for review at age 3 years, 153 were unable to complete the lung function assessment (fig 1). Two hundred and seventy five children (64%) successfully completed pulmonary function assessment; of these, 76 were accompanied by a parent.

There was no significant difference in sR_aw between children who had never wheezed (n = 152, GM 1.03 kPa·s, 95% CI 1.00 to 1.06) and those with unconfirmed wheeze (n = 36, GM 1.02 kPa·s, 95% CI 0.96 to 1.07; p = 1.00). However, sR_aw was significantly higher in children with confirmed wheeze (n = 87, GM 1.17 kPa·s, 95% CI 1.11 to 1.22; p < 0.001) compared to both the children with no history of wheeze or those with unconfirmed wheeze (fig 2).

DISCUSSION

In our study we have reported that children with parentally reported wheeze which was not confirmed by a physician (unconfirmed wheeze) had identical lung function to those children who had no history of wheeze. However, children with physician confirmed wheeze had significantly poorer lung function compared to either those who had never wheezed or who had unconfirmed wheeze.

Limitations

Although parents in the study were asked to contact the study team if they heard their child wheeze, no doubt some “wheezy” episodes were not reported to the study team and the children were not seen by their physician at the time, either because the parents did not recognise wheeze or because they failed to act for whatever reason. These children would therefore have been categorised as either “never wheezed” or “unconfirmed wheeze” depending on whether the parent ever reported the episode to the study team. Despite this, there is still a clear difference in sR_aw between both the “never wheezed” and “confirmed wheeze” groups, and the “unconfirmed wheeze” and “confirmed wheeze” groups

A further limitation of the study is the large number of physicians who were involved in the assessment of wheeze in the children. It has previously been reported that inter-observer agreement on wheeze even among physicians is only moderate—that is, somewhere between chance and total agreement.¹⁶ Furthermore, the wheeze was not assessed concurrently by physician and parent, limiting the opportunity to distinguish transient wheeze from misinterpretation of wheeze.

Interpretation

The most likely explanation for our findings is that parents may be confusing other respiratory noises with wheeze. Other studies have previously reported discrepancies between parental and physician reports of wheeze. Cane et al investigated parents’ understanding of “wheeze” in children.¹¹ A questionnaire administered to parents of children attending a chest clinic revealed that parental understanding of wheeze differed from epidemiological definitions. Furthermore, parents of 139 children admitted to a paediatric A&E agreed with clinicians on the presence of wheeze in less than half of the cases. The most disagreement occurred over reports of cough and other upper airway noise.

When asked to identify audible breathing noises on video, only 59% of parents correctly identified wheeze.¹⁷ In a study of 92 infants with noisy breathing, 59% of parents initially described their child’s symptoms as wheeze, but after being shown video clips of various respiratory sounds including wheeze, ruttle, and stridor, only 36% still used the term wheeze to describe the symptoms.¹²

An alternative explanation for our results is that those children with only unconfirmed wheeze had very mild or extremely transient symptoms of wheeze which were not present when seen by the physician. These very mild symptoms appeared not to have caused any significant effect on specific airway resistance when measured at age 3 years. However, it is unlikely that symptoms would have completely disappeared within the hours from onset and the time of clinical examination, when symptoms caused sufficient parental concern for them to contact the study team in the first instance.

Conclusions

This study adds further weight to the argument that many parents have little understanding of what physicians mean by the term “wheeze”. Clearly parental perception of “wheeze” is very variable. The study shows the importance of using objective measures wherever possible in the diagnosis and also the follow up of children with reported wheeze. Epidemiological studies must be wary when interpreting questionnaire data of parentally reported symptoms and disease, as up to 30% of parentally reported wheeze was probably not a wheeze in the first place. There are also important implications for the recruitment of children into studies on the safety and efficacy of pharmaceutical interventions, which should be limited to those children with physician confirmed wheeze only.