Results

STUDY 1A

Parents recorded that their own child was wheezy in 160 of 235 questionnaires administered. Of these 10 were completed in translations. Of the remainder, 41 defined wheeze but recorded their child was not wheezy and 34 were incomplete. Data were not kept on non-respondents.

Most respondents were mothers (76%; CI 70–83%), had no other children with wheeze or asthma (79%; CI 73–86%), and did not consider themselves wheezy (68%; CI 60–75%). Children were aged between 5 months and 16 years (median 4 years 6 months). Responses to questions about age of onset of wheezing were grouped in age bands, as many were imprecise. A total of 88% (CI 83–93%) of children had started wheezing before 5 years of age.

Table 1 shows four groupings of responses to open questions: sound, difficulty in breathing (DIB), cough, and unwell. Interrater agreement for grouping was greater than 95%.

Open question groups were then classified as “sound”, “sound with other cue(s)”, “cough”, “cough with other cue(s)”, and “DIB and/or unwell” (non-sound) (table 2). This grouping allows for a comparison of parents' responses with definitions in the epidemiology questionnaires which are based on sound (table 4). In response to the open questions, 36% of parents defined wheeze as a “sound”, of which 11% (CI 6–15%) mentioned “whistling”. A total of 36% knew their child was wheezy by “sound” alone, 28% by sound with another cue, 9% by cough with another cue, and 26% reported only non-auditory cues (DIB/unwell). There was no relation between the length of time a child had been considered wheezy, and responses to “How do you know your child is wheezy?” grouped as sound, DIB, and sound with DIB (χ² < 2.0; p > 0.5). When asked for the main thing they noticed, 66% of parents used non-auditory cues.

In response to the closed questions (table 3), parents were invited to indicate as many cues as needed to describe how they knew their child was wheezy. A total of 38% (CI 30–45%) selected three; 23% did not select “what you hear”. When asked to select one of the cues listed as the main way of knowing their child was wheezing, 40% (CI 32–48%) selected “what you hear” and 41% (CI 33–49%) chose a non-auditory cue (what they felt and what they saw). Significantly more respondents indicated sound to the closed than to the open “main thing” questions (40% versus 18%; p < 0.002).

Although wheeze translates as “squeaking”, results on the 10 translations used suggested that sound was reported less than for the group as a whole, although numbers are too few to say with certainty.

STUDY 1B

Twelve studies reported in the literature were included. Table 4lists definitions of wheeze and response options. All 12 defined wheeze as an auditory sign, a “sound”. Eight studies gave a definition of wheeze of which seven used the word “whistling”. These included the following features: audible pitch (noisy breathing, high pitch) with a sound descriptor (whistle, squeak, musical) and location (chest, throat). One study also included “tightness” and “breathlessness”.3 In the four remaining studies, “whistling” in the chest was given as an alternative to wheeze. A “not sure” response option was reported in only one of the 12 studies.28

STUDY 2

In 10 months 296 forms were completed at A&E. Wheeze or asthma was reported in 139 patients by parents and/or clinicians. The remainder had other respiratory symptoms and were excluded from analysis. The children were aged between 4 months and 15 years 9 months (median 2 years 6 months). Interrater agreement for scoring responses was greater than 95%. Parents and clinicians agreed in 45% of occasions that the patient was wheezy or had asthma. Most disagreement was over reports of cough and upper airway noise (table 5).

Discussion

From the UK childhood surveys examined here, the prevalence of “current wheeze” (within the last 12 months) ranges from 5.5%23 to 19.6%19 and doctor diagnosed asthma from 1.2%28 to 22.9%.22 However, as we have suggested, reliance on parents' reports in such surveys may be problematic. There may be false negative responses because parents perceive wheeze as something other than the questionnaire definition. There could be false positive responses if parents mislabel noisy breathing owing to stridor, snoring, or other upper airway noise as “wheeze”. “Noisy breathing” appeared frequently in the definitions used by the epidemiology questionnaires we reviewed.

To try to clarify what parents understand by wheeze, we chose to ask parents from a large sample of children attending a chest clinic. Within this population are likely to be a large group of children who have wheezed. It is reasonable to assume that parents of these children are more likely to know what wheeze is than a sample of parents from the general population. Although translations were provided, many bilingual parents still preferred to answer in English. Parents who declined to answer a questionnaire included those who spoke languages where translations were unavailable or were illiterate. This is a limitation of this study and of written surveys in general. As “wheeze” does not always directly translate in some languages, it may not make sense to discuss a “non-existent” concept from outside an English framework. However, the problem of defining concepts is not limited to translating between different languages; there are regional differences in the language used to describe symptoms31and differences are likely to exist between different social groupings (age, class, gender) as well as between individuals.15 The definitions reviewed in these epidemiology studies show some differences in emphasis although they, and the professional nomenclature of lung sounds,32 remain focused on the auditory.

When our respondents were asked to define wheeze, responses of “DIB and/or being unwell” and “sound” scored similarly (36% each). However, 66% gave “DIB and/or being unwell” as the main way they knew their child was wheezy. From the equivalent closed question 34% chose “seeing DIB”. Similarly, “what you hear” in the closed question was selected more frequently than the equivalent responses grouped as “sound” to the open question. In other words, responses were more in agreement with the audible clinical concept of wheeze when prompted with a selection. However, in response to the “main thing” questions some respondents to both open and closed questions (86 and 24 of 160 respectively) gave more than one response or left the answer void. Most parents indicated that there were many ways they perceived wheeze in their child (open and closed questions). It is possible that identifying a single cue was for them not appropriate. Perhaps our most important finding was that 23% of respondents when prompted to select as many cues as needed, did not select “what you hear”.

The phrase “wheeze or whistling” in some epidemiology studies implies that wheeze is something other than whistling.33This is different from saying that wheeze is whistling and table 4reflects this distinction. Our parents rarely mentioned “whistling”, even though 45% of children had wheezed since infancy and presumably had contact with health care workers who might have used the word. Of those who do describe wheeze as “sound”, few mention the “whistling” that epidemiology questionnaires usually use to define wheeze. This is not to say that parents would not have answered positively to questions about wheezing in the chest in such surveys. We did not use a prompt of “whistling” in the closed questions but had we done so, we probably would have had positive responses.

Two epidemiology studies said that wheeze came from either the throat or the chest.22 28 Stating the chest alone is probably incorrect as “wheeze” may best be heard at the throat.34 Only one study provided a “don't know” option to questions about wheeze.28 Such an option would allow for the possibility of subjects not understanding or sharing the given definition, nor knowing whether their child has been “wheezy”.

Sometimes questionnaires ask questions relating to separate symptoms, for example, cough, wheeze, and breathlessness.22 In our study parents seemed to assess wheeze by multiple cues. In questionnaires however, symptoms are listed separately and clinicians also perhaps tend to consider them as discrete.35 Although wider definitions can result in misclassifications,36 a triad of reported cough, wheeze, and breathlessness22seems to relate best to a diagnosis of asthma when objective measurements have not been made. DIB was offered by respondents more frequently than breathlessness as an answer to our open questions. In a questionnaire, DIB may be a better term than breathlessness as a way of describing altered breathing.

While the first part of the study compared the understandings of parents with definitions given by epidemiologists, the second part compared the labelling of observations of wheeze or asthma by parents and clinicians. Of the A&E attendees whom clinicians diagnosed as wheezy or having asthma, 39% of parents described their children's problem with terms other than wheeze or asthma (cough, DIB, chesty or noisy breathing). These descriptors reflect what has been described in the first part of our study and cannot be considered “incorrect”. If there is a discrepancy in parents' and doctors' findings when the episode is severe enough for the child to be bought to A&E, then it would be reasonable to suggest a similar or greater discrepancy when a child is less unwell and is seen in the community. However, clinicians' findings in A&E reflect only what is present at the time of examination and not what parents observed before they arrived. Similarly the experience of the clinicians will have varied. It is already known that there is observer variability in clinicians' assessments of clinical features of chest disease.32 37 38 In our study, 19% of parents complained of cough where clinicians found “wheeze”. Parents are likely to report cough alone when wheeze is heard only on auscultation. Some parents reported wheeze or asthma where clinicians found no respiratory abnormalities on examination. This part of our study suggests there are both false positive and false negative reports of wheeze.

Our findings are similar to those of a study in Copenhagen where, using semistructured interviews, the differences in “conceptual vocabulary” between the parents of asthmatic children and clinicians indicated problems in doctor–patient communication.39Asthma vocabulary has also been shown to be different between and within ethnic groups.40 Perhaps more widespread use of objective tests for asthma is needed. The reliance of history taking when making a diagnosis of asthma has been questioned.41 42 A recent review suggested “Let parents use their own words. Do not offer the word ‘wheeze’ but wait to see if the parents use it. If they do, clarify what it means”.43 We suggest that lay understandings of wheeze and the language used to describe it, be considered in both epidemiology and clinical practice.