Our study shows that parental understanding of "wheezing" differs from most epidemiological definitions, and is distinct from health care professionals' perceptions. These results were obtained using a common Portuguese translation of the term used in a large-scale international questionnaire, and they are consistent with findings from previous studies mostly conducted in English-speaking countries [1, 10, 12, 15, 20]. Importantly, a relevant subgroup reported not being familiar with the term "wheezing", and we identified social, clinical and geographical characteristics associated with this outcome. The impact of these different perceptions must be considered when developing and using questionnaire instruments in epidemiological or intervention studies, and can also influence clinical practice.
Impact of parental perceptions of "wheezing"
In epidemiological studies and clinical practice, "wheezing" is usually defined as a whistling sound located in the chest [9, 10, 12–14]. We found that 34% parents or caregivers reported not knowing this term, and of those that did recognize it, 31% interpreted it differently. Such differences in conceptual understandings of "wheeze" by parents have been identified in previous studies, both quantitatively and qualitatively [10, 12, 15, 20, 21]. Few of these, however, were performed outside the UK . Our study was conducted in a Portuguese population, and we chose a translated term for "wheezing" that is commonly used in both clinical practice and large-scale epidemiological studies. In a recent population-based survey from a respiratory cohort, Michel et al reported a slightly lower proportion of parents not identifying "wheezing" as sound . In their study, however, a definition of "wheezing" was given before the questionnaire, which may explain differences between estimates. Parent's use of other respiratory sounds as synonyms for "wheezing" was also remarkably similar between studies. This suggests that variations of parents' understanding and interpretation of "wheeze" is present cross-culturally, in different settings, and is not exclusively a linguistic issue.
These findings may impact the accuracy with which "wheezing" prevalence rates are estimated through questionnaires alone. Evidence is conflicting when comparing parental assessments of wheezing and other respiratory sounds with different putative clinical "gold-standards". Studies in the acute care setting have shown considerable variation in agreement between clinicians and parents when using the term "wheezing" for the description of acute respiratory symptoms [10, 17]. Importantly, one study showing good agreement was conducted in Portuguese- and Spanish-speaking countries [17, 22]. Use of video recordings of children presenting with different respiratory sounds improves the accuracy of parental report and confirms that misclassification of sounds by parents is frequent, as there is limited agreement between written and video questionnaires for the term "wheeze" when using the English language [15, 23]. Additionally, children with clinically confirmed wheeze in the first years of life later have poorer lung function than those with parent-reported wheezing . Contradictory results may arise from the absence of an accurate "gold-standard" for "wheezing", which also reflects the heterogeneity and different dimensions of asthma and wheezing disorders [25, 26]. Michel et al have modeled bias from different degrees of parental misunderstandings of "wheezing" using various hypothetical scenarios, all of which showed considerable impact on epidemiological survey results . Our study adds to these results by showing that a considerable proportion of parents reported not knowing "wheezing". It is reasonable to assume that these parents are at a risk of misclassifying "wheezing" items in questionnaires, although we did not assess the direction or magnitude of bias. This adds complexity to the interpretation of questionnaire results. Further research is needed to assess what is the impact of different perceptions of "wheezing" in the accuracy with which parents recognize and report this symptom, and whether it varies in different settings, languages or cultures.
Variables associated with parental perceptions of "wheezing"
We identified subgroups of parents which were more frequently unfamiliar with the term "wheezing", based on characteristics which may be classified into three categories: social, clinical and linguistic. Younger and less educated parents or caregivers fit in the first category, and the second includes children attending well-child visits, with no prior history of respiratory disease themselves or their parents. A third category includes non-national first language (Portuguese in this case), and differences in understanding of the term according to geographical location. We hypothesize that the latter are due to region-specific terminologies regarding "wheezing" and other respiratory sounds, since they were independent from setting. Additionally, geographical location was also associated with the pattern of "wheezing" synonyms mentioned by parents. Linguistic considerations have been shown to be important when assessing respiratory symptoms, and there is less congruence across languages for "wheeze" than other terms . Our findings are likely applicable to other languages, including those for which no term for "wheeze" exists, and this evidence strengthens the need for adequate linguistic validation of multicentre and international respiratory questionnaires [27–29]. Overall, the subgroups we identified may be more prone to have biased estimates, and careful interpretation of survey results is warranted. Additional guidance in questionnaires may facilitate the understanding of the term in these populations.
The educational level of parents was the only variable associated with both being unfamiliar with "wheezing" and describing it inadequately, and no other associations were found with the latter. Accuracy of "wheezing" description has been shown to vary based on ethnic, cultural, clinical and linguistic parameters, as well as with the child's respiratory history, i.e. frequency and severity of previous "wheezing" . The fact that we excluded participants which reported not knowing wheezing may have limited the power to investigate these associations. Most predictors identified by Michel et al were consistent with the parameters we found associated with not knowing "wheezing", which reflects an overlap between not knowing the term and defining it inadequately . Educational levels of parents were associated with all study outcomes, in line with findings from qualitative studies highlighting the relevance of social, cultural and linguist backgrounds in parental understanding of respiratory symptoms [20, 21, 30].
Cues used by parents and health professionals to identify "wheezing"
Our findings show parents use multiple cues to identify "wheezing", which is consistent with previous results . Visual and tactile cues were often reported, and their association with the use of non-"wheezing" synonyms suggests that parents may confound different respiratory sounds. We also found differences between physicians and nurses/physiotherapists regarding the definition and location of "wheezing", which may have an impact in clinical practice. Of interest is the fact that some health professionals also often used visual cues to define "wheezing". This supports the variability in assessing this symptom, and highlights the difficulty in capturing the concept of "wheezing" with a single definition. The relevance of these cues and their validity when assessing wheezing and asthma in epidemiological studies should be considered for future questionnaires, with more precise and explicit symptom definitions.
This study did not use a large population-based approach, but relied on convenience sampling. However, we sought a priori to recruit participants from different social, clinical and geographical backgrounds, in different clinical settings. There was a high prevalence of wheezing in children of participants, possibly due to the large hospital-based population. We could expect, however, that this would overestimate adequate knowledge of "wheezing". We studied the understanding of respiratory symptoms, but did not compare them to any objective finding. Furthermore, our questionnaire used closed directed questions, which may have missed qualitative aspects of parental or caregiver perception. Our purpose was to mimic approaches susceptible of being used in larger-scale questionnaires, as well as to perform quantitative analysis. Our results were mostly based on parents of younger aged children. Other studies have assessed the accuracy with which children and adolescent perceive and self-report respiratory symptoms .