Self-reported asthma was biased in relation to disease severity while reported year of asthma onset was accurate

doi:10.1016/j.jclinepi.2005.03.019

Journal of Clinical Epidemiology

Volume 59, Issue 1, January 2006, Pages 90-93

https://doi.org/10.1016/j.jclinepi.2005.03.019 Get rights and content

Abstract

Background and Objectives

The aims of the study were to assess the accuracy of self-reported asthma and notified year of asthma onset.

Methods

The study was performed on a sample of 365 subjects, 18–60 years old, with clinically diagnosed onset of asthma between 1983 and 1986. All subjects were investigated 10 years later, in 1996, with a respiratory questionnaire about the items of asthma and year of onset. The material was analyzed with logistic regression models.

Results

Of the 289 subjects who returned the questionnaire, asthma was reconfirmed in 251 subjects. In a logistic regression model, asthma severity was significantly associated with confirmation of asthma. The median difference between the “true” year of onset and the reported year 10 years later, the recall period was zero, with a 10th to 90th interpercentile range of –2 to 6 years. The recall period was not associated with asthma severity, bronchial hyperresponsiveness, smoking, atopy, or sex.

Conclusion

Self-reported asthma is biased in relation to disease severity, meaning that subjects with mild disease were less prone to report their asthma. Reported year of asthma onset among adults seems to be rather accurate, with no obvious dependent misclassifications.

Introduction

In epidemiologic studies of causes of asthma, operational definitions with high specificity are preferred [1]. Questions based on self-reporting or a physicians' diagnosis of asthma seem to be among those with the highest specificity [2]. High specificity implies a low sensitivity, that is, asthma definitions relying on self-reporting or a doctors' diagnosis increase the fraction of false negatives. Such bias is probably associated with the severity of disease; thus, subjects with mild asthma are probably more likely not to report their asthma diagnosis (or not at all visiting a physician) [3]. In addition to self-reporting of diagnoses other methods such as scoring of asthma symptoms have been suggested [4].

In studies assessing either the incidence of asthma or the risk for adult-onset asthma due to different occupational exposures or other time-dependent predictors, one option is to use the age of asthma onset [2]. In epidemiologic studies of asthma year of asthma onset has been used by several authors [5], [6], [7], [8], [9], [10]. Age of disease onset has also been used in studies of other diseases such as migraines [11], [12] and Raynaud's phenomenon [13].

Self-reported age of onset may involve a reporting error; the difference between the reported year and the “true” year has been termed “recall period” [11]. In a study of migraines it was found that older subjects tend to have a longer recall period [11]. Regarding asthma, no such studies have been identified.

The aim of the present study was to assess whether self-reporting of asthma and the recall period were associated with asthma severity, sex, age, atopy, or smoking.

Section snippets

Methods

The baseline study was performed in the city of Göteborg, and aimed to identify all subjects with recent-onset asthma, aged 18–60 years during the period 1983–1986, as previously described [14]. Altogether, 420 subjects with newly started asthma were identified. Forty-two had an earlier history of childhood asthma, and for the purposes of this study they were excluded. Further, seven subjects were deceased and six had moved abroad, altogether resulting in a population of 365 subjects with

Results

Of the 289 subjects that responded to the questionnaire, the question Do you have or have you ever had asthma? was answered “yes” by 250 subjects (87%) and the question Have you been told by a physician that you have asthma? was answered “yes” by 229 subjects (80%); 251 subjects (confirmers) responded affirmatively to one or the other or both of these items. Among the subjects responding to the questionnaire 59% (n = 171) had mild asthma, 39% (n = 113) had moderate asthma, and 2% (n = 5) had

Discussion

This study shows, not surprisingly, that self-reported asthma is biased by disease severity, that is, the false negatives are among those with mild disease. The reported year of asthma onset seems to be slightly misclassified, but less then anticipated.

One advantage of the present study is that the diagnoses of asthma at baseline are based on a combination of asthma symptoms and clinical physiologic parameters. The cohort was also selected from the only hospital and from primary care centers in

Acknowledgments

The study was supported by the Swedish Council Health Care Science (Vårdal), Herman Kreftings Fund for Asthma Research, Torsten and Ragnar Söderberghs Medical Foundation, and Swedish Council for Worklife and Social Research (FAS).

References (19)

K. Torén et al.
Asthma and asthma-like symptoms in adults assessed by questionnaires
Chest
(1993)
M. Grassi et al.
Asthma-like symptoms assessment through ECRHS screening questionnaire scoring
J Clin Epidemiol
(2003)
R. de Marco et al.
Incidence and remission of asthma: a retrospective study on the natural history of asthma in Italy
J Allergy Clin Immunol
(2002)
W. Stewart et al.
Estimating age incidence from survey data with adjustments for recall errors
J Clin Epidemiol
(1989)
B. Balder et al.
Predictors of self-assessed work ability among subjects with recent-onset asthma
Respir Med
(1998)
K.T. Copeland et al.
Bias due to misclassification in the estimation of relative risk
Am J Epidemiol
(1977)
L. Nathell et al.
Determinants of undiagnosed asthma
Allergy
(2001)
S.J. Brisman et al.
Occurrence of self-reported asthma among Swedish bakers
Scand J Work Environ Health
(1995)
E.A. Eisen et al.
A strategy to reduce healthy worker effect in a cross-sectional study of asthma and metalworking fluids
Am J Ind Med
(1997)

There are more references available in the full text version of this article.

Cited by (96)

Identification of true chemical respiratory allergens: Current status, limitations and recommendations
2024, Regulatory Toxicology and Pharmacology
Asthma in the workplace is an important occupational health issue. It comprises various subtypes: occupational asthma (OA; both allergic asthma and irritant-induced asthma) and work-exacerbated asthma (WEA). Current regulatory paradigms for the management of OA are not fit for purpose. There is therefore an important unmet need, for the purposes of both effective human health protection and appropriate and proportionate regulation, that sub-types of work-related asthma can be accurately identified and classified, and that chemical respiratory allergens that drive allergic asthma can be differentiated according to potency. In this article presently available strategies for the diagnosis and characterisation of asthma in the workplace are described and critically evaluated. These include human health studies, clinical investigations and experimental approaches (structure-activity relationships, assessments of chemical reactivity, experimental animal studies and in vitro methods). Each of these approaches has limitations with respect to providing a clear discrimination between OA and WEA, and between allergen-induced and irritant-induced asthma. Against this background the needs for improved characterisation of work-related asthma, in the context of more appropriate regulation is discussed.
Identification by cluster analysis of patients with asthma and nasal symptoms using the MASK-air® mHealth app
2023, Pulmonology
The self-reporting of asthma frequently leads to patient misidentification in epidemiological studies. Strategies combining the triangulation of data sources may help to improve the identification of people with asthma. We aimed to combine information from the self-reporting of asthma, medication use and symptoms to identify asthma patterns in the users of an mHealth app.
We studied MASK-air® users who reported their daily asthma symptoms (assessed by a 0-100 visual analogue scale – “VAS Asthma”) at least three times (either in three different months or in any period). K-means cluster analysis methods were applied to identify asthma patterns based on: (i) whether the user self-reported asthma; (ii) whether the user reported asthma medication use and (iii) VAS asthma. Clusters were compared by the number of medications used, VAS asthma levels and Control of Asthma and Allergic Rhinitis Test (CARAT) levels.
We assessed a total of 8,075 MASK-air® users. The main clustering approach resulted in the identification of seven groups. These groups were interpreted as probable: (i) severe/uncontrolled asthma despite treatment (11.9-16.1% of MASK-air® users); (ii) treated and partly-controlled asthma (6.3-9.7%); (iii) treated and controlled asthma (4.6-5.5%); (iv) untreated uncontrolled asthma (18.2-20.5%); (v) untreated partly-controlled asthma (10.1-10.7%); (vi) untreated controlled asthma (6.7-8.5%) and (vii) no evidence of asthma (33.0-40.2%). This classification was validated in a study of 192 patients enrolled by physicians.
We identified seven profiles based on the probability of having asthma and on its level of control. mHealth tools are hypothesis-generating and complement classical epidemiological approaches in identifying patients with asthma.
Nonrespiratory Diseases in Adults Without and With Asthma by Age at Asthma Diagnosis
2023, Journal of Allergy and Clinical Immunology: In Practice
Chronic nonrespiratory diseases are seemingly more prevalent in subjects with than without asthma, and asthma seems to differentiate by age of onset. However, studies with comparison of nonrespiratory diseases in subjects with and without asthma, considering asthma age of onset, are scarce.
To compare the quantity and type of chronic nonrespiratory diseases in adults with and without asthma considering age at asthma diagnosis.
In 2016, a FinEsS questionnaire was sent to 16,000 20- to 69-year-old adults randomly selected in Helsinki and Western Finland populations. Physician-diagnosed asthma was categorized to early (0-11), intermediate (12-39), and late-diagnosed (40-69 years).
A total of 8199 (51.5%) responded, and 842 (10.3%) reported asthma and age at diagnosis. In age and sex-adjusted binary logistic regression model, the most represented nonrespiratory disease was treated gastroesophageal reflux disease in early-diagnosed (odds ratio, 1.93; 95% CI, 1.17-3.19; P = .011) and osteoporosis in both intermediate-diagnosed (odds ratio, 3.45; 95% CI, 2.01-5.91; P < .001) and late-diagnosed asthma (odds ratio, 2.91; 95% CI, 1.77-4.79; P < .001), compared with subjects without asthma. In addition, gastroesophageal reflux disease, depression, sleep apnea, painful condition, and obesity were significantly more common in intermediate- and late-diagnosed asthma compared with without asthma, and similarly anxiety or panic disorder in intermediate-diagnosed and hypertension, severe cardiovascular disease, arrhythmia, and diabetes in late-diagnosed asthma. In age-adjusted analyses, having 3 or more nonrespiratory diseases was more common in intermediate (12.1%) and late-diagnosed asthma (36.2%) versus without asthma (10.4%) (both P < .001).
Nonrespiratory diseases were more common in adults with asthma than in adults without asthma. The type of nonrespiratory diseases differed, and their frequency increased by increasing age at asthma diagnosis.
Age at asthma diagnosis is related to prevalence and characteristics of asthma symptoms
2022, World Allergy Organization Journal
Although asthma may begin at any age, knowledge about relationship between asthma age of onset and the prevalence and character of different symptoms is scarce.
The aim of this study was to investigate if adult-diagnosed asthma is associated with more symptoms and different symptom profiles than child-diagnosed asthma.
A FinEsS postal survey was conducted in a random sample of 16 000 20-69-year-old Finnish adults in 2016. Those reporting physician-diagnosed asthma and age at asthma diagnosis were included. Age 18 years was chosen to delineate child- and adult-diagnosed asthma.
Of responders (N = 8199, 51.5%), 842 (10.3%) reported asthma diagnosis. Adult-diagnosed asthma was reported by 499 (59.3%) and child-diagnosed by 343 (40.7%). Of responders with adult-diagnosed and child-diagnosed asthma, 81.8% versus 60.6% used asthma medication (p < 0.001), respectively. Current asthma was also more prevalent in adult-diagnosed asthma (89.2% versus 72.0%, p < 0.001). Risk factors of attacks of breathlessness during the last 12 months were adult-diagnosis (OR = 2.41, 95% CI 1.64–3.54, p < 0.001), female gender (OR = 1.49, 1.07–2.08, p = 0.018), family history of asthma (OR = 1.48, 1.07–2.04, p = 0.018) and allergic rhinitis (OR = 1.49, 1.07–2.09, p = 0.019). All the analysed asthma symptoms, except dyspnea in exercise, were more prevalent in adult-diagnosed asthma in age- and gender-adjusted analyses (p = 0.032-<0.001) which was also more often associated with 5 or more asthma symptoms (p < 0.001) and less often with non-symptomatic appearance (p < 0.001) than child-diagnosed asthma.
Responders with adult-diagnosed asthma had more often current asthma and a higher and multiform asthma symptom burden, although they used asthma medication more often compared to responders with child-diagnosed asthma.
Asthma Remission by Age at Diagnosis and Gender in a Population-Based Study
2021, Journal of Allergy and Clinical Immunology: In Practice
Citation Excerpt :
This also applied to our findings, suggesting that misinterpretation of early-onset relapsed asthma to adult-diagnosed asthma would not have caused a major bias. On the other hand, retrospective self-reported asthma age of onset assessment is previously found to be very specific.39,44 For these reasons, we consider the reported asthma diagnoses in the present study to be precise and comprehensive, and only little misclassification of early-diagnosed asthma to late-diagnosed asthma or asthma to COPD in the older age groups.
Child-onset asthma is known to remit with high probability, but remission in adult-onset asthma is seemingly less frequent. Reports of the association between remission and asthma age of onset up to late adulthood are scarce.
To evaluate the association between asthma remission, age at diagnosis and gender, and assess risk factors of nonremission.
In 2016, a random sample of 16,000 subjects aged 20 to 69 years from Helsinki and Western Finland were sent a FinEsS questionnaire. Physician-diagnosed asthma was categorized by age at diagnosis to early- (0-11 years), intermediate- (12-39 years), and late-diagnosed (40-69 years) asthma. Asthma remission was defined by not having had asthma symptoms and not having used asthma medication in the past 12 months.
Totally, 8199 (51.5%) responded, and 879 reported physician-diagnosed asthma. Remission was most common in early-diagnosed (30.2%), followed by intermediate-diagnosed (17.9%), and least common in late-diagnosed asthma (5.0%) (P < .001), and the median times from diagnosis were 27, 18.5, and 10 years, respectively. In males, the corresponding remission rates were 36.7%, 20.0%, and 3.4%, and in females, 20.4%, 16.6%, and 5.9% (gender difference P < .001). In multivariable binary logistic regression analysis, significant risk factors of asthma nonremission were intermediate (odds ratio [OR] = 2.15, 95% confidence interval: 1.37-3.36) and late diagnosis (OR = 11.06, 4.82-25.37) compared with early diagnosis, chronic obstructive pulmonary disease (COPD) (OR = 5.56, 1.26-24.49), allergic rhinitis (OR = 2.28, 1.50-3.46), and family history of asthma (OR = 1.86, 1.22-2.85). Results were similar after excluding COPD.
Remission was rare in adults diagnosed with asthma after age 40 years in both genders. Late-diagnosed asthma was the most significant independent risk factor for nonremission.
Relationship between health-related quality of life and subjective wellbeing in asthma
2021, Journal of Psychosomatic Research
Citation Excerpt :
However, the raw AQLQ-S total scores in our study were similar (mean 14.2 versus 13.9) to baseline scores of patients with mixed levels of asthma control (54.4% well-controlled asthma, 22.4% poorly controlled, and 23.1% very poorly controlled) reported in other studies [61]. It has also been reported that patients with mild disease might be less likely to self-report asthma compared to those with more severe asthma [62]. Although AQLQ-S is weakly correlated to physiological markers of severity (i.e. percent predicted forced expiratory volume in 1 s (FEV1)), it shows strong correlations to the number of asthma medications, symptom frequency, and prior hospitalisations [40,43,63], which are probably more accurate indications of asthma severity [40].
Poor health-related quality of life (HRQoL) is associated with a high risk of acute exacerbations in patients with asthma. The use of health state utility instruments to measure HRQoL is common, but it has been criticised for failing to adequately capture the impact of the condition from the patient perspective. We aimed to assess the ability of subjective wellbeing and health state utility instruments to capture important HRQoL dimensions for asthma patients.
Regression analyses were conducted using data (n = 856 asthma patients) from a large multi-national cross-sectional survey to explore the relationship between subjective wellbeing (ONS4, PWI, SWLS), health state utility (15D, AQoL-8D, EQ-5D-5L, HUI3, SF-6D, SF-6Dv2), and an asthma-specific instrument (AQLQ-S). The latent structure of different instruments was investigated using an exploratory factor analysis.
All instruments could distinguish the impact of asthma on HRQoL. Among health utility instruments, 15D, SF-6D and AQoL-8D were more sensitive based on the average standardised regression coefficients. AQOL-8D had the strongest correlation with subjective wellbeing. Nevertheless, the exploratory factor analysis showed a complementary relationship between subjective wellbeing and HRQoL instruments. The most important subjective wellbeing domains were standard of living, achieving in life, and personal relationships; although contributions of these domains to overall life satisfaction differed among asthma patients identified by a latent class analysis.
The use of subjective wellbeing alongside AQLQ-S and health state utility instruments provides a comprehensive approach to assess the impact of asthma on the patient's life. Further research is needed to incorporate these measures in health care evaluation.

View all citing articles on Scopus

View full text

Original ArticleSelf-reported asthma was biased in relation to disease severity while reported year of asthma onset was accurate

Abstract

Background and Objectives

Methods

Results

Conclusion

Introduction

Section snippets

Methods

Results

Discussion

Acknowledgments

Chest

J Clin Epidemiol

J Allergy Clin Immunol

J Clin Epidemiol

Respir Med

Bias due to misclassification in the estimation of relative risk

Am J Epidemiol

Determinants of undiagnosed asthma

Allergy

Occurrence of self-reported asthma among Swedish bakers

Scand J Work Environ Health

A strategy to reduce healthy worker effect in a cross-sectional study of asthma and metalworking fluids

Am J Ind Med

Original Article
Self-reported asthma was biased in relation to disease severity while reported year of asthma onset was accurate