Relation of occupational exposure to respiratory symptoms and asthma in a general population sample: self-reported versus interview-based exposure data

Our plain-English summary of Relation of occupational exposure to respiratory symptoms and asthma in a general population sample: self-reported versus interview-based exposure data, by P.S. Bakke, R. Hanoa, A. Gulsvik (American Journal of Epidemiology, 2001).

The study

• Authors: P.S. Bakke, R. Hanoa, A. Gulsvik
• Journal: American Journal of Epidemiology
• Published: 2001;154(5):477-83
• Cited by: 33 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

The objective of this study was to examine how the consistency of self-reported exposure to dust or gas, asbestos, and quartz varied between subjects with and those without respiratory symptoms and asthma in a Norwegian community sample () in 1987-1988. Exposure characterization obtained in a structured work history interview was used as the "gold standard." The authors also wanted to assess how the exposure-disease relation differed when the exposure was based on self-reported versus interview-obtained data. The prevalence of self-reported exposure to dust or gas, asbestos, and quartz was 28%, 5%, and 4%, respectively. The sensitivity of the self-reported exposure data varied from 21% to 64% and was higher in those with than in those without the respiratory disorders. The specificity varied from 78% to 100% and was lower in those with than in those without the respiratory disorders. The sex-, age-, and smoking-adjusted odds ratios of the respiratory disorders in those with exposure to dust or gas and to asbestos were only slightly reduced when misclassification was taken into account. The corresponding numbers for exposure to quartz were halved and lost their statistical significance when the misclassification was allowed for. In this general population sample, the self-reported occupational, airborne exposure data were differentially misclassified by disease status.

In plain English

This two-phase cross-sectional survey of 714 participants in Hordaland county, Norway, in 1987–1988 compared two ways of measuring past workplace exposure to dust or gas, asbestos and quartz. Participants answered yes-or-no questionnaire questions. Separately, an occupational physician who did not know those answers assessed exposure in a structured work history interview, which the authors used as the "gold standard." Exposure was 1.4 to 2.9 times more common by interview than by self-report. Self-reported asbestos exposure had a sensitivity of 28 percent and a specificity of 98 percent. For asbestos, these did not differ significantly between people with and without symptoms or asthma, except for chronic cough. The outcomes were respiratory symptoms and asthma, not asbestosis or mesothelioma. After adjustment for sex, age and smoking, the associations of asbestos and of dust or gas with these disorders were only slightly reduced when interview data replaced self-reports. In Table 5, two of the five significant quartz associations (morning cough and chronic cough) lost significance; the text described this as half. The authors concluded that self-reported exposure was misclassified differently depending on disease status.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Two-phase cross-sectional general population survey comparing self-reported (questionnaire) occupational exposure with exposure assessed in a structured work history interview, used as the gold standard
• Who was studied: 714 participants who attended the second phase (84 percent of those invited) in 1987–1988. They were drawn from a 1.9 percent random sample (4,992 people) of the general population of Hordaland county, Norway. The authors invited 91 percent of eligible people who reported a physician's diagnosis of asthma or emphysema and 22 percent of those who did not. Mean age was 42.1 years and 50.6 percent were men.
• Exposure: Self-reported ever-exposure at work to much dust or gas, to asbestos dust (fiber type not specified), and to quartz dust, answered yes or no. This was compared with exposure of at least 6 months to an airborne pollutant as coded by an occupational physician in a structured interview covering all jobs lasting more than 6 months.
• Main results:
  • Exposure prevalences based on the structured interview were 1.4 to 2.9 times higher than the self-reported prevalences. The difference was larger for asbestos and quartz than for dust or gas.
  • Overall, self-reported asbestos exposure had a sensitivity of 28 percent and a specificity of 98 percent compared with the interview.
  • For asbestos, the sensitivity and specificity of self-reported exposure did not vary significantly between those with and without the respiratory disorders. The only exception was chronic cough, for which sensitivity was significantly higher in those with the symptom.
  • Across the exposures, self-reported data had higher sensitivity and lower specificity in subjects with respiratory disorders than in those without.
  • The sex-, age- and smoking-adjusted odds ratios for the respiratory disorders in those exposed to dust or gas and to asbestos were only slightly reduced when interview-based exposure was used. None of the significant self-reported associations for these exposures lost significance.
  • For phlegm when coughing, the adjusted odds ratio for asbestos exposure was 2.33 (95% confidence interval 1.22, 4.45) with self-reported exposure and 2.05 (95% confidence interval 1.36, 3.08) with interview-based exposure.
  • For quartz, the text states that half of the significant self-reported exposure-disease associations lost significance when exposure was interview based.
  • In Table 5, five self-reported quartz associations were significant, and two of these lost significance when exposure was interview based. For morning cough, the odds ratio was 3.51 (1.61, 7.64) self-reported and 1.70 (0.99, 2.98) by interview. For chronic cough, it was 3.10 (1.38, 6.94) self-reported and 1.52 (0.76, 3.04) by interview. The quartz associations with phlegm, breathlessness and wheezing remained significant.
• The authors' conclusion: The authors concluded that self-reported occupational airborne exposure data in this general population sample were differentially misclassified by disease status. They stated that significant adjusted exposure-disease relations still persisted after this misclassification was taken into account.
• Limitations the authors state:
  • The interview data used as the gold standard are not necessarily correct; another expert might have reached a different exposure characterization.
  • The authors had no data on use of protective breathing equipment.
  • The self-reported exposure questions asked about ever being exposed, while the expert coded exposure job by job.
  • Response bias could have influenced the results.
  • Exposed subjects might have overreported respiratory symptoms.
  • The findings may apply only to communities with similar industrial structures.
  • The size of the reductions in odds ratios should be interpreted cautiously because of wide confidence intervals.

Topics

Medical terms in this study, explained in our Lexicon: Occupational Exposure, Work History, Odds Ratio, Statistical Significance.

Related studies

• Vapor, dust, and smoke exposure in relation to adult-onset asthma and chronic respiratory symptoms: the Singapore Chinese Health Study — T.D. LeVan, W.P. Koh, et al. (2006)
• Longitudinal pattern of reported respiratory symptoms and accelerated ventilatory loss in asbestos-exposed workers — C.A. Brodkin, S. Barnhart, et al. (1996)
• Exposure to asbestos-containing vermiculite ore and respiratory symptoms among individuals who were children while the mine was active in Libby, Montana — L.C. Vinikoor, T.C. Larson, et al. (2010)
• Surveillance of Australian workplace Based Respiratory Events (SABRE): notifications for the first 3.5 years and validation of occupational asthma cases — D. Elder, M. Abramson, et al. (2004)
• Irregular opacities in the lung, occupational asthma, and airways dysfunction in aluminum workers — K.H. Kilburn, R.H. Warshaw (1992)
• Prevalence of obstructive lung disease in a general population: relation to occupational title and exposure to some airborne agents — P.S. Bakke, V. Baste, et al. (1991)

Browse more research in American Journal of Epidemiology, or return to medical research.