Accuracy of a mobile app to identify suspect asbestos-containing material in Australian residential settings

Our plain-English summary of Accuracy of a mobile app to identify suspect asbestos-containing material in Australian residential settings, by M.H. Govorko, L. Fritschi, A. Reid (Journal of Occupational and Environmental Hygiene, 2018).

The study

• Authors: M.H. Govorko, L. Fritschi, A. Reid
• Journal: Journal of Occupational and Environmental Hygiene
• Published: 2018;15(8):598-606
• Cited by: 2 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

In situ asbestos in the built environment is a remaining source of exposure in countries that have prohibited the manufacture and use of asbestos. However, it is difficult to identify in situ asbestos-containing material in residential settings. The objective of this study was to evaluate the accuracy of the mobile phone application ("app"), ACM Check, in identifying in situ asbestos located inside and outside of homes compared with onsite inspections conducted by an experienced environmental consultant. A cross-sectional study was undertaken that involved participants completing ACM Check on their homes built pre-1990 and located throughout metropolitan Perth, Western Australia, and an onsite inspection conducted at each home by an environmental consultant. Cohen's kappa statistic was calculated to evaluate the strength of agreement between the two methods. The 40 houses sampled were built between 1898 and 1988 with a median year of 1966. Thirty-eight (95%) homes had at least one type of material categorized as positive for asbestos by both ACM Check and the environmental consultant (κ = 1.00). Agreement between the two methods differed when categorizing specific materials as positive or negative for asbestos with substantial agreement for fencing (κ = 0.918), outbuilding walls (κ = 0.844), backing board to electrical meter box (κ = 0.826), exterior wall cladding (κ = 0.771), and interior walls (κ = 0.754), and fair agreement for outbuilding roofs (κ = 0.375) and interior flooring (κ = 0.304). ACM Check is a tool that can be used by tradespeople, home renovators, and householders to screen residential settings for the presence of in situ asbestos-containing material. Mobile phone apps have the potential to be developed or modified for use in other countries to help users identify asbestos and reduce their risk of asbestos exposure.

In plain English

This cross-sectional study tested ACM Check, a mobile phone app the authors developed to help people find possible asbestos-containing material in homes. Owners of 40 houses built before 1990 in metropolitan Perth, Western Australia, completed the app. An experienced environmental consultant then inspected each house. Both methods rated 38 of the 40 houses (95%) as having at least one asbestos-containing material. Across all 560 material ratings, the two methods agreed 505 times (90.2%). Compared with the consultant, the app had a sensitivity of 86% and a specificity of 91.2%. Agreement was substantial for fencing, outbuilding walls and electrical meter box backing boards, but only fair for outbuilding roofs and interior flooring. The app rated wall tile backing as positive in 21 houses where the consultant did not. Of 27 laboratory-tested samples, 19 contained asbestos, most of them both chrysotile and amosite. Against the laboratory results, the app had 100% sensitivity but 25% specificity. The authors concluded that the app showed promising results. They noted that only one consultant was used and that participants were self-selected.

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: Cross-sectional study comparing householders' use of the ACM Check app with onsite inspections by an environmental consultant, plus laboratory analysis of some samples; agreement measured with Cohen's kappa
• Who was studied: 40 owner-occupied houses built between 1898 and 1988 (median 1966) in metropolitan Perth, Western Australia, studied August 2016 to February 2017; 41 owners completed the app and one property was excluded
• Exposure: In situ asbestos-containing materials in homes (for example asbestos cement fencing, wall cladding, eaves, meter box backing boards, flooring), as identified by the authors' ACM Check app across 14 locations and by a consultant; laboratory analysis identified chrysotile and amosite
• Main results:
  • 38 of 40 houses (95%) had at least one material rated positive for asbestos by both the app and the consultant (kappa 1.00 per Table 2)
  • Across all 560 material observations, the app and consultant agreed in 505 (90.2%). Kappa was 0.718 (p less than 0.001), and the app had 86% sensitivity and 91.2% specificity compared with the consultant
  • For asbestos inside the house, agreement was only fair (kappa 0.318, p=0.013): the app rated 25 homes (62.5%) positive and the consultant rated 12 homes (30%) positive
  • The inside discrepancy was mainly due to the app rating wall tile backing as positive (21 false positives compared with the consultant)
  • Substantial agreement for fencing (kappa 0.918), outbuilding walls (kappa 0.844) and electrical meter box backing board (kappa 0.826)
  • Only fair agreement for outbuilding roof (kappa 0.375) and interior flooring (kappa 0.304)
  • Corrugated asbestos cement sheet fencing was the most common asbestos-containing material, present at 33 homes (82.5%)
  • Of 27 laboratory samples, the consultant classified 22 as asbestos-containing and the laboratory found 19 positive, 17 of them with both chrysotile and amosite
  • Agreement between the consultant and laboratory analysis was substantial (kappa 0.701, p less than 0.05)
  • Agreement between the app and laboratory analysis was fair (kappa 0.319, p less than 0.05), with 100% sensitivity but 25% specificity
  • Only 3% of assessments were false negatives, mostly materials in the eaves
• The authors' conclusion: The authors concluded that ACM Check can provide promising results to help people detect the presence or absence of in situ asbestos in the most common sites in Australian homes
• Limitations the authors state:
  • A single environmental consultant served as the reference; the authors said multiple consultants would have allowed a more rigorous evaluation
  • Participants were self-selected and may already have known whether their property had asbestos, which may have biased their app answers
  • Laboratory sampling was opportunistic and too small to calculate agreement for specific materials or locations
  • Low numbers of asbestos-containing materials in some areas may have affected the kappa values

Topics

Medical terms in this study, explained in our Lexicon: Asbestos-Containing Material (ACM), Asbestos Exposure.

Related studies

• Using a Mobile Phone App to Identify and Assess Remaining Stocks of In Situ Asbestos in Australian Residential Settings — M. Govorko, L. Fritschi, et al. (2019)
• Does the Presence of Asbestos-Containing Materials in Buildings Post-construction and Before Demolition Have an Impact on the Exposure to Occupants in Non-occupational Settings? — R. Singh, A.L. Frank (2023)
• Asbestos-containing materials in abandoned residential dwellings in Detroit — A. Franzblau, A.H. Demond, et al. (2020)
• Letter to Editor regarding Franzblau et al. 2020: Asbestos-containing materials in abandoned residential dwellings in Detroit, Science of Total Environment, v714,1-8 — A.L. Frank (2020)
• Response to Letter to the Editor regarding Franzblau et al., Asbestos-containing materials in abandoned residential dwellings in Detroit, from Prof. Arthur Frank, MD, PhD — A. Franzblau, A.H. Demond, et al. (2020)
• Worker exposure to asbestos during removal of sprayed material and renovation activity in buildings containing sprayed material — N.W. Paik, R.J. Walcott, et al. (1983)

Browse more research in Journal of Occupational and Environmental Hygiene, or return to medical research.