Two biochemical markers effectively used to separate smokeless tobacco users from smokers and nonusers

Our plain-English summary of Two biochemical markers effectively used to separate smokeless tobacco users from smokers and nonusers, by D.B. Holiday, J.W. McLarty, R.H. Yanagihara, et al (Southern Medical Journal, 1995).

The study

• Authors: D.B. Holiday, J.W. McLarty, R.H. Yanagihara, L. Riley, S.B. Shepherd
• Journal: Southern Medical Journal
• Published: 1995;88(11):1107-13
• Cited by: 11 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Data from asbestos workers were used to devise a cutpoint classifier to identify subjects as Nonuser (non-tobacco user), Smokeless (exclusive smokeless tobacco user), and Smoker (ignited tobacco user). In some clinical trials and smoking cessation programs, Smokeless should be separated from Smoker. One therefore needs a marker for smoke exposure, such as thiocyanate, since nicotine levels, as measured by cotinine, could be similar in both groups. Levels of cotinine (ng/mL) and thiocyanate (mumol/L) levels (mean +/- SD) were, respectively: 320.9 +/- 201.1 and 145.9 +/- 63.7 for the Smoker group; 339.1 +/- 327.5 and 32.0 +/- 16.9 for the Smokeless group; and 0.6 +/- 2.6 and 58.2 +/- 33.2 for the Nonuser group. For Nonuser, Smokeless, and Smoker, respectively, the self-reported status was 45.1%, 10.8%, and 44.1%, which was adjusted to 42.2%, 11.6%, and 46.2%; the classifier yielded sensitivities of 100%, 76.1%, and 92.2%; specificities of 96.1%, 97.6%, and 96.4%; and predictive values of 94.9%, 80.6%, and 95.6%. The classifier successfully identified Nonusers, separated Smokeless from Smoker, and determined the prevalence of false reports in our cohort.

In plain English

This study used blood tests from 614 former asbestos workers enrolled in a lung cancer chemoprevention trial in eastern Texas, nearly all of them men. The authors built a simple rule to separate people who used no tobacco, people who used only smokeless tobacco, and smokers. The rule combined two blood markers: cotinine, which comes from nicotine, and thiocyanate, which is raised by tobacco smoke. Cotinine below 22.2 ng/mL meant Nonuser. Above that level, thiocyanate below 37.8 micromoles per liter meant Smokeless, and above it meant Smoker. After checking questionable answers, the researchers judged 20 of 614 self-reports (3.3%) to be false. Against these corrected categories, the rule identified all 259 nonusers, 54 of 71 smokeless users (76.1%) and 262 of 284 smokers (92.2%). The study did not analyze asbestos exposure. The authors concluded that the rule identified nonusers, separated smokeless users from smokers and measured false reporting in their group. They noted that it missed nearly a quarter of smokeless users, that its accuracy figures may be too optimistic, and that the findings apply only to similar populations.

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: Analysis of blood marker and questionnaire data collected within a clinical trial, used to develop and test a cutpoint rule for classifying tobacco use
• Who was studied: 614 former asbestos workers (598 men, 16 women) enrolled in the Beta-Carotene, Retinol, and Lung Cancer Chemoprevention Project, mostly from rural eastern Texas; 23 of 637 recruited subjects were excluded for reasons unrelated to tobacco use
• Exposure: Tobacco use (no tobacco, exclusive smokeless tobacco such as chewing tobacco or snuff, or smoking of cigarettes, pipes or cigars), measured by self-report and by serum cotinine and thiocyanate. Occupational asbestos exposure was a condition for entering the trial; fiber types and asbestos doses were not studied.
• Main results:
  • The best cutpoints were cotinine 22.2 ng/mL and thiocyanate 37.8 micromoles per liter. The rule agreed with self-reports in 91.5% of cases and with the corrected (adjusted) categories in 93.7% of cases.
  • Self-reports were 277 (45.1%) Nonuser, 66 (10.8%) Smokeless and 271 (44.1%) Smoker. After 20 false reports were reassigned (20/614, 3.3%), the categories were 259 (42.2%) Nonuser, 71 (11.6%) Smokeless and 284 (46.2%) Smoker. Of 277 self-reported Nonusers, 18 were reclassified, and 2 of 66 self-reported Smokeless users were reassigned to Smoker.
  • Mean cotinine (ng/mL) and thiocyanate (micromoles per liter) in the adjusted groups were: Smokeless 339.1 ± 327.5 and 32.0 ± 16.9; Smoker 320.9 ± 201.1 and 145.9 ± 63.7; Nonuser 0.6 ± 2.6 and 58.2 ± 33.2.
  • Nonuser: the rule identified all 259 true nonusers (sensitivity 100%), with specificity 96.1% and predictive value 94.9%.
  • Smokeless: the rule identified 54 of 71 true smokeless users (sensitivity 76.1%), with specificity 97.6%. Of 67 people the rule classed as Smokeless, 54 (80.6%) were truly Smokeless.
  • Smoker: the rule identified 262 of 284 true smokers (sensitivity 92.2%), with specificity 96.4% and predictive value 95.6%.
  • In the adjusted categories, a greater share of true smokeless users (9.9%) than of true smokers (3.9%) had reported no tobacco use.
• The authors' conclusion: The authors concluded that the classifier identified nonusers, separated smokeless tobacco users from smokers, and determined the prevalence of false reports in their cohort, which they put at about 3% to 4% overall.
• Limitations the authors state:
  • The same data were used to set the cutpoints and to measure accuracy, so the reported performance may be too optimistic.
  • The cutpoints missed nearly a quarter of actual smokeless tobacco users, and about one fifth of those classified as smokeless users were misclassified.
  • The conclusions apply only to populations similar to this predominantly male group of former asbestos workers.
  • Absolutely verifiable tobacco-use reports were not available for some subjects.
  • Thiocyanate was not adjusted for sex, depth of inhalation or diet, which also affect its levels.

Topics

Medical terms in this study, explained in our Lexicon: Clinical Trial.

Related studies

• Retinoid signaling and activator protein-1 expression in ferrets given beta-carotene supplements and exposed to tobacco smoke — X.D. Wang, C. Liu, et al. (1999)
• Tobacco-borne siderophoric activity — M.W. Qian, J.W. Eaton (1989)
• Tobacco and asbestos litigation — T.R. Goffe (1986)
• The joint effect of asbestos exposure, tobacco smoking and alcohol drinking on laryngeal cancer risk: evidence from the French population-based case-control study, ICARE — G. Menvielle, A. Fayossé, et al. (2016)
• Urinary Metabolite Diagnostic and Prognostic Liquid Biopsy Biomarkers of Lung Cancer in Nonsmokers and Tobacco Smokers — B. Dalal, T. Tada, et al. (2024)
• Mailing strategies and costs of recruiting heavy smokers in CARET, a large chemoprevention trial — B. Valanis, J. Blank, et al. (1998)

Browse more research in Southern Medical Journal, or return to medical research.