Validity of self-reported smoking cessation with carbon monoxide cut-off point – A systematic review and meta-analysis of diagnostic test accuracy studies

Abstract

Background and Objective: Breathalyzer is a small, portable, easy-to-use device that is used to determine the abstinence status of a smoker exhaling carbon monoxide, as well as to validate the self-reported smoking abstinence of an individual. Our study objective is to determine the different carbon monoxide cut-off points of a Breathalyzer that can be used to validate self-reported smoking cessation. Materials and Methods: Observational and interventional studies using Breathalyzer as a diagnostic test published between January 1980 and August 2020 were eligible. QUADAS-2 was used to assess the quality of the included studies. Pooled sensitivity and specificity were calculated using random-effects model. Sub-group analysis was carried out by using differing cut-off values and brands. Results: A total of 22 (18,337 participants) manuscripts were included for review, out of which 10 (10,385 participants) were considered in the meta-analysis. A cut-off of 7–10 ppm using piCO+ Smokerlyzer device and any of the Bedfont devices gave a pooled sensitivity through sensitivity analysis of 93.5 (95% confidence interval [CI]: 88–97) and specificity ranged from of 89.3 (95% CI: 77.7–95.2) to 93.6 (87.1–96.9) for any of the devices. When meta-analysis was performed by combining all devices and cut-off ranges from 2 to 10 ppm through sensitivity analysis, at 6–7 ppm, the sensitivity obtained was 95 (84.2–98.5) and specificity was 88.5 (71.7–95.9). Conclusion: Self-reported abstinence was validated by piCO+ Smokerlyzer by Bedfont Scientific at 6–7 ppm when all devices and various cut-off points were combined for analysis.

Keywords

• breathalyzer test
• carbon monoxide
• meta-analysis
• smoking cessation
• validation study