The mobile-phone-based iCO(TM) Smokerlyzer(®): Comparison with the piCO(+) Smokerlyzer(®) among smokers undergoing methadone-maintained therapy

INTRODUCTION: The mobile-phone-based Bedfont iCO(TM) Smokerlyzer(®) is of unknown validity and reproducibility compared to the widely-used piCO(+) Smokerlyzer(®). We aimed to compare the validity and reproducibility of the iCO(TM) Smokerlyzer(®) with the piCO(+) Smokerlyzer(®) among patients reducing or quitting tobacco smoking. METHODS: Methadone-maintained therapy (MMT) users from three centers in Malaysia had their exhaled carbon monoxide (eCO) levels recorded via the piCO(+) and iCO(TM) Smokerlyzers(®), their nicotine dependence assessed with the Malay version of the Fagerström Test for Nicotine Dependence (FTND-M), and daily tobacco intake measured via the Opiate Treatment Index (OTI) Tobacco Q-score. Pearson partial correlations were used to compare the eCO results of both devices, as well as the corresponding FTND-M scores. RESULTS: Among the 146 participants (mean age 47.9 years, 92.5% male, and 73.3% Malay ethnic group) most (55.5%) were moderate smokers (6-19 cigarettes/day). Mean eCO categories were significantly correlated between both devices (r=0.861, p<0.001), and the first and second readings were significantly correlated for each device (r=0.94 for the piCO(+) Smokerlyzer(®), p<0.001; r=0.91 for the iCO(TM) Smokerlyzer(®), p<0.001). Exhaled CO correlated positively with FTND-M scores for both devices. The post hoc analysis revealed a significantly lower iCO(TM) Smokerlyzer(®) reading of 0.82 (95% CI: 0.69–0.94, p<0.001) compared to that of the piCO(+) Smokerlyzer(®), and a significant intercept of -0.34 (95% CI: -0.61 – -0.07, p=0.016) on linear regression analysis, suggesting that there may be a calibration error in one or more of the iCO(TM) Smokerlyzer(®) devices. CONCLUSIONS: The iCO(TM) Smokerlyzer(®) readings are highly reproducible compared to those of the piCO(+) Smokerlyzer(®), but calibration guidelines are required for the mobile-phone-based device. Further research is required to assess interchangeability.