The effect of deriving breathhold time and alveolar volume in different ways and of subtracting the back pressure of carbon monoxide on the calculated diffusing capacity (DLCO) was examined in 398 young adult smokers, ex-smokers, and nonsmokers. Significantly higher values were obtained when the recommendations of the ATS-DLD Epidemiology Standardization Project were followed for calculating breathhold time and alveolar volume, respectively, compared to following those of the ECCS Standardized Lung Function Testing Project. We therefore recommend that in reporting DLCO results, an explicit statement be included on the methods used to calculate both. Significantly higher values for DLCO were also obtained when CO back pressure was taken into account. The magnitude of this effect in smokers makes it difficult not to recommend its inclusion in epidemiologic studies. In clinical laboratories, however, when correction for CO back pressure is omitted, it is unlikely to result in an underestimate of DLCO of more than 5%, even in smokers.