I read with interest the article in CHEST (September 2014) by Clark et al1 evaluating pulmonary function measures with respect to pulmonary CT scan findings among former vermiculite workers who were exposed to amphibole fibers. The study shows no statistically significant differences in lung function parameters between the pleural plaques only (PPO) group and the normal CT scan (NCTS) group. The authors indicate confidence “that the potential for type 2 error was minimal.”1 Determination of whether differences between independent groups can be statistically detected (ie, avoiding false negatives) depends upon the sample size, the true difference between groups, and the within-group SD. The mean percentage point difference between PPO and NCTS is 5.29, 5.72, 4.14, and 5.98 for FVC, FEV1, total lung capacity, and diffusing capacity of the lung for carbon monoxide, respectively. Based on the within-group SDs, a control to case ratio of 0.18, and an α level of 0.05, the given sample sizes for PPO and NCTS indicate that this study had only 15% to 34% power to detect the differences noted previously. This post hoc power analysis is limited to the data available in the published tables, and it could be further informed by the covariates in the model. Nevertheless, 80% is typically considered to be a minimal level of power necessary to avoid excess type 2 error.