The present study had several limitations. First, it was not possible to establish a causal relationship between low-level cadmium exposure and the development of COPD, because this was a cross- sectional study. Second, there were only a small number of COPD patients among the males who had never smoked. However, the ORs of the quartile groups indicated a positive correlation between blood cadmium levels and COPD prevalence in males who had never smoked. Third, the effects of drugs or occupational exposures could not be excluded, which could have affected lung function and/or cadmium levels. Fourth, it is known that pulmonary function test results vary depending on the skill of the individual conducting the test. However, results should vary for those with and those without COPD. Therefore, a non-differential bias towards the null hypothesis may exist. Fifth, individuals who did not have a blood cadmium level or did not undergo pulmonary function tests were excluded from this study. Therefore, the results may be subject to selection bias. Sixth, KNHANES did not measure urinary cadmium levels, and so blood cadmium levels were used instead. While a blood cadmium level is the optimum marker for recent cadmium exposure