To more exactly assess the relationship between binaural
hearing impairment and smoking and eliminate the
confounding effect of age and duration of exposure
to noise, we used logistic regression analysis. In this
analysis (after eliminating confounding variables) there
was still a statistically significant relationship between
smoking and hearing impairment. To perform this
analysis, we divided the workers into two groups: those
suffering from hearing impairment and those without
any impairment. Then they were divided into two groups
according to their age (≤40 years old and more than 40
years old) and their exposure duration to noise (