DISCUSSION:

Regardless of which comparison population was used, this cohort experienced a twofold excess of larynx cancer, which was a statistically significant excess. Therefore, this study provides further evidence that exposure to acid mists is associated with larynx cancer. Although the numbers in this study were small, exposure were well characterized and an adjustment was made for smoking and drinking habits.

The fact that all the larynx cancers occurred among smokers or former smokers is not surprising, because virtually all larynx cancers in the general population also occur among smokers. Virtually no larynx cancers would have been expected to occur among the never-smokers in the cohort.

It is possible that acid mists acted as laryngeal carcinogens by facilitating the already strong carcinogenic effect of tobacco smoke.

About 75% of this cohort was exposed to sulfuric acid, and the remainder was largely exposed to hydrochloric acid. The data in our study were too sparse to be able to separate the effects of the different types of acid.

Among the live individuals in the study, nine men also reported benign growths on the vocal cords, one of which subsequently developed into larynx cancer. However, no expected numbers of such growths could be calculated because no standard rates were known, so it cannot be determined if these growths numbered more than might have been expected.

Further analyses were conducted by duration of exposure and time-since-first-exposure (potential latency), by dividing the person-years into less than five years exposure versus more than five years exposure, and less than 20 years since first exposure versus more than 20 years since first exposure. No clear differences were revealed between duration or potential latency groups with these analyses, which were based on small numbers.

QUESTIONS

After considering the above, try and answer the following questions:

  1. QUESTION 11. Aside from the problem of imprecision, can an analysis of larynx cancer risk by duration of exposure be considered a kind of "dose-response" analysis? Can the lack of trend in the duration analysis in this study be taken as a lack of a dose-response?