Whether dioxin, or more specifically 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), has caused cancer in humans has been the subject of many epidemiology studies. A recent review concluded that the findings of these studies were not consistent, although the evidence was considered weak to strong for nonHodgkhVs lymphoma and too inconsistent for soft tissue sarcomas.1 More recently, however, two studies have been published that report consistent findings indicative of dioxin carcinogenicity. These two studies are an update, by Zober et al,2 of a study of workers exposed as a result of a 1953 accidental release of dioxin in Germany, and the National Institute for Occupational Safety and Health (NIOSH) Dioxin Registry study by Fingerhut et al.3 We will consider how these new studies add to our knowledge of the potential dioxin health effects. In 1977, a case report appeared, followed by a related case-control study in 1979, that suggested a connection between herbicide exposure and soft tissue sarcoma.4'5 Dioxin can be a contaminant in some herbicides. This finding appeared to receive some support from three occupational studies68 that reported a total of four soft tissue sarcomas, a relatively rare tumor, among small groups of workers exposed to relatively high levels of dioxin. These findings, as well as public concerns, motivated the formation of the NIOSH Dioxin Regis?

Three principles are presented that will improve the quality of epidemiologic studies now being done. These principles are: (1) that outcome foreknowledge should be disclosed and discussed; (2) that key data in the study should be well understood to include what influences it and its inherent variability; and (3) that detailed exposure characterization, whether it be occupational or personal risk factors, should be an integral part of all studies. The quality of epidemiologic studies resides in meticulous planning and execution, because not considering important confounders or poorly measuring an effect can mask or magnify associations.

A study of 200 persons working with benzene showed no differences in commonly measured hematologic outcomes when compared with 268 nonbenzene workers in the same plant. Exposures ranged from 0.01 ppm to a high of 1.40 ppm 8-hour time weighted average over a 10-year period. Several other factors (age, sex, race, and smoking), however, were associated with these outcomes, indicating the importance of considering confounding factors when comparing hematology results. Exposure to low levels of benzene does not appear to produce an increased level of abnormal hematology measures detectable in routine medical surveillance.

Recent reports have alleged that silver presents a toxic hazard to exposed workers. To define the potential risks of long-term exposure to silver better, a cross-sectional investigation was conducted of 27 Caucasian males occupationally exposed to primarily insoluble silver compounds and 27 matched referents. Physical examination and electron microscopy of skin biopsies revealed no cases of generalized argyria. Measurements of facial discoloration, judged from color photographs by panels of laymen and physicians, showed no significant difference between the two groups. Although 29% of the silver workers and none of the referents exhibited ocular silver deposition, optometric and contrast sensitivity test results revealed no significant deficits in visual performance. The kidney and respiratory findings were essentially normal in both populations. Despite the increased presence of silver in the blood, feces, and hair of the recovery workers versus the referents, there was no evidence that chronic silver exposure adversely affected the health of these employees.

To assess the potential chronic health effects of methylene chloride, the mortality experience of a maturing 1964 to 1970 cohort of 1,013 hourly men was evaluated through 1984. On average, employees were exposed at a rate of 26 ppm (eight-hour time-weighted average) for 22 years; median latency was 30 years. Compared with the general population, no statistically significant excesses were observed for such hypothesized causes as lung cancer (14 observed v 21.0 expected), liver cancer (0 v 0.8), and ischemic heart disease (69 v 98.1); dose-response relationships based on career methylene chloride exposure and latency were not demonstrated. Among nonhypothesized causes, a significant deficit was reported for total deaths (176 v 253.2). None of the industrial referent comparisons achieved statistical significance. Sufficient power was available to detect relative risks of 1.6 for lung malignancy and 1.3 for ischemic heart disease. In contrast, there was inadequate power to identify meaningful risk levels for hepatic cancer. With 14 combined lung and liver cancer deaths observed v 36.3 predicted (P less than .0001), the mortality estimate projected from a mathematical model derived from an animal bioassay substantially overestimated cancer mortality for these sites. This inconsistency emphasizes the need to incorporate epidemiologic evidence in assessing the human health risks associated with long-term exposure to this widely used solvent.

The mortality experience of a cohort of approximately 9,000 traced men employed at a Tennessee chemical plant was examined between 1972 and 1982. Statistically significant total mortality deficits of 41% to 46% and 14%, respectively, were observed compared with general population and occupational controls. Cancer deaths were 22% (significant) below expectation based on state and national vital statistics, whereas comparison with an employed group showed no difference. Significantly low mortality differentials were also reported for other major causes, including diseases of the circulatory, respiratory, and digestive systems. Analysis of the data by length of employment, payroll status, and chemical production division demonstrated no unusual patterns. The cohort's favorable mortality experience may be attributed to such factors as employee selectivity, health maintenance, accessibility to medical care, and less cigarette smoking.

An epidemiologic investigation of a 1964 cohort of 478 photographic processors in nine Eastman Kodak Color Print and Processing (CP&P) laboratories in the continental United States was undertaken. Analysis of pre-1965 environmental data indicated that chemical exposures were within the standards applicable at the time. The results showed no significant excess mortality, sickness-absence or cancer incidence. The processing cohort's mortality experience after 16 years of follow-up compared favorably with that of two industrial control populations. The present study represents (to our knowledge) the first published occupational epidemiology investigation that specifically examines the health of workers engaged in large-scale commercial photofinishing operations.

Epidemiology Section, Health, Safety and Human Factors, Laboratory, Eastman Kodak Company, Rochester, NY 14650

Preliminary data showing over-representation of the Eastman Kodak Company (Rochester) on death certificates of brain tumor patients, and higher risk for older workers when compared to the general population, led to a case-control epidemiologic study. Chemical exposure histories of 56 workers with brain tumors were compared with those of other Kodak employees. No differences were found in exposure to a variety of chemicals. In addition, employees with brain tumors were compared to other upstate New York brain tumor patients; there was no difference in histology. However, the Kodak employees had diagnoses more frequently confirmed by histologic examination and more thorough diagnostic studies. Thus, the apparent initial excess of diagnosed tumors may have resulted from a "diagnostic sensitivity bias" arising from more complete medical evaluation of Kodak employees.

A large male employee population with continuous low level work exposures to methylene chloride for up to 30 years was examined for mortality experience using non-concurrent prospective and proportionate mortality techniques. Mortality of the study group was consistent with industrial controls and less than the New York State general population. Life table analysis of the 1964 methylene chloride exposed cohort was followed through 1976 and demonstrated age group specific probabilities or 13 year survivorship equal to or better than that of three control male populations: Kodak Park, New York State, and United States.