Industrial hygiene involvement in occupational epidemiology.
Explore the source record for details and available documents.
Epidemiologic Methods↗
Biomedical subjects
H Checkoway
Publications and source records attributed to H Checkoway.
A simple computer program for generating person-time data in cohort studies involving time-related factors.
The use of grouped data methods, such as standardized rate ratios and Poisson regression, for the analysis of cohort studies has a number of attractive features. This approach, however, has not been widely used in the past because of the difficulty of generating person-time data required for the computation of rates, particularly when stratification on time-related factors is involved. This paper presents a simple Statistical Analysis System (SAS) program for the generation of such data in a form that can be read directly by GLIM and used in a Poisson regression analysis. In addition to its simplicity, the program has the advantage of considerable flexibility and involves no restrictions on the number of time-related factors or the number of levels of each factor. Furthermore, it can be modified easily for multiple disease outcomes and for analyses of the latency period, or empirical induction time.
Humans↗
Ocular sarcoidosis. A case-control study among black patients.
A case-control study was conducted to investigate possible environmental risk factors for ocular sarcoidosis. The factors studied were exposures to pine products and occupational histories of employment in the tobacco, lumber, and textile industries. The exposure histories of 29 female and 15 male cases, patients with newly diagnosed ocular sarcoidosis that were attending the Glaucoma/Uveitis Clinic of the North Carolina Memorial Hospital, were compared to those of 110 female and 28 male controls selected from out-patient clinics. Elevated risks for sarcoidosis were detected among those females that were exposed to burning pine or were employed in the tobacco industry. For males, positive associations were observed for chewing pine products and for employment in the lumber and textile industries. These preliminary data are consistent with incomplete phagocytic clearance of nonbiodegradable foreign bodies present within our environment that may become antigenic to the immunologically susceptible host. Confirmation of refutation of these hypotheses awaits additional clinical and epidemiological research within the sarcoidosis belt.
Adolescent↗
Time-related factors as potential confounders and effect modifiers in studies based on an occupational cohort.
Time-related factors which are potential confounders and effect modifiers in studies based on an occupational cohort are reviewed. The most frequently considered ones include age at first exposure, duration of exposure, interval from exposure to disease recognition, and age at risk. These factors are related to the "healthy worker effect," which appears to be more pronounced among workers with the longest durations of employment and older ages, at date of hire, but weaker with longer length of follow-up and older age at risk. Hence, use of an internal comparison group may not eliminate bias since confounding will occur if the exposed and unexposed groups differ in their distributions across these factors. It is also shown, using the multistage model of carcinogenesis, that these factors may be important effect modifiers. Fortunately, generally straightforward methods of control exist both for stratified analyses and for the commonly used mathematical modeling approaches. Although no firm recommendations can be made, it would appear to be important to control for length of follow-up in the design or analysis of most studies based on an occupational cohort, and controlling for age at first exposure may also be desirable under many circumstances.
Adult↗
Epidemiologic programs for computers and calculators. Use of Poisson regression models in estimating incidence rates and ratios.
Summarizing relative risk estimates across strata of a covariate is commonly done in comparative epidemiologic studies of incidence or mortality. Conventional Mantel-Haenszel and rate standardization techniques used for this purpose are strictly suitable only when there is no interaction between relative risk and the covariate, and tests for interaction typically are limited to examination for departures from linearity. Poisson regression modeling offers an alternative technique which can be used for summarizing relative risk and for evaluating complex interactions with covariates. A more general application of Poisson regression is its utility in modeling disease rates according to postulated etiologic mechanisms of exposures or according to disease expression characteristics in the population. The applications of Poisson regression analysis to problems of summarizing relative risk and disease rate modeling are illustrated with examples of cancer incidence and mortality data, including an example of a nonlinear model predicted by the multistage theory of carcinogenesis.
Adolescent↗
Radiation, work experience, and cause specific mortality among workers at an energy research laboratory.
A retrospective cohort mortality study was conducted among 8375 white male employees who had worked at the Oak Ridge National Laboratory for at least one month between 1943 and 1972. This plant has been the site of energy related research, including uranium and plutonium reactor technology and radioisotope production. Radiation doses, primarily from gamma rays, were generally low; the median cumulative exposure for workers was 0.16 rems. Historical follow up was conducted for the years 1943-77 and ascertainment of vital status was achieved for 92.3% of the cohort. Standardised mortality ratios (SMRs) were computed to contrast the subjects' cause specific mortality experience with that of the United States white male population. The observed number of 966 deaths from all causes was 73% of the number expected. Mortality deficits were also seen for arteriosclerotic heart disease (SMR = 0.75; 344 observed) and all cancers (SMR = 0.78; 194 observed). These results are indicative of the healthy worker effect and the favourable influence on health of the cohort's relatively high socioeconomic status. Non-statistically significant raised SMRs were seen for all leukaemias (SMR = 1.49, 16 observed), cancer of the prostate (SMR = 1.16, 14 observed), and Hodgkin's disease (SMR = 1.10, 5 observed). Internal comparisons of mortality (standardised rate ratios, SRRs) were made between subgroups of the cohort according to radiation dose level and duration of employment in various job categories. No consistent gradients of cause specific mortality were detected for radiation exposure. Leukaemia mortality was highest among workers with greater than or equal to 10 years employment in engineering (SRR = 2.40) and maintenance (SRR = 3.12) jobs. The association of leukaemia with employment in engineering was unexpected; maintenance jobs entail potential exposures to radiation and to a wide range of organic chemicals; metals, and other substances.
Adolescent↗
Assessing the possible extent of confounding in occupational case-referent studies.
Case-referent studies nested within occupational cohort studies are efficient designs for estimating relative risks associated with exposures and for obtaining data on nonoccupational confounding factors. Frequently, however, only incomplete data on confounders can be obtained from sources such as next-of-kin or co-workers. A method for assessing the potential extent of confounding in cohort studies described previously by Axelson [Scand J Work Environ Health 4 (1978) 98-102] can be adapted for use in nested case-referent studies. An example is shown to illustrate that the adapted Axelson method is preferable to the usual confounder adjustment techniques when data on the confounder are incomplete.
Epidemiologic Methods↗
Identifying unexposed workers from occupational cohorts.
Explore the source record for details and available documents.
Epidemiologic Methods↗
Mortality among workers in the Florida phosphate industry. I. Industry-wide cause-specific mortality patterns.
A retrospective cohort mortality study was conducted on 17,601 white and 4,722 nonwhite male workers in the Florida phosphate mining and chemical processing industry. Concerns about potential risks from naturally-occurring sources of ionizing radiation and anecdotal reports of lung cancer among workers prompted this investigation. Historical follow-up was conducted for the years 1949 to 1978. In comparison with U.S. rates, small excess of mortality rates of lung cancer were observed for white (standardized mortality ratio [SMR] equals 1.22) and nonwhite workers (SMR = 1.24); however, these excesses disappeared when contrasts were made with prevailing rates in Florida. Emphysema mortality was also slightly elevated in comparison with U.S. rates; the SMRs were 1.48 and 1.73 for white and nonwhite workers, respectively. Neither disease was related to overall length of employment. Detailed analyses of mortality in relation to work assignments are presented in the companion article.
Adult↗
Mortality among workers in the Florida phosphate industry. II. Cause-specific mortality relationships with work areas and exposures.
Small excesses of lung cancer and emphysema mortality had been detected among an historical cohort of 17,601 white and 4,722 nonwhite Florida phosphate industry workers. Internal mortality rate comparisons were made between worker subcohorts classified according to length of employment in 16 work areas and according to employment duration in jobs grouped by potential exposures to 10 agents, including alpha radiation. The only consistent associations seen were increased lung cancer mortality rates among long-term workers in plant-wide services and skilled crafts jobs. There was no evidence to support causal associations with exposures characteristic of the phosphate industry.
Adult↗
An evaluation of the associations of leukemia and rubber industry solvent exposures.
Excessive leukemia mortality has appeared consistently in epidemiological studies of British and U.S. rubber industry workers. Attempts to identify causative factors have focused on exposure to benzene and other solvents. Interpretations of findings from these studies have often been influenced by expectations of a benzene/nonlymphocytic leukemia association, seen from previous work in other settings. However, data from the rubber industry studies have not been consistent with this expectation, as lymphocytic and nonlymphocytic leukemia have shown similar mortality excesses. Data from a small case-control study of lymphocytic leukemia are presented to illustrate an approach that considers multiple solvent exposures. The associations with lymphocytic leukemia risk observed for a number of solvents, most notably carbon tetrachloride and carbon disulfide, were stronger than those detected for benzene.
Adult↗
Cancer mortality and solvent exposures in the rubber industry.
Some evidence suggests that solvent exposures to rubber industry workers may be associated with excess cancer mortality, but most studies of rubber workers lack information about specific chemical exposures. In one large rubber and tire-manufacturing plant, however, historical documents allowed a classification of jobs based on potential exposures to all solvents that were authorized for use in the plant. A case-control analysis of a 6678 member cohort compared the solvent exposure histories of a 20% age-stratified random sample of the cohort with those of cohort members who died during 1964-1973 from stomach cancer, respiratory system cancer, prostate cancer, lymphosarcoma, or lymphatic leukemia. Of these cancers, only lymphosarcoma and lymphatic leukemia showed significant positive associations with any of the potential solvent exposures. Lymphatic leukemia was especially strongly related to carbon tetrachloride (OR = 15.3, p less than .0001) and carbon disulfide (OR = 8.9, p = .0003). Lymphosarcoma showed similar, but weaker, associations with these two solvents. Benzene, a suspected carcinogen, was not significantly associated with any of the cancers.
Adult↗
Lymphocytic leukemia and exposures to benzene and other solvents in the rubber industry.
Previous epidemiological studies on one cohort of rubber industry workers indicated that observed mortality excesses from lymphocytic leukemia were related to solvent exposures. The results of a detailed examination of the solvent exposure histories of 15 cases of lymphocytic leukemia from this cohort and 30 matched industry controls are presented. Extensive review of historical solvent use, raw materials specifications and job description records were used to reconstruct exposures to specific agents. Cases were 4.5 times as likely as controls to have had direct exposures to both benzene and other solvents. Analysis by manufacturing production source of solvent revealed that cases spent greater proportions of their work experience in jobs with potential exposures to coal tar-based benzene and xylene. No differences were seen for petroleum-derived solvents.
Adult↗
A hematology survey of workers at a styrene-butadiene synthetic rubber manufacturing plant.
A previous epidemiologic study of the U.S. rubber industry indicated that there has been an excess of leukemia and lymphoma mortality among hourly workers at one styrene-butadiene rubber manufacturing plant. This investigation was a combined industrial hygiene and hematology cross-sectional survey at the same plant. The objectives of the survey were to quantify exposure levels for styrene, butadiene, benzene and toluene, and to relate these levels to hematologic variation. Personal air samples and blood specimens were obtained from 157 production workers. All exposure levels for the four chemicals assayed were well below recommended standards. The higher mean styrene (13.67 ppm) and butadiene (20.03 ppm) concentrations were found in the Tank Farm area; in all other departments the mean levels for the four chemicals were less than 2 ppm. The Tank Farm workers had slightly lower levels of circulating erythrocytes, hemoglobin, platelets and neutrophils, and slightly higher mean corpuscular red cell volumes and neutrophil band counts than the other workers. Overall, in this population there was no pronounced evidence of hematologic abnormality, as determined from examination of peripheral blood.
Adult↗
Low level ionizing radiation and human mortality.
Explore the source record for details and available documents.
Adult↗
A case-control study of bladder cancer in the United States rubber and tyre industry.
A case-control study of bladder cancer was conducted in five United States rubber and tyre companies to determine if there were high-risk jobs and work areas within the industry. The study included 220 male cases of bladder cancer, of whom 107 were identified from hospital record reviews and 113 from death certificates. Each case was matched individually with two industry controls by sex, race, year of birth, and company. One control was matched additionally by year of hire and duration of employment. Comparisons of cases and controls not matched by year of hire and age of hire showed no differences for those variables, which suggests that age and calendar period of first exposure to the industry were not risk determinants. When the work histories of both cases and controls were contrasted it was found that cases were more likely than controls to have worked in milling (odds ratio (OR) = 1.91) and calender operation (OR = 2.21) jobs. The relative risk estimates for milling and calender operation both exhibited linear trends of increase with duration of exposure. Milling and calender operation jobs entail potential exposures to volatilised reaction products from heated rubber stock. A better understanding of aetiological associations with job type will require more detailed characterisation of the work environment with regard to the sources and levels of aromatic amines and other suspected bladder carcinogens.
Adult↗
Bias due to misclassification in the estimation of relative risk.
Lack of bias in the estimation of relative effect in epidemiologic studies depends on the internal validity of the study. This paper conveys in graphic and tabular form the direction and magnitude of bias due to misclassification of study subjects. A series of computer-generated graphs shows that the departure of the estimate of effect (relative risk or odds ratio) from its true value is a function of sensitivity and specificity (measures of classification validity), disease frequency, and exposure frequency. The discussion of bias emphasizes misclassification of the "outcome" variable; i.e., disease occurrence in a cohort study and exposure rate in a case-control study. Examples are used to illustrate that the magnitude of the bias can be large under circumstances which occur readily in epidemiologic research. When misclassification is equal for the two compared groups, the estimate is biased toward the null value, and in some instances beyond; when differential misclassification occurs (as in selective recall in case-control studies) the bias can be in either direction, and may be great. Formulas are derived to estimate the underlying true value of the relative risk or odds ratio using the investigator's observations together with the estimated sensitivity and specificity of the classification procedure.
Classification↗
Latency analysis in occupational epidemiology.
Allowance for prolonged disease induction and latency times is an important consideration in occupational epidemiology studies of cancer and other delayed effects of exposure. Two useful approaches for assessing prolonged induction and latency periods are (1) exposure lagging and (2) considering exposures only within moving time windows. The exposure weighting scheme proposed by Jahr2 to assess exposure burdens is another method that accounts for induction and latency, although not explicitly. These three approaches, which are shown to be special cases of exposure weighting, are illustrated with an analysis of lung cancer mortality among a cohort of workers from an asbestos textile plant.
Adult↗