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.

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.

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.

Nutrient intakes among selected North American populations in the Lipid Research Clinics Prevalence Study: composition of fat intake.

Components of fat and their relationship to total energy are described for 2,368 white male and 2,200 white female adults, aged 20-59 years, for whom 24-hour dietary recalls were completed between 1972 and 1975 in nine North American populations as a part of the Lipid Research Clinics (LRC) Program Prevalence Study. Men had higher absolute intakes of total fat and cholesterol than women, although both sexes consumed diets similar in relative composition of polyunsaturated, saturated, and monounsaturated fatty acids. Marked geographical differences in intakes of energy, total fat, and dietary cholesterol were observed. Compared with data from dietary surveys conducted in the 1960's the LRC data showed that consumption of cholesterol had decreased by the early 1970's, whereas consumption of polyunsaturated fatty acids appears to have increased, resulting in a higher ratio of polyunsaturated to saturated fatty acids. However, comparison of current dietary guidelines with these data, which are based on a single dietary recall, showed that few LRC participants met the recommendations of the Senate Select Committee on Nutrition and Human needs for dietary fat intake.

Grouping and linear regression.

With a large number of observations, the method of grouping is often employed to provide simpler graphs or tables. When one investigates the relationship between two variables, one usually groups based on the magnitude of the independent variable, and then plots the dependent variable averages against independent variable averages to get a clearer graph. If grouping is based on the magnitude of the dependent variable, the plot of group means as indicated above does not appropriately describe the relationship of the dependent variable to the independent variable. These results are demonstrated theoretically for the special case of bivariate normality (and thus linear regression), but would be expected to be similar for other distribution assumptions. An example is given from an epidemiological study.

Common methods, different populations. The Lipid Research Clinics Program Prevalence Study.

The salient characteristics of the Lipid Research Clinics (LRC) Program Prevalence Study are presented, including an overview of its structure, the populations included and the common procedures used. Highlighted are the two sequential screening examinations that included 68,317 participants in 12 LRCs in the U.S. Canada and the U.S.S.R. for the first examination and 15,816 for the second. Several thousand more will be added as the thirteenth LRC completes its screening in Israel. Major data items are described briefly, as are data quality enhancement procedures. Procedures for plasma lipid determination, lipoprotein cholesterol, other clinical chemistries, nutritional intake assessment, and resting and stress electrocardiography are discussed. A discussion of the methodologic strengths and weaknesses of the findings of the LRC Program Prevalence Study is also included. This paper provides a frame of reference for the individual papers in this volume.

Associations of plasma high-density lipoprotein cholesterol with clinical chemistry data.

We examined data from the participants in the Lipid Research Clinics Program Prevalence Study for associations between high-density lipoprotein (HDL) cholesterol and clinical chemistry tests. There is a negative relationship between serum thyroxine and HDL cholesterol: men 20-69 years and women 20-44 years with low thyroxine levels have significantly higher HDL cholesterol than those with high thyroxine levels. Women 45-69 years with hyperglobulinemia (> 3.2 g/dl) have significantly lower HDL cholesterol levels than those with lower globulin levels. There is a weak negative association between HDL cholesterol and serum uric acid in men 20-44 years and a stronger association in women. A weak negative association between HDL cholesterol and plasma glucose is present only in men 20-44 years and women 45-69 years. Subjects with high serum bilirubin or serum aspartate aminotransferase (AST, SGOT) values have higher HDL cholesterol levels that are statistically significant in men 20-69 years and women 20-44 years (bilirubin) and men 45-69 years and women 45-69 years (AST). There is a negative association between alkaline phosphatase and HDL cholesterol in women and young men. These results suggest that thyroid hormones may be involved in the regulation of HDL cholesterol, and that there are associations between HDL cholesterol and the clinical chemistry tests that are not necessarily explained by disease. However, in the whole population, the plasma constituents measured as clinical chemistry tests, or the mechanisms that regulate their levels, are not important determinants of plasma HDL cholesterol levels.