Notes Washed Up in a Bottle

Don. W. MacCorquodale M.D. M.S.P.H.

Periodic Notes From the Field on Philosophy and Science.

PESTICIDE EXPOSURE AND MORTALITY

In order to gain insight into cancer and other disease patterns, the authors assembled a cohort of private and commercial pesticide applicators and their spouses with detailed information on life style and agricultural exposures (Blair A et al. Mortality among participants in the agricultural health study. Annals of Epidemiology 2005; 15: 279-285). The Agricultural Health Study is a prospective study of slightly more than 57,000 licensed pesticide applicators, including about 52,000 private applicators (almost entirely farmers), nearly 5,000 commercial applicators from Iowa (not included in this analysis), and about 32,5000 spouses of private applicators.

Over 82% of the applicators completed a 21 page enrollment questionnaire. A second questionnaire covering lifestyle, pesticide application and other agricultural activities was given to participating applicators. Wives filled out another questionnaire.

The enrollment questionnaire sought information on crops, livestock, pesticide application, alcohol consumption, tobacco use, fruit and vegetable intake, medical conditions, and the usual demographic characteristics. The female questionnaire covered reproductive history.

Deaths were identified through the National Death Index and state mortality bases. Less than 1% of the cohort has been lost to follow-up. Standardized mortality ratios, SMRs, were calculated for major causes of death and selected cancers to compare deaths among applicators and their spouses with mortality patterns for the general population in each state.

The average age at entry was 48 years. The average follow-up time was 5.3 years with more than 400,000 person-years accumulated and 2,055 deaths. The SMR for all cause mortality was 0.5. Statistically significant deficits were found for all cancers combined and many individual causes of death, including diabetes, cardiovascular disease, COPD, nephritis, suicide, and cancers of the buccal cavity and pharynx, esophagus, pancreas, lung, prostate, bladder. No statistically significant excesses were observed.

The authors suggested that some observed deficits were undoubtedly due to the healthy worker effect. Of equal importance, they found that only 15% of farmers and 10% of their spouses were tobacco users at the time of enrollment. Alcohol use did not appear to differ from that of the general population. The investigators noted that typically farming requires more physical activity than many other occupations, and physical activity is known to be protective against a number of diseases. Historically, however, rates of injuries and accidental death among farmers are among the highest of any occupational group. However, the authors found an SMR for accidental death for their sample of only 1.0 for applicators and of 0.6 for spouses.

COMMENT: I found this paper fascinating. I suspect that many lay people regard exposure to pesticides as very dangerous, which is not born out in this study. It would have been most interesting to see if mortality varied by degree of exposure to pesticides, but of course, that would have been very difficult to quantify.

MERCURY EXPOSURE FROM DENTAL FILLINGS DURING PREGNANCY

The authors conducted a population based, case-control study of mercury-containing fillings placed during pregnancy and the risk of low birth weight (Hujoel PP et al. Mercury exposure from dental fillings during pregnancy and low birth weight risk. Am J Epidemiol 2005; 161: 734-740.

The study linked dental utilization data from a not-for-profit dental insurance company and Vital Records birth certificates for Washington State . More than 29,000 matches were successful.

Low birth weight was defined as any liveborn infant weighing less than 2,500 g. Only singleton births were included. The number of procedures, i. e., resin-based fillings, amalgams, and crowns, received since the last menstrual period was calculated and summarized using three statistics: the insertion of at least one filling during pregnancy (yes/no); the number of teeth filled per woman, and the number of surfaces filled per woman.

The birth certificate records and dental utilization provided information on a number of potentially confounding variables including: dental radiation doses, maternal age at the time of delivery, ethnicity, marital status, parity, maternal smoking during pregnancy (self-reported), gestational or established diabetes, adequacy of prenatal care, maternal educational attainment, self-reported alcohol consumption during pregnancy, prepregnancy weight, preeclampsia, chronic hypertension, and eight different types of dental care procedures.

Mercury amalgam fillings had been placed during pregnancy in 3.5 percent of the women with a low birth weight infant and in 5.3 percent of the women with an infant weighing 2,500 g or more at birth. There were no substantial differences with regard to the number of resin-based fillings, sealants or crowns during pregnancy. Mothers of low birth weight infants were more likely to be less than 20 years of age, less likely to be Caucasian, and more likely to have had inadequate prenatal care. Cigarette smoking during pregnancy, established or gestational diabetes, zero parity, and preeclampsia were also identified as risk factors for low birth weight.

The odds for a low birth weight infant were significantly lower for women who had one or more dental amalgams placed during pregnancy, OR = 0.65, 95% CI: 0.46, 0.92. After adjustment for risk factors for low birth weight, placement of amalgam fillings was no longer statistically significant, OR = 0.68, 9s5% CI: 0.42, 1.09.

COMMENT: The findings of this study do not, of course, prove that placing dental amalgams during pregnancy is without risk regarding low birth weight; however, they are consistent with the concept of no effect.

I believe that newspapers and magazines would perform a useful public service by refusing to publish the results of initial observational studies regarding health effects. I remember when the first reports appeared in the lay press about possible adverse health effects from mercury amalgam fillings when a dear friend – and an intelligent one – asked me if he should have all of his fillings removed and replaced with resin fillings. Luckily, I was able to persuade him to leave them alone. I would not be surprised to learn that others concerned about such fillings were not so lucky.

CALCIUM, VITAMIN D AND COLORECTAL CANCER

The findings of some studies have suggested that calcium and vitamin D are protective against colorectal cancer, however the evidence is inconsistent. Jennifer Lin and her colleagues examined the data from a large female cohort, The Women's Health Study, to assess calcium and vitamin D intake and the risk of colorectal cancer (Lin J et al. Intake of calcium and vitamin D and risk of colorectal cancer in women. Am J Epidemiol 2005; 161: 755-764). Almost 40,000 women were enrolled in the latter study in 1993. About 3,000 women were not included in this analysis because they failed to provide sufficient dietary information at baseline.

A food frequency questionnaire was administered at baseline. The participants were asked every year after enrollment if they had been diagnosed with colorectal cancer. Cox proportional hazard regression was used to estimate relative risks and 95% confidence intervals. Multivariate models were adjusted for body mass index, physical activity, family history of colorectal cancer, multivitamin use, smoking status, alcohol consumption, saturated fat and meat intake, and total energy intake.

The multivariate risks comparing the highest with the lowest quintile were 1.20, 95% CI: 0.79, 1.85, for total calcium and 1.34, 95% CI: 0.84, 2.13, for total vitamin D. The findings provided little support for an association of calcium or vitamin D with colorectal cancer risk.

COMMENT: It is refreshing to find negative results published. If this apparent trend continues, future meta-analyses will not be as biased as they are today.

SOME OBSERVATIONS BY D. A. HENDERSON

Jonathan M. Samet recently interviewed D. A. Henderson, and the interview was published in Epidemiology (Samet J M. A conversation with D. A. Henderson. Epidemiology 2005: 266-269). I found the last item of the interview especially interesting, and I will quote the segment in its entirety:

JS: Let me ask a question that you are probably as well positioned as anybody to answer. Have the academic world of epidemiology and the applied world gone in quite separate direction?

DH: Yes, to a regrettable degree. It is not only epidemiology, but schools of public health. My concern, when I accepted the deanship at Hopkins , was that the schools of public health were becoming primarily graduate schools, not professional schools. Could a faculty of surgery do research on surgery and teach surgery, but perform no operations? They would be considered irrelevant. In schools of public health, we have all too many faculty who have never worked in operating a public health program or even participating in one. Such individuals seem to be overrepresented on committees on appointments and promotions, and judge faculty only on the number of peer-reviewed papers, debasing those with real-world experience. If schools of public health and departments of epidemiology are to prosper, they have to be relevant to the real world.