Statistical techniques applied to environmental health sciences

National Institute Of Environmental Health Sciences

Investigators

Linked publications & trials

Abstract

We have recently guided the data analysis for several projects, primarily involving data from the Agricultural Health Study (AHS), a cohort of licensed pesticide applicators (mostly farmers) from Iowa and North Carolina. We evaluated associations between specific pesticides and incident hyperthyroidism in private pesticide applicators in the AHS. We used Cox proportional hazards models to estimate hazard ratios for associations between pesticide use at enrollment (19931997) and hyperthyroidism (n=271) among 35,150 applicators (mostly men) adjusting for potential confounders. Ever use of several pesticides (organophosphate insecticide malathion, fungicide maneb/mancozeb, herbicides dicamba, metolachlor, and atrazine) was associated with reduced risk of hyperthyroidism; hazard ratios ranged from 0.50 (95% CI 0.30 to 0.83) for maneb/mancozeb to 0.77 (95% CI 0.59 to 1.00) for atrazine. These results require confirmation in additional studies. Olfactory impairment is common among older adults and predictive of risk for certain neurodegenerative diseases. We examined high pesticide exposure events (HPEEs) reported at enrollment in relation to self-reported olfactory impairment at the latest AHS follow-up (20132015) using multivariable logistic regression considering farmers without HPEEs as unexposed. About 10.6% of farmers reported olfactory impairment. A history of HPEEs was associated with a higher likelihood of reporting olfactory impairment two decades later odds ratio (OR)=1:49 95% confidence interval (CI):1.28,1.73. The association appeared stronger the longer the delay between an HPEE and washing with soap and water e.g., OR=2:07 (95% CI: 1.48, 2.89) for 46 hrs. OR=1:39 (95%CI: 1.11, 1.75) for <30min. Our study was limited by use of self-reported olfactory impairment; future studies should assess olfactory impairment objectively. In earlier work, we reported lower mortality rates through 2007 among AHS participants compared to the general population. We extended the mortality analysis among AHS participants in three subgroups(51,502 private pesticide applicators, their 31,867 spouses, and 4,677 commercial pesticide applicators in Iowa and North Carolina) through 2015. In addition, we compared results using several analytic approaches (standardized mortality ratios (SMRs), causal mortality ratios (CMRs), and relative SMRs) using state-specific mortality rates of the general populations as the referent. Over the average 16 years of follow-up, 9,305 private applicators, 3,384 spouses, and 415 commercial applicators died. SMRs and CMRs indicated lower overall mortality in all three study subgroups (SMRs ranging 0.610.69 and CMRs 0.740.89. SMR analyses indicated fewer-than-expected deaths from many causes, but greater-then-expected deaths from transportation-related injuries and mechanical forces were elevated in private applicators. CMRs indicated higher-than-expected deaths from prostate cancer, lymphohematopoietic cancers, Parkinsons and Alzheimers disease, and chronic glomerulonephritis in private applicators, and non-Hodgkin lymphoma in spouses (ranging 1.191.53). Relative SMR and CMR provided comparable findings.