David M. Umbach

Rotenone, Paraquat, and Parkinson’s Disease

Background Mitochondrial dysfunction and oxidative stress are pathophysiologic mechanisms implicated in experimental models and genetic forms of Parkinson’s disease (PD). Certain pesticides may affect these mechanisms, but no pesticide has been definitively associated with PD in humans. Objectives Our goal was to determine whether pesticides that cause mitochondrial dysfunction or oxidative stress are associated with PD or clinical features of parkinsonism in humans. Methods We assessed lifetime use of pesticides selected by mechanism in a case–control study nested in the Agricultural Health Study (AHS). PD was diagnosed by movement disorders specialists. Controls were a stratified random sample of all AHS participants frequency-matched to cases by age, sex, and state at approximately three controls: one case. Results In 110 PD cases and 358 controls, PD was associated with use of a group of pesticides that inhibit mitochondrial complex I [odds ratio (OR) = 1.7; 95% confidence interval (CI), 1.0–2.8] including rotenone (OR = 2.5; 95% CI, 1.3–4.7) and with use of a group of pesticides that cause oxidative stress (OR = 2.0; 95% CI, 1.2–3.6), including paraquat (OR = 2.5; 95% CI, 1.4–4.7). Conclusions PD was positively associated with two groups of pesticides defined by mechanisms implicated experimentally—those that impair mitochondrial function and those that increase oxidative stress—supporting a role for these mechanisms in PD pathophysiology.

Designing and analysing case-control studies to exploit independence of genotype and exposure

Genetic susceptibility and environmental exposures play a synergistic role in the aetiology of many diseases. We consider a case-control study of a rare disease in relation to a categorical exposure and a genetic factor under the assumption that the genotype and the exposure occur independently in the population under study. Using a logistic model for risk, we describe maximum likelihood methods based on log-linear models that explicitly impose the independence assumption, something the usual logistic regression analyses cannot do. The estimator of the genotype-exposure interaction effect depends only on data from cases. Estimators for genotype and for exposure effects depend also no data from controls, but only through their respective marginal totals. All three estimators have smaller variance than they would were independence not enforced. These results have important implications for design: (i) Case-only studies can efficiently estimate gene-by-environment interactions. (ii) Studies where controls are genotyped anonymously can estimate genotype, exposure, and interaction effects as efficiently as designs where genotype and exposure data are linked. This feature addresses a growing concern of human subjects review boards. (iii) Exposure and interaction effects, but not genotype effects, can be estimated from studies where genetic information is only collected from cases (although one can recover the genotype effect if external gene prevalence data exist). Such designs have the compensatory benefit that the response rate (hence, validity) is higher when controls are not subjected to intrusive tissue sampling. However, the independence assumption can be checked only with linked genotype and exposure data for some controls. We illustrate the methods by applying them to recent study of cleft palate in relation to maternal cigarette smoking and to a variant of the transforming growth factor alpha gene in the child.

Prevalence of Medication Treatment for Attention Deficit–Hyperactivity Disorder Among Elementary School Children in Johnston County, North Carolina

OBJECTIVES This study estimated the prevalence of medication treatment for attention deficit-hyperactivity disorder (ADHD) among elementary school children in a North Carolina county. METHODS Parents of 7333 children in grades 1 through 5 in 17 public elementary schools were asked whether their child had ever been given a diagnosis of ADHD by a psychologist or physician and whether their child was currently taking medication to treat ADHD. Parents of 6099 children (83%) responded. RESULTS By parental report, 607 children (10%) had been given an ADHD diagnosis and 434 (7%) were receiving ADHD medication treatment. Seventy-one % of the diagnosed children were receiving medication. Treatment rates varied by sex, race/ethnicity, and grade. CONCLUSIONS If treatment patterns observed in this study are representative, the public health impact of ADHD may be underestimated.

Lead exposure and amyotrophic lateral sclerosis.

Background. Previous interview-based studies have suggested that exposure to neurotoxicants including metals might be related to ALS. Methods. We evaluated the relation of lead exposure to ALS, using both biological measures and interviews, in a case-control study conducted in New England from 1993 to 1996. Cases (N = 109) were recruited at two hospitals in Boston, MA. Population controls (N = 256) identified by random-digit dialing were frequency-matched to cases by age, sex, and region of residence within New England. Results. Risk of ALS was associated with self-reported occupational exposure to lead (odds ratio [OR] = 1.9; 95% confidence interval [CI] = 1.1–3.3), with a dose response for lifetime days of lead exposure. Blood and bone lead levels were measured in most cases (N = 107) and in a subset of controls (N = 41). Risk of ALS was associated with elevations in both blood and bone lead levels. ORs were 1.9 (95% CI = 1.4–2.6) for each &mgr;g/dl increase in blood lead, 3.6 (95% CI = 0.6–20.6) for each unit increase in log-transformed patella lead, and 2.3 (95% CI = 0.4–14.5) for each unit increase in log-transformed tibia lead. Conclusions. These results are consistent with previous reports and suggest a potential role for lead exposure in the etiology of ALS.

Smoking duration, intensity, and risk of Parkinson disease

Objective: To evaluate the relative importance of smoking duration vs intensity in reducing the risk of Parkinson disease (PD). Methods: The study included 305,468 participants of the NIH-AARP Diet and Health cohort, of whom 1,662 had a PD diagnosis after 1995. We estimated odds ratios (OR) and 95% confidence intervals from multivariate logistic regression models. Results: Compared with never smokers, the multivariate ORs were 0.78 for past smokers and 0.56 for current smokers. Among past smokers, a monotonic trend toward lower PD risk was observed for all indicators of more smoking. Stratified analyses indicated that smoking duration was associated with lower PD risk within fixed intensities of smoking. For example, compared with never smokers, the ORs among past smokers who smoked >20 cigarettes/day were 0.96 for 1–9 years of smoking, 0.78 for 10–19 years, 0.64 for 20–29 years, and 0.59 for 30 years or more (p for trend = 0.001). In contrast, at fixed duration, the typical number of cigarettes smoked per day in general was not related to PD risk. Close examination of smoking behaviors in early life showed that patients with PD were less likely to be smokers at each age period, but if they smoked, they smoked similar numbers of cigarettes per day as individuals without PD. Conclusions: This large study suggests that long-term smoking is more important than smoking intensity in the smoking–Parkinson disease relationship.

The Use of Case-Parent Triads to Study Joint Effects of Genotype and Exposure

Most noninfectious disease is caused by low-penetrance alleles interacting with other genes and environmental factors. Consider the simple setting where a diallelic autosomal candidate gene and a binary exposure together affect disease susceptibility. Suppose that one has genotyped affected probands and their parents and has determined each probands exposure status. One proposed method for assessment of etiologic interaction of genotype and exposure, an extension of the transmission/disequilibrium test, tests for differences in transmission of the variant allele from heterozygous parents to exposed versus unexposed probands. We show that this test is not generally valid. An alternative approach compares the conditional genotype distribution of unexposed cases, given parental genotypes, versus that of exposed cases. This approach provides maximum-likelihood estimators for genetic relative-risk parameters and genotype-exposure-interaction parameters, as well as a likelihood-ratio test (LRT) of the no-interaction null hypothesis. We show how to apply this approach, using log-linear models. When a genotype-exposure association arises solely through incomplete mixing of subpopulations that differ in both exposure prevalence and allele frequency, the LRT remains valid. The LRT becomes invalid, however, if offspring genotypes do not follow Mendelian proportions in each parental mating type-for example, because of genotypic differences in survival-or if a genotype-exposure association reflects an influence of genotype on propensity for exposure-for example, through behavioral mechanisms. Because the needed assumptions likely hold in many situations, the likelihood-based approach should be broadly applicable for diseases in which probands commonly have living parents.

Isoflavones in urine, saliva, and blood of infants: data from a pilot study on the estrogenic activity of soy formula

In the United States, about 25% of infant formula sold is based on soy protein, which is an important source of estrogenic isoflavones in the human food supply. Nevertheless, few studies report isoflavone levels in infants. We did a partly cross-sectional and partly longitudinal pilot study to examine childrens exposure to isoflavones from different feeding methods. A total of 166 full-term infants between birth and 1 year of age were recruited into soy formula, cow milk formula, or breast milk regimens according to their feeding histories. A total of 381 urine, 361 saliva, and 88 blood samples were collected at 382 visits. We used automated online solid-phase extraction coupled to high-performance liquid chromatography–tandem mass spectrometry (HPLC–MS/MS) for measuring three isoflavones (daidzein, genistein, and equol) in urine, and used similar LC/MS/MS techniques for saliva and blood spots. Concentrations of daidzein and genistein were undetectable in most blood or saliva samples from children fed breast milk or cow milk formula. The proportion of non-detectable values was somewhat lower in urine than in the other matrices. Concentrations of equol were detectable only in a few urine samples. For both daidzein and genistein, urine contained the highest median concentrations, followed by blood and then saliva. Urinary concentrations of genistein and daidzein were about 500 times higher in the soy formula-fed infants than in the cow milk formula-fed infants. The correlations between matrices for either analyte were strikingly lower than the correlation between the two analytes in any single matrix. We did not find significant correlations between isoflavone concentrations and the levels of certain hormones in children fed soy formula. Our results, based on much larger numbers of infants, strongly confirm previous reports, but whether phytoestrogens in soy formula are biologically active in infants is still an open question. We plan further longitudinal studies focusing on physical and developmental findings reflecting the effects of estrogen exposure.

Studying the Epidemiology of Attention-Deficit Hyperactivity Disorder: Screening Method and Pilot Results

Objective: As part of a larger epidemiologic study of risk factors for attention-deficit hyperactivity disorder (ADHD), this pilot study combined parent and teacher information to estimate ADHD prevalence among elementary school children in a North Carolina county. The methods developed for this study and the pitfalls we encountered illustrate the challenges involved in conducting population-based studies of ADHD. Methods: We employed 2-stage screening using DSM-IV criteria. Teachers completed behaviour-rating scales for all children. We then administered a structured telephone interview to parents of potential cases. We screened 362 of 424 (85%) children in grades 1 to 5 in 4 schools. Results: According to parent reports, 43 children (12%) had previously been diagnosed with ADHD by a health professional. Thirty-four children (9%) were taking ADHD medication. Forty-six children (12.7%) met study case criteria for ADHD, based on combined teacher and parent reports. Of the 46 cases, 18 (39%) had not been previously identified. Eight previously diagnosed children, however, did not meet case criteria. After we adjusted for nonresponse, the estimated prevalence was 16% (95%CI, 12% to 20%). Conclusions: These data suggest that the DSM-IV prevalence of ADHD has been substantially underestimated, although the true prevalence in this population may be less than the 16% estimated here. Population-based studies of ADHD are feasible and may provide important information about practice and treatment patterns in community settings, as well as a broader understanding of the etiology and life course of this common disorder.

Association of Cigarette Smoking with Amyotrophic Lateral Sclerosis

We explored the relationship between amyotrophic lateral sclerosis (ALS) and cigarette smoking in a case-control study conducted in New England from 1993 to 1996. Recently diagnosed ALS cases (n = 109) were recruited from two major referral centers. Population controls (n = 256) were identified by random telephone screening. Data were analyzed by logistic regression. After adjusting for age, sex, region and education, ever having smoked cigarettes was associated with an increase in risk for ALS (odds ratio 1.7; 95% confidence interval 1.0–2.8). Average cigarettes smoked per day, years smoked and pack-years were all greater in cases than controls, but dose-response trends were not observed. Similar numbers of cases and controls had ever used alcohol, and only a small, nonsignificant association of drinks per month with ALS was observed. The association of cigarette smoking with ALS was not affected by adjusting for alcohol use. In contrast, the weak relationship of ALS with alcohol use was apparently due to confounding by smoking.

Application of the GA/KNN method to SELDI proteomics data

SUMMARY Proteomics technology has shown promise in identifying biomarkers for disease, toxicant exposure and stress. We show by example that the genetic algorithm/k-nearest neighbors method, developed for mining high-dimensional microarray gene expression data, is also capable of mining surface enhanced laser desorption/ionization-time-of-flight proteomics data. AVAILABILITY The source code of the program and documentation on how to use it are freely available to non-commercial users at http://dir.niehs.nih.gov/dirbb/lifiles/softlic.htm