How useful is linear regression analysis in detecting the existence of dose-response relationships in large-scale epidemiologic studies when only a fraction of the population is sensitive? The case of methylmercury

The existence of a dose response in epidemiologic studies is generally determined from the linear regression slope after controlling for covariates. This approach assumes the entire population is equally sensitive to the toxicant and that response is a function only of dose and a random error function. However, sensitive subpopulations have been identified for a variety of toxicants possibly including methylmercury (MeHg). The study of MeHg exposure in the Seychelles Islands has failed to find significant effects (dose-response slope not significantly different from zero) while other studies have found such effects. Using data on the error function in developmental test scores and MeHg exposure distributions from that study, and assuming plausible dose-response relationships for sensitive subpopulations, we conducted Monte-Carlo simulations of the power of linear regression analysis to detect a dose-response relationship from the total sample (n=700), and to compare dose-response slopes in the total and sensitive populations. Linear regression did not reliably detect a dose-response relationship for most scenarios when sensitives were 5% of the total and for some scenarios when sensitives were 10% of the total. We also found that the dose-response slope for the total population underestimated the sensitive dose-response slope in all cases by about an order of magnitude. These findings may have important implications for detection and quantification of dose-response relationships from epidemiologic studies.

Cardiovascular stress responses and their relation to symptoms in Gulf War veterans with fatiguing illness

OBJECTIVE

The objective of this study was to examine whether inappropriate cardiovascular responses to stressors may underlie symptoms in Gulf War veterans with chronic fatigue.

METHODS

Psychophysiological stress testing was performed on 51 Gulf War veterans with chronic fatigue (using the 1994 case definition of the Centers for Disease Control and Prevention) and 42 healthy veterans. Hemodynamic responses to cold pressor, speech, and arithmetic stressors were evaluated using impedance cardiography.

RESULTS

Veterans with chronic fatigue had diminished blood pressure responses during cognitive (speech and arithmetic) stress tests due to unusually small increases in total peripheral resistance. The cold pressor test, however, evoked similar blood pressure responses in the chronic fatigue and control groups. Low reactivity to cognitive stressors was associated with greater fatigue ratings among ill veterans, whereas an opposite relation was observed among healthy veterans. Self-reported neurocognitive decline was associated with low reactivity to the arithmetic task.

CONCLUSIONS

These results suggest a physiological basis for some Gulf War veterans' reports of severe chronic fatigue. A greater deficit with responses processed through cerebral centers, as compared with a sensory stimulus (cold pressor), suggests a defect in cortical control of cardiovascular function. More research is needed to determine the specific mechanisms through which the dissociation between behavioral and cardiovascular activities identified in this study may be contributing to symptoms in Gulf War veterans.

Physical activity before and after exercise in women with chronic fatigue syndrome

We measured physical activity after strenuous exercise in 20 women with chronic fatigue syndrome (CFS), compared to 20 sedentary healthy volunteers who exercised no more than once per week. Activity was measured for 2 weeks using a portable waist-worn vertical accelerometer. After the first week of activity monitoring, all participants returned for a maximal treadmill test, followed by continued activity monitoring for the second week. Five activity measures were derived from the data: (i) average activity; (ii) total activity; (iii) duration of waking day; (iv) duration; and (v) number of daily rests. A repeated measures ANCOVA was used to determine post-treadmill group differences accounting for pre-treadmill differences. There was a significant reduction in overall average activity after the treadmill test, with the greatest decrease on days 12 through 14. This reduction was accompanied by a significant increase in the duration of the waking day and number of daily rests. Thus, marked exertion does produce changes in activity, but later than self-report would suggest, and are apparently not so severe that CFS patients cannot compensate.

The Effects of Fatigue on Neuropsychological Performance in Patients With Chronic Fatigue Syndrome, Multiple Sclerosis, and Depression

The effects of fatigue on neuropsychological performance were examined in patients with fatiguing illnesses. Repeated testing with the Paced Auditory Serial Addition Test (PASAT; Gronwall, 1977) was employed over the course of a demanding neuropsychological testing session. It was hypothesized that if fatigue affects performance, one would expect to observe "blunting" of the PASAT practice effect. Fifteen of the study participants live with chronic fatigue syndrome (CFS), 15 with multiple sclerosis (MS), 14 with depression (DEP), and 15 are healthy, sedentary controls. Overall PASAT performance was significantly reduced for CFS and DEP participants compared to controls, whereas mean performance did not differ across the three fatiguing illness groups. Degree of improvement across trials (i.e., practice effect) for the groups did not differ from controls'. Neither subjective fatigue or DEP were significantly related to PASAT performance. These findings suggest that fatigue does not universally impair performance during neuropsychological assessment even in groups in which fatigue is a prominent symptom.

Estimation of fish consumption and methylmercury intake in the New Jersey population

Despite scientific attention to the toxicology of methylmercury (MeHg), little is known about population-based exposure to this compound. In this study, fish consumption and MeHg intake were estimated based on a seven-day recall survey of fish consumption among 1000 randomly selected New Jersey residents. Survey data were reported on a per-meal basis, and the fish species/dishes consumed at each meal were identified. Portion sizes for each meal were reported or estimated. To correct a possible bias due to underrepresentation of infrequent consumers, several schemes for weighting fish consumption data were investigated. MeHg concentration for most fish species was estimated from the National Marine Fisheries Service database or from recent United States Food and Drug Administration data. Commercial fish accounted for about 95% of all consumption. Mean fish consumption is estimated for all New Jersey adult consumers at 50.2 g/day (90th percentile = 107.4 g/day) and for women 18-40 years old (childbearing age) at 41.0 g/day (90th percentile = 88.1 g/day). Mean MeHg intake is estimated for all New Jersey adult consumers at 7.5 micrograms/day (90th percentile = 1.79 micrograms/day) and for women 18-40 at 6.3 micrograms/day (90th percentile = 14.8 micrograms/day). When MeHg concentrations are adjusted to account for the possible overestimation of current concentrations by the 20-year-old National Marine Fisheries Service database, it is estimated that MeHg intakes may be about 70-80% of unadjusted estimates. Based on these analyses, it is estimated that 21-30% of New Jersey women 18-40 and 5-8% of all New Jersey adults exceed their respective U.S. Environmental Protection Agency Reference Doses for MeHg. Because of uncertainty associated with the Reference Doses, exceeding the U.S. Environmental Protection Agency guidelines does not necessarily correspond to adverse effects on consumers or their fetuses.

Prevalence of and potential risk factors for symptoms associated with insecticide use among animal groomers

Pet groomers make numerous insecticide applications during the flea season, but few studies have examined their health complaints. The Pesticide Control Program of the New Jersey Department of Environmental Protection conducted a health and safety survey of this population. All licensed pet applicators in New Jersey were contacted, as were New Jersey veterinarians listed as pet-animal practitioners by the American Veterinary Medical Association. Approximately 36% of the respondents indicated that during the 1994 flea season, they had experience at least one of the 17 symptoms associated with insecticide application. Central nervous system symptoms (headache, dizziness, or confusion) and skin symptoms (skin rash or numbness/tingling) were reported most frequently. Logistic regression results suggest that applications per season, years as an applicator, certain hygiene variables, certain classes of products, and status of applicator (veterinary vs veterinary) are potentially important risk factors.