Researching a Broad Range of Health Outcomes
Health outcomes of interest to environmental epidemiologists range from well-characterized diseases, such as cancer, to syndromes or constellations of clinical measures of physiologic or neurobehavioral function. Changes in these health outcomes can indicate that environmental factors are involved. However, not all relations between health end points and environmental exposures have been recognized. In order to identify changes in chronic diseases, the expected or baseline rates of diseases need to be determined, as has already been done for cancer. This is difficult because case definitions for many chronic diseases and syndromes are not uniform or well defined, registries and codings are not consistent, many demographic variables may influence the outcome, and time trends may not be reliable.
Many environmental toxins, acting either independently or in combination with other agents, may affect several organ systems. We consider problems in defining and studying a variety of health end points, especially those not associated with cancer. We identify gaps in current knowledge, suggest several noncancer end points that require additional research, and highlight some of those research opportunities. Finally, we discuss biologic markers for these end points and considerations of special populations at risk.
A comprehensive assessment of the evolving literature on health outcomes that may be associated with environmental exposures is beyond the scope of this study. Readers are referred to reports from the National Research Council on some of these health outcomes (NRC, 1989a,b, 1991, 1992a,b,c, 1993).
Most environmental epidemiology has been concerned with relatively few health outcomes, such as the neurologic and mental outcomes associated with lead and methyl mercury, respiratory disease associated with air pollution, and malignant neoplasms associated with exposures to various chemical agents in the environment. However, there is increasing evidence that a much broader array of health outcomes—such as neurologic, respiratory, and reproductive end points—may also be associated with environmental exposures. Examples of the environmental associations that have been investigated are a link between congenital malformations and trichloroethylene in drinking water, increased hospitalization of children and adults for asthma associated with air pollution, and the neurologic consequences of exposures to lead and solvents in the environment (NRC, 1991). The factors causing many noncancer outcomes, the underlying mechanisms of causation, and relative contributions of the various causal agents have not been clearly delineated. Further, an individual or community may experience more than one health outcome. Persons may be exposed to multiple chemicals (sometimes from the same source), each producing different outcome events, or a single chemical can be responsible for different outcomes in the same or different individuals.
Research into many outcomes potentially related to environmental exposure is at an early stage of development. In part, this is because there has been little effort to measure the incidence and prevalence of many chronic diseases. It is difficult to identify, monitor, and study populations at risk. Finally, more-refined measures of disease that can be applied to groups of persons are needed.
This chapter provides a brief overview of health outcomes that have been associated with environmental exposure. We also review outcomes for which evidence from toxicologic or occupational studies suggests associations with environmental exposure, but where current environmental-epidemiologic data are inadequate to provide definitive information about risk.
A diverse and growing literature characterizes the effects of air pollution on human health. Progress over past decades has elucidated the link of an array of respiratory outcomes to air pollution. Adverse effects on the respiratory tract of air pollutants such as ozone or airborne particles are highly nonspecific and not easily detected clinically. Nevertheless, for many common diseases, even small relative risks (RRs) may translate into significant numbers of cases of disease because of the large size of the exposed population.
For example, studies of the effects on animals of long-term exposure to ozone at concentrations similar to those seen in some US cities showed chronic lung damage that increased monotonically with cumulative dose; there was no evidence of a threshold (American Thoracic Society, 1996a). If similar effects also occur in humans, a substantial proportion of the population in the United States is exposed every summer to ozone concentrations that may produce chronic lung damage. Analyses of data in 8 cities have detected associations of airborne particles with small increases in the risk of mortality (Schwartz, 1991). In a community study, the closing of a steel mill for a year was associated with more than a 40% decrease in hospital admissions for asthma in children; the next year the mill reopened, and hospitalization rates rose to their previous level (Pope 1989, 1991). Another recent study found that rates of acute bronchitis in children were about twice as high in a town with particle concentrations at the US air-quality standard as they were in a community with near background concentrations; intermediate communities had intermediate rates (Dockery et al., 1989). Braun-Fahrlander et al. (1992) found major changes in respiratory symptoms among schoolchildren at concentrations commonly seen in the US urban population.
Acute Health Effects
Acute health effects of air pollution have long been established, although primarily by studies at concentrations of pollutants far higher than those now typical in developed countries. Recently, however, exposure to air pollution has been associated with acute outcomes—such as reversible reductions in lung function, increased respiratory symptoms and illness, emergency-room visits and increased hospitalization, and deaths from respiratory and cardiovascular diseases—at exposure levels far lower than those at which earlier data were collected. These are discussed briefly below in order of increasing severity. Several recent reports provide more-complete treatments of the evidence on acute health effects (Dockery and Pope, 1994; American Thoracic Society, 1996b).
Exercising volunteers exposed to ozone at concentrations below current air-quality standards show reversible decrements in lung function (McDonnell et al., 1985). Studies of ozone-exposed children in summer camps (Spektor et al., 1988; Lioy et al., 1985) have found that they have similar responses at even lower levels, at least in the summer months, when they are outdoors during much of the day. Smaller but still significant effects of ozone on lung function have been reported in schoolchildren during the school year, when outdoor activity is less (Kinney et al., 1989). Studies of lavage fluid from the lungs of volunteers have also provided evidence of inflammatory processes in the lung following ozone
exposure (American Thoracic Society, 1996a). Short-duration exposures of asthmatics to SO2 also reduce lung function; in the protocols, delivering the gas during exercise increases the amount of SO2 received (American Thoracic Society, 1996b).
Short periods of moderately elevated particle concentrations have been associated with pulmonary-function deficits (Dassen et al., 1986; Brunekreef et al., 1991). Subsequent studies that have examined daily time series, rather than episodes, have also associated ambient PM10 concentrations with short-term variation in peak expiratory flow rate at concentrations below current national ambient-air quality standards (Pope and Dockery, 1992; Pope et al., 1991).
Daily symptom incidence and duration of respiratory illness have been linked to exposure to airborne particles. Supporting evidence includes results from the Six Cities Study conducted by Harvard investigators (Schwartz et al., 1989), in which particle concentrations never exceeded 75% of the current air-quality standard. In this study, daily diary responses concerning respiratory symptoms were significantly associated with particle concentrations. Similar findings come from Switzerland (Braun-Fahrlander et al., 1992) and Provo, Utah (Pope and Dockery, 1992; Pope et al., 1991). Acid-aerosol exposure has been associated with increased symptoms in a diary study of asthmatics (Ostro et al., 1991). For several of the respiratory symptoms, the odds ratios from Utah, the Six Cities Study, and Switzerland are similar.
Exposure to airborne particles has been associated with increased rates of bronchitis in children (Dockery et al., 1989) at concentrations below current standards and also with increased rates of croup attacks in children (Schwartz et al., 1991). These effects do not appear to be limited to children. For example, Ostro and Rothschild (1989) have reported an association between airborne particles and ozone and respiratory illness severe enough to restrict activity in adults. A meta-analysis has reported a significant association between NO2 exposure and respiratory illness in children (Hasselblad et al., 1992).
Bates and Sizto (1987) reported that exposures to both ozone and sulfate were associated with increased incidences of hospitalization for respiratory illness in Ontario. Pope (1989), as mentioned above, found sharp changes in hospitalization rates for children when a steel mill closed and then reopened. Hospital admissions were also increased in the German smog episode of 1985 (Wichmann et al., 1989) at a time of sharp increases in both total suspended particles and SO2. Sunyer et al. (1991) reported that airborne particles and SO2 were associated with hospitalization for respiratory illness in Barcelona, Spain. Hospital emergency-room visits were associated with sulfates and SO2 in a study in Vancouver, Canada (Bates et al., 1990), and respirable particles were associated with hospital
emergency-room visits in Israel (Gross et al., 1984). Schwartz et al. (1993) reported that inhalable-particle concentrations that never exceeded the current air-quality standards were associated with increased emergency-room visits for asthma.
Studies have associated airborne particles with increased mortality (Dockery and Pope, 1994; American Thoracic Society., 1996b). Those studies have shown similar patterns of disease in areas with different mean temperatures and climatic conditions and in areas with air-pollution peaks in both winter and summer. Other studies have used somewhat different methods, so their estimates of effect size are not directly comparable. However, Fairley (1990) and Schwartz and Marcus (1990) both reported that optical measures of airborne particles were associated with daily mortality. Qualitatively, the effect-size estimates as assessed by the regression coefficients seemed similar. In contrast, Hatzakis et al. (1986) reported that mortality in Athens, Greece, was primarily associated with SO2 and not with particulate matter, though data for the winter season alone (Katsounnayi et al., 1990) showed a principal association with particulate matter. Kinney and Ozkaynak (1991) have reported associations of both ozone and airport visibility (as a proxy for particles) with daily mortality in Los Angeles, Calif. The ozone association was stronger in that study.
Chronic Effects of Air Pollution
Spektor et al. (1991) have reported that long-term exposure of children to particulate air pollution was associated with impaired lung function. Data from the NHANES II survey (Schwartz, 1989) showed that chronic exposure to particles or ozone was associated with lower lung function in children; the ozone effect was stronger. Chestnut et al. (1991) have reported that long-term TSP exposure was associated with lung-function decrements in adults.
Some studies have associated differences in long-term exposure to air pollution with increased rates of chronic respiratory illness. For example, Euler et al. (1987, 1988) have reported that an index of cumulative exposure to TSP was associated with increased rates of chronic bronchitis in a cohort of Seventh-Day Adventists in California. The association remained when ozone exposure was also examined. A weaker association was found with ozone exposure. Detels et al. (1987) have reported differences in respiratory health among communities exposed to different levels of air pollution in the Los Angeles area.
Several studies (e.g., Lave and Seskin, 1977; Chappie and Lave, 1982; Lipfert, 1980; Ozkaynak et al., 1986) have sought to associate long-term differences in air-pollution concentrations with differences in age-adjusted
mortality rates across major urban areas. These studies have applied regression methods to data from multiple locations. Such studies are constrained by the difficulty of adequately controlling for potential confounding effects of other risk factors, such as smoking; this problem is avoided by time-series analyses within a single urban area. In general, research methods have improved over time, though much uncertainty remains. The magnitude of the excess mortality suggested by these studies is somewhat larger than that suggested by acute studies of daily mortality.
Studies have also been directed at specific diseases, for example, lung cancer. Archer (1990) examined deaths from lung cancer and other respiratory illnesses in 2 heavily populated valleys in Utah. Mortality rates were essentially identical in the 1940s and early 1950s. However, a steel mill was opened in one of these areas in the late 1950s. By the 1960s, a trend toward higher lung cancer and respiratory mortality was evident in this valley, and the trend was quite pronounced in the 1970s and 1980s.
Neurologic and neurobehavioral changes can result from effects of agents on autonomic, peripheral, and central components of the nervous system. Thus, neurotoxins can have a wide range of effects, including changes in motor and sensory function or behavior, central nervous system damage, and cognitive, memory, and developmental changes. The effects on the autonomic system are primarily biochemical. Chronic exposures may affect the peripheral nervous system; even minor damage to myelin can affect nerve conduction velocities. Since the nervous system has only a modest capacity to regenerate, subtle damage can have serious and long-lasting effects (Tilson and Mitchell, 1992). Neurotoxic effects may be acute, chronic, or delayed. Time relationships may be more complicated for some neurotoxic exposures. For example, solvent exposure just before testing may interfere with performance tests for chronic neurotoxic effects (Melius and Schulte, 1981). Thus, the study design should use the proper exposure time range and the proper response time range for the situation being evaluated, e.g., chronic effects in a chronically exposed population. The full range of neurologic and behavioral effects of a toxic substance is rarely known (Xintaras et al., 1979).
During recent decades, increasing attention has been paid to subtle behavioral changes that may occur at doses of agents lower than those causing physical signs and symptoms (Gochfeld et al., 1991; Tilson and Mitchell, 1992). These changes may occur after inapparent and chronic exposures, and they may lead to functional impairment that would otherwise be subclinical. Such impairments, unnoticeable in an individual, can have a substantial impact on the population. For example, Needleman et
al. (1982) has estimated that a downward shift in IQ of 4 points would double the proportion of children with IQ less than 80. Environmental exposures have been suggested for such diseases as parkinsonism, amyotrophic lateral sclerosis, Alzheimer's disease, and several peripheral neuropathies (Tanner et al., 1987; Ngim and Devathasan, 1989; Kalfakis et al., 1991; Bos et al., 1991; McLachlan et al., 1991). Improved means to evaluate more-subtle neurobehavioral and neurophysiologic effects (Valciukas and Lilis, 1980; Xintaras et al., 1979) have led to major advances in assessing neurotoxic effects. These methods are often specific for certain types of neurobehavioral effects, e.g., visual versus auditory memory. Hence, a careful selection of tests is required for field surveys. Consideration must also be given to possible confounders, such as age, alcohol intake, and education. Sometimes a battery of tests may be needed to screen workers for the effects of a neurotoxic substance. Such a battery should be specific enough to measure functions related to known effects of the substance and heterogeneous enough to cover a variety of neurobehavioral functions.
Neurologic and behavioral effects have been assessed for exposures to some indoor air pollutants. Otto et al. (1992) exposed 66 healthy young male subjects with no history of chemical sensitivity to air, to clean air, and to a complex mixture of volatile organic compounds (VOCs). Participants reported more fatigue and mental confusion after exposure to the organic compounds. However, performance on 13 neurobehavioral tests was not affected. In another part of the study, eye and throat irritation, headache, and drowsiness increased or showed no evidence of adaptation during exposure, even though the intensity of odors decreased by 30% (Hudnell et al., 1992). The investigators concluded that these results indicate that irritation intensity and other symptoms are not related in a simple way to odor intensity. The findings suggest that the symptoms may not be a psychosomatic response to the detection of an unpleasant odor and that environmental odor pollution may affect neurobehavior. Instead, subthreshold levels of VOCs may interact additively and stimulate trigeminal nerve receptors. Many nonspecific-symptom clusters have an odor component. Noxious environmental odors might trigger symptoms by a variety of physiologic mechanisms, including exacerbation of underlying medical conditions, innate odor aversions or aversive conditioning, stress-induced illness, and possible phenomenal reactions. Whereas relatively consistent patterns of subjective symptoms have been reported among individuals who live near environmental odor sources, documentation of objective correlates to such symptoms would require the development of new research tools.
An example of a common environmental neurotoxicant that produces a variety of chronic effects is lead. Lead has been known to cause serious
cognitive damage to children since the pioneering work of Byers and Lord (1943). Historically, assessment of the neuropsychologic effects of lead has been hampered by inadequate markers of exposure; i.e., blood lead is a short-term marker, and levels may return to normal after exposure has ended (Needleman, 1986), even though subjects' past exposures have caused persistent physiologic effects. One approach to the exposure-assessment issue is use of lead in shed deciduous teeth as a marker of past exposures. In a cohort study, first- and second-grade students who were considered asymptomatic for lead were classified by dentin lead levels and then evaluated with a battery of neuropsychologic tests (Needleman et al., 1979). Children with high dentin lead scored significantly less well on the Wechsler Intelligence Scale for Children (revised), on 3 measures of auditory speech processing, and on a measure of attention (Needleman et al., 1982). The authors concluded that lead, at doses below those that produce clinical symptoms, is associated with impaired neurobehavioral functioning (Needleman, 1986).
Although recent studies have found negative associations between blood lead concentrations and the full-scale intelligence quotient (IQ), not all have been statistically significant after control for some potential confounders (Needleman et al., 1979; Fulton et al., 1987; Hatzakis et al., 1987; Fergusson et al., 1988; Yule et al., 1981; Lansdown et al., 1986; Schroeder et al., 1985; Hawk et al., 1986; Bellinger et al., 1987). These studies are supported by studies in animals. In primates, Rice and Wiles (1979) found learning and attention-deficit disorders. Studies in rodents have shown cognitive disorders and interference with the dopaminergic system in the brain (Cory-Schlecta et al., 1981). Inhibition of long-term potentiation in the hippocampal region by lead (Munoz et al., 1988) has also been demonstrated in rats at moderate blood lead levels, which is again consistent with learning disorders. Meta-analyses of the human data have reported evidence for the association (Needleman and Gatsonis, 1990; Schwartz et al., 1985). These associations may or may not indicate a cause-effect relation (see discussion in NRC, 1993).
Several lessons may be drawn from these studies. First, a toxicant may have numerous neurologic effects, though it may be difficult to measure real but subtle behavioral changes. Second, markers of neurotoxic effect are often obtained simultaneously with markers of dose, making it difficult to discern the temporal or pathologic sequence of exposure, dose, and response. Third, neurotoxicant-induced alterations of neurobehavioral, neurophysiologic, and neurochemical function are believed to precede morphologic evidence of toxicity and to be more sensitive. However, functional indicators can be compromised by the adaptive capacity of the individual, especially with moderate to low levels of exposure. If the function of the nervous system is viewed as an adaptive process, it is
logical to predict that at sufficiently high exposures the functional reserve of the individual will be depleted and performance will deteriorate (Tilson and Mitchell, 1983; Bleeker and Agnew, 1987). This implies a threshold at the individual level. However, if person-to-person variations in biologic effect are great, the overall group effect of exposure may have no detectable threshold and individual dose-response relationships may be lost because of influences by genetics, age, sex, prior experience, overall health, and adaptive capacities.
Reproductive and Developmental Outcomes
The term reproductive and developmental toxicity refers to maternal, paternal, pregnancy, and fetal effects. The exposure of pregnant women to drugs or other toxicants has long been recognized to produce adverse pregnancy outcomes, and effects on both males and females of exposures before conception have also been reported.
Fecundity (biologic capacity to have a child, whether or not one does so) may be affected by several mechanisms, including decreased fertility (reproductive capacity expressed as number of liveborn children) of either partner or the desire to have children. A report that exposure to the pesticide dibromochloropropane affected the fecundity of exposed men (Whorton et al., 1977) spurred studies of more-subtle indicators to determine whether other chemicals may also adversely affect male reproductive capacity. Epidemiologic reports have linked paternal occupational exposures to adverse pregnancy outcomes (Davis et al., 1992). Data from the National Toxicology Program indicate that chemicals affecting the fecundity of male rodents usually affect the fecundity of the female as well. Research to assess the fecundity and fertility of women exposed to potential toxicants is growing.
Assessment of the effects of potential toxicants on male fecundity is conducted primarily through endocrine and semen analyses. The basic methods (Schrader et al., 1987, 1992) attempt to evaluate effects on the neuroendocrine system, testes, and accessory sex glands. Hormone levels are used to assess the neuroendocrine system. Sperm count, morphology, and morphometry are used to assess testicular function. Sperm-cell function, including motility and viability, and semen biochemistry are useful in evaluating accessory sex gland function.
Well-designed human field studies are needed to evaluate the many chemicals in our environment. Better laboratory methods are needed to assess semen, genetic damage in the sperm, and sexual function. Research
is also needed in recruitment strategies to increase participation rates in field studies, thus decreasing potential bias. Research is needed to understand the implications of animal data for human risk.
Gaps in knowledge about fecundity are even larger for females than for males. Data from well-designed human studies of potential exposures are needed. Data on the ''normal" values and their interindividual and intraindividual variations are needed for designing future studies. Most early studies of women and reproductive toxicity were centered around adverse outcomes of pregnancy, with little interest in the women's fecundity. The data reported by the National Toxicology Program's Continuous Breeding Protocol indicated that many toxicants being tested were affecting female rat fecundity.
New methods are now being assessed to detect neuroendocrine function, biochemical changes, and ovulation in humans (Hughes, 1988; Wright et al., 1992). Research needs include the development of practical, noninvasive methods for field studies of exposed humans, including better epidemiologic methods for the recruitment and study of human populations. Further research should address female sexual function, accessory sex organs, and genetic damage to the ova.
Adverse pregnancy outcomes include miscarriage and low birthweight, as well as malformations and functional abnormalities. Most of the findings on chemicals causing adverse pregnancy outcomes have been established in animal studies. Two major exceptions illustrate that human fetuses are at risk from chemical exposures: the therapeutic use of thalidomide (McBride, 1961) and environmental exposures to mercury in Minamata Bay, Japan (Koos and Longo, 1976).
A few noninvasive laboratory methods are useful for the assessment of adverse effects on human pregnancies. Recently developed "ultra-sensitive" assays of human chorionic gonadotropin can detect pregnancy losses around the time of the first postconception menstrual cycle, and studies using this technology may provide insight into early pregnancy loss due to environmental exposures (NRC, 1989a).
Birth Defects and Developmental Effects
Developmental effects, including birth defects, have been studied as possible effects of toxic pollutants from point sources. This is because such effects can be readily monitored with either existing surveillance systems or special studies and because they have shorter latency periods than
cancer. Some of the studies evaluating exposures that might increase the risk of birth defects, congenital anomalies, and low birthweight were reviewed in volume 1 (NRC, 1991). One of these sets of studies, performed by the California Department of Health Services (CDHS), was updated in a special issue of Epidemiology (Swan et al., 1992; Deane et al., 1992; Zierler, 1992).
CDHS investigated cardiac defects among babies born in Santa Clara County in 1981 to September 1, 1982. The investigators compared the rates of cardiac defects in the study area (suspected of water contamination) with the rates for the rest of the county. Investigators found an increased risk of cardiac defects in the study area (RR = 2.6, p = 0.01). The results could not be due to recall bias, because the information is documented by charts. However, the investigators also compared the times when the birth defects occurred with when the water contamination occurred and concluded that the timing could not link them. Investigators mention other possible exposures, such as air contaminants and other contaminants in the water, that might have increased the rates of cardiac defects.
In a second study, CDHS compared the Los Paseos census tract, which received water from one contaminated well, with a control census tract that had demographic characteristics similar to those of Los Paseos. Women who had been pregnant in 1980 and/or 1981 were contacted by mail, telephone, or home visit and interviewed to determine rates of spontaneous abortion, congenital anomalies, and low birthweight, as well as various risk factors. After adjustment for confounding variables, the Las Paseos spontaneous abortion rate was 2.3 times that in the control community (odds ratio, OR = 2.3, 95% confidence interval, CI = , 1.3-4.2). Drinking tap water was also found to be associated with spontaneous abortions in both the control area and case area. The congenital anomaly rate was 3 times that of the control area (OR = 3.1, 95% CI = 1.1-10.4). However, the authors commented that the observed pattern is not consistent with a single teratogen, but rather with several teratogens affecting the fetus at different times during gestation. They did not find any low birthweight babies in the Los Paseos area. Tests for chemicals in the control area did not show chemical contamination. The investigators concluded that the exposure data were insufficient to determine whether the leak into the contaminated well was a cause of the increased rates of spontaneous abortions and congenital malformations.
The relation between environmental exposures and birth defects remains important but difficult to study. An innovative study from New York state used existing data to detect some associations between congenital malformations (all types pooled) and residential proximity to hazardous-waste sites (Geschwind et al., 1992). This study examined unusu-
ally large numbers of subjects and used more than one exposure variable. The investigators linked a congenital-malformations registry and a hazardous-waste site inspection program to evaluate this relation. The congenital-malformations registry collects information from hospitals, medical facilities, and private physicians on all children born alive in New York state with a diagnosis before 2 years of age of congenital malformations, chromosomal anomalies, or persistent metabolic defects. Ongoing audits show that the data are 95% complete and accurate. They linked the information from this database to residence within 1-mile of 590 waste sites in New York state, excluding those in New York City and those in rural areas that were not divided into census tracts. Investigators examined data on 9,313 newborns with congenital malformations and 17,802 healthy controls. Regressions controlled for maternal age, race, education, complications during pregnancy, parity, population density, and sex of child. They found a statistically significant OR of 1.12 (95% CI = 1.06-1.18) for bearing children with any kind of congenital malformation. Within the 1 mile area, babies born near off-site chemical leaks had higher rates of malformation than did those in areas without chemical leaks (RR = 1.17, 95% CI = 1.08-1.27).
Some studies indicate that the development of children exposed in utero to polychlorinated biphenyls (PCBs) is moderately impaired. Groups of children in Michigan and North Carolina have been studied from birth to age 4 years. In Michigan, higher PCB levels in umbilical cords were linked to reduced head size, diminished chest girth, and shorter gestation; these effects also occur in children whose mothers have been exposed to PCBs occupationally (Tilson et al., 1990). Higher PCB levels were also associated with lower scores on standardized tests for infant development and reduced activity. In North Carolina, a study of 912 infants followed since birth has shown that levels of PCBs commonly encountered in the United States "produce detectable effects on motor maturation and some evidence of impaired infant learning" (Tilson et al., 1990).
Lead has long been linked to adverse pregnancy outcomes and, indeed, was used as an abortifacient early in this century. Fetal loss has been associated with lead exposure, although control for potential confounding factors in those studies was poor. More recently, occupational lead exposure has been associated with several kinds of adverse pregnancy outcomes (Schwartz, 1992). Prospective studies of pregnant women who were not occupationally exposed have associated higher blood or cord lead levels with lower birthweight, small size for gestational age, and shorter duration of gestation. Higher placental lead levels have been associated with pregnancy loss (Schwartz, 1992). Not all these outcomes have been seen in every study, however.
Hepatic and Renal Outcomes
The liver and kidneys are directly involved in the body's handling of chemical exposures. The liver is the primary organ for metabolism of these substances, and the kidney is the primary organ for excretion of them and any toxic metabolic products. Both tend to concentrate toxic substances to levels well above those in the blood. Thus, the liver and kidney may have unusual patterns of exposure. The health effects of environmental toxicants on the liver can include enlargement, necrosis, fibrosis, cirrhosis, veno-occlusive disease, granulomas, lymphocyte infiltration, and cancer (Van Thiel, 1986; Tamburro and Liss, 1986).
Human exposures to low levels of many environmental and occupational chemicals can cause adaptive changes in the liver that may indicate exposure or early disease (Rubin, 1987; Van Thiel, 1986). In some situations, adaptive responses, though not toxic themselves, can indicate exposure. For example, various pesticides may have effects on liver function that indicate exposure. Several biologic indexes provide markers of hepatic function. The kidney has the dual role of separating unwanted from wanted substances and excreting the former. The high blood flow through the kidneys exposes them to the variety of chemicals absorbed as a result of breathing, ingesting, or skin absorption. Identifying and preventing environmentally induced renal disease is important because most types of renal disease are irreversible once substantial loss of renal function has occurred. Renal toxicants may produce glomerular nephritis, tubular necrosis, Fanconi syndrome, interstitial nephritis, hypersensitivity reaction, kidney stones, and cancer.
The influence of environmental exposures on the kidney has been widely described in the occupational literature (Littorin et al., 1984; Bernard et al., 1979; Friberg et al., 1985, Druet et al., 1982). A major impediment to recognition of early kidney disease is the lack of clinical and laboratory tests that are sensitive and specific (Goyer, 1987). The most-common clinical indicator of renal disease, serum creatinine, is not generally abnormal (between 1.5 and 2.0 mg/dL of blood) until 50% or more of renal function is lost.
A wide range of immunologic changes can result from exposures to xenobiotics. These have been extensively reviewed in recent reports from NRC and others (NRC, 1992a; Luster et al., 1989; Koller, 1987). The immune system defends the body by responding to exposure or stress. Distinguishing the changes that are defensive from those that are toxic is a current challenge.
As markers of adverse immunologic change are refined and validated, environmental-epidemiologic studies will be able to incorporate them in studies of exposure and effect. For further information on this developing field, readers are referred to Biologic Markers in Immunotoxicology (NRC, 1992a).
Biologic Markers in Environmental Epidemiology
Environmental-epidemiologic research of the future will include biologic markers of exposure, effect, and susceptibility, which were reviewed in volume 1 and in other NRC reports. These markers can reduce misclassification of exposure, refine the classification of disease, identify mechanisms of causation, and pinpoint high-risk populations and persons. The primary gap in existing information on various markers is the need for validation (NRC, 1989b). This requires that assays be assessed for sensitivity, specificity, and reproducibility. Once they are validated, laboratories must be able to conduct the relatively large number of assays generated by epidemiologic studies (Schulte, 1991; Hulka, 1991). In addition, costs of most biologic markers must be reduced before they will become relevant for epidemiologic study designs. Further, epidemiologic use will require information on how a marker varies with a wide range of factors, such as age, race, sex, time of day, season, drinking, smoking, use of medication, exercise, and concurrent diseases. With such information in hand, it will be easier to tell which markers to test for, how to use them, and how to interpret the findings.
Developments in molecular biology over the past decade may fundamentally alter the field, as well as provide a richer array of end points for statistical analysis. Axelson (1991) has argued that biomarkers could lead to new definitions of disease entities that combine clinical or histopathologic criteria with biochemical or genetic characteristics. Thus, recent work on the metabolism of debrisoquine, antipyrine, and other compounds has indicated that persons with some enzyme patterns have increased relative risks of cancer of the lung, pancreas, and stomach. Subjects who were extensive debrisoquine metabolizers and who also had occupational exposure to asbestos had an 18-fold increase in the rate of lung cancer.
There has been little research on indicators or markers of early or preclinical environmentally induced disease. Identification of such indicators can serve as both an early warning of potential problems and a guide to interventions for prevention or control. The key question in this regard is when a biologic change indicates a disease (Goyer and Rogan, 1986). From the epidemiologic perspective, a biologic change that is a candidate as an early indicator of disease must significantly differ between those
who will develop disease (in the absence of intervention) and those who will not.
The use of laboratory tests to assess biologic change requires the availability of validated biologic markers, the collection and storage of biologic specimens, and methods to integrate biologic measurements with observations, such as data from questionnaires. The validation of markers pertains to both the laboratory and the field. Before use in environmental epidemiology, population prevalence, sensitivity, and specificity of a marker need to be determined. This determination is also influenced by how markers vary among groups characterized by demographic, behavioral, or genetic factors. Once a study using biologic markers is planned, it is necessary to consider how specimens will be collected, transported, and stored, which can affect the ability to detect associations between exposures and outcomes.
Ultimately, biologic markers will be used in environmental epidemiology only if they are shown to contribute to the understanding of environmental hazards. Markers are not an end, but only a means to an end. Researchers should be able to identify what research questions each marker will answer and how the biomarker data will be used before collection of biologic specimens. Failure to consider these factors can lead to wasteful efforts.
Field studies of chemically exposed populations have made little use of immune-system responses to define exposures (NRC, 1989b; Karol, 1987; Grammer et al., 1988; Stejskal et al., 1986; Zeiss et al., 1977). There are several possibilities for developing the field of immunotoxicology (NRC, 1992). Better use of biomarkers of immunologic function requires an understanding of the interval between exposure and the appearance of the marker. Ideally, a specific exposure will cause the expression of a specific marker at a known time after exposure. However, human populations are generally exposed to mixtures, and a marker or battery of markers may be able to depict only qualitatively whether individuals have been exposed to immunomodulating substances (Biagini et al., 1986). Exposure assessment might be improved by using ordinal- and interval-scale characterizations of dose-dependent immunologic responses as measures of exposure. Karol (1983, 1987) demonstrated that an immunologic response (immunoglobulin G cytophilic antibodies) was dose-dependent over the range of 1-30 mg/kg in guinea pigs exposed to toluene diisocyanate vapors.
A marker cannot be used to estimate exposure without some data on reference levels in the unexposed population. There is a need to standardize the collection of such information and obtain appropriate demographic and other information about individual subjects. Even when information from the general population is available, data that assess immunologic markers of exposure or effect must be obtained from spe-
cific control populations that are adequately characterized in such variables as age, race, sex, geographic region, socioeconomic status, nutritional status, life style, and sexual habits.
The selection of immunologic markers to be used in a battery of tests to define exposure requires careful study. A battery of immunologic markers for assessing exposure might be quite different from one used to assess toxicity. Conversely, there may be value in developing a single battery that is less than optimal for specific tests but is familiar, as well as widely known and widely used. If the battery is to be comprehensive, it is often useful to employ some nonimmunologic markers as well. For example, the determination of a covalently bound DNA adduct, coupled with some measure of immunoreactivity, might provide a useful measure of exposure and immunologic response.
Since the immune system generally has some degree of functional reserve capacity, it is not likely that gross indicators, such as cell counts, will be useful as early indicators of exposure (Bick, 1985). One purpose of individual exposure assessment is to identify members of a possibly exposed population who are actually exposed.
The normal range of some immunologic parameters in humans includes extensive interindividual and intraindividual variability. In a 14-month study of healthy individuals between 21 and 70 years of age, the coefficient of variation for 13 of 16 immunologic parameters was greater than 30% (Dorey and Zighelboim, 1980). However, for all the parameters measured in a short period, intraindividual variability was significantly less than interindividual variability. The interindividual variability was due, in general, to subjects who consistently exhibited functional levels and cell numbers that were significantly higher or lower than the population mean. No significant correlation was found between age and most of the immunologic parameters examined, though this does not rule out significant age-related changes in other immune end points. Animal studies have shown age-related changes in immunity (Bick, 1985). Large intraindividual coefficients of variation determined for various immunologic parameters do not preclude the use of such markers to assess exposure or effect. In industrial and environmental exposure measurements, it is not unusual to find coefficients of variation as high as those found with immunologic parameters. Some implications of large coefficients of variation in immunologic markers that may be used to define exposure are that large samples must be used to reduce random variation, comparisons between exposed and nonexposed persons might involve stratification or adjustment for populations whose results significantly deviate from the mean, and construct validity must be established for multiple markers to ensure that confounders are distributed similarly between exposed and nonexposed groups.
In summary, immunologic markers of exposure can be used in several ways to distinguish exposed from nonexposed individuals and to distinguish among those with different levels of exposure within exposed groups. The use of immunologic markers should be considered in the context of other information, such as exposure history or environmental or breathing-zone measurements. The use of biologic markers may need consideration during the design of a study, during implementation, and during the interpretation of results. However, before any study, the investigator should ask what benefit biologic markers might provide over exposure assessment by ambient personal or environmental monitoring. If this question cannot be answered in a convincing way, biologic markers should not be used in environmental-epidemiologic research.
The frequency of many diseases is likely to increase in certain population groups. For example, an increasingly large proportion of the population is over age 65 (Cooper et al., 1991). Older persons have an increased vulnerability to many stresses. Whether and to what extent an elderly person is more susceptible to the toxic effects of potentially hazardous compounds from the environment is largely unknown. Elderly individuals may be at increased risk from toxicant exposure because of age-related changes in the body's protective mechanisms (NRCb, 1989). Older persons have had a long period of exposure to chemicals, more time for latent adverse effects to manifest themselves, and more time for cumulative effects to emerge.
The role of aging in environmental toxicity has been a subject of extensive research. In aged populations, 2 classes of adverse effects may exist: those caused by aging alone and those caused by interactions between aging and toxic agents (Williams et al., 1987). It may not be possible to discern with confidence the health decrements from these. A problem, then, is the identification of experimental approaches that will give insight into the relation between age and toxicity (NRC, 1989b).
A major impediment to understanding the role of environmental agents in the aging process is the paucity of reliable and valid biomarkers of aging per se (Reff and Schneider, 1982; Ingram, 1988; McClearn, 1988). Harrison (1988) suggested 4 criteria for determining whether a particular physiologic assay may be useful as a biomarker of aging in an individual organism: the results should change significantly with age, the changes should be repeatable in the same individual, assays of independent physiologic parameters should give similar estimates of age for the same individual, and the degree of aging as determined by the assays should correlate with subsequent longevity.
Children also have heightened sensitivity to many environmentally induced diseases. The susceptible period of childhood can be viewed as beginning with influences on the sperm and egg and, hence, on preconception parental exposure and life styles, continuing with the impacts on embryos and fetuses, and including the particular susceptibility of neonates and infants. Research is needed to catalog environmental factors that influence each of these stages (NRC, 1989b). The immature enzymatic detoxification systems of embryos, fetuses, and neonates put them at increased risk of the effects of pharmacologic and environmental substances that pass through the placenta, such as heavy metals, PCBs, and immunogens.
Information is scanty about racial and ethnic differences in susceptibility to disease, though these also merit careful consideration. Lead poisoning affects children of all races but occurs with especially high frequency in inner-city African Americans. Other diseases also occur more commonly among different ethnic groups and reflect genetic, nutritional, and environmental factors that need to be systematically evaluated. Personal choice of life style, such as smoking or use of alcoholic beverages, and personal exposures, such as use of medication, may change susceptibility.
The recent and rapid increase in knowledge about molecular biology and genetics has identified, at the gene or molecular level, an increased number of susceptible groups. These have implications for research and for societal response to environmental hazards. The key to future epidemiologic research is the linkage of susceptibility markers to exposures and outcomes in ways that will answer questions about why, in similarly exposed groups, some people develop disease and others do not. Susceptibility markers can be used to differentiate subgroups of populations in order to identify risks.
The ability to distinguish sensitive subgroups in populations has legal, ethical, and social implications that need to be assessed in detail. For example, labeling individuals as susceptible has been alleged to have effects on psychosocial status and function, job security, property values, obtaining insurance, and finding a mate (Ashford et al., 1990). Recent evidence that each human has, on average, 5-10 genetic abnormalities may diminish the social and psychologic impact of knowing about a specific abnormality in a specific person.
The types, nature, and extent of environmental influences on humans are unclear. Substantial evidence associates air-pollution exposure with some respiratory outcomes. The toxicities of lead are increasingly well understood. The rate of environment-related risks of other outcomes is
less well understood, and further work is necessary to delineate environmental risks. Researchers should use a range of strategies to study possible relations of health outcomes to environmental exposures. These strategies include the following:
- Agencies concerned with the prevention of disease and the promotion of health should make a deliberate effort to identify diseases and syndromes that appear to be increasing or to have an important impact on public health. Specific causes of death and illness, as well as such functional afflictions as impaired vision and effects on reproductive health, need to be considered. Nomenclature and case definitions should be standardized for these diseases and syndromes.
- When time trends reveal recent shifts in disease patterns, researchers should explore possible etiologic explanations. Health-information databases should be designed to facilitate linkage to exposure information and other relevant factors, and vice versa, so that analyses can be extended from local patterns to relations with possible causal factors
- Researchers should search for and study cohorts with exposures to toxicants that might be associated with specific health outcomes in order to test hypotheses about association. Whenever possible, high-exposure populations and those with a range of exposures should be selected
- Opportunities should be exploited to add environmental assessments to existing research on end points where there is a potential environmental risk. This could be a cost-effective approach to epidemiology.
- Broad evaluations of symptom prevalence and disease history should be a part of selected environmental investigations, and findings that suggest effects on specific organ systems should be followed with more-focused studies.
- Tools to identify susceptible subpopulations need substantial further development and should be used to identify associations of exposures with specific effects. A broad range of criteria-such as age, race, metabolic phenotypes, and socioeconomic status-should be used.
- Data gaps exist in reproductive epidemiology, and we urge renewed research focus on this field. Important data gaps occur in the study of pregnancy, including early pregnancy loss, and extrapolations from animal studies have generally not been validated in humans. Data gaps also exist on male-mediated adverse pregnancy outcomes. An important research need is the development of measures and tools to identify toxic effects on fecundity.
- Longitudinal studies should be conducted to assess the significance of biologic markers and preclinical effects suggestive of disease, including their use in calculating and assessing past exposures.
- Biologic markers need to be validated for specificity for exposure
- and health outcome. Many of the markers that have been identified and characterized in the laboratory have not been specifically tied to chemical, biologic, or physical exposures. Validation of these will strengthen the ability of environmental epidemiology to discern relations between exposure and disease. Research is needed to identify additional direct and indirect markers of exposure and disease.
- Environmental exposure to some substances is now routinely monitored. The planning of such monitoring should incorporate factors that will increase its utility for epidemiologic studies.
American Thoracic Society, Committee of the Environmental and Occupational Health Assembly: R. Bascom, P.A. Bromberg, D.A. Costa, R. Devlin, D.W. Dockery, M.W. Frampton, W. Lambert, J. M. Samet, F. E. Speizer, and M. Utell. 1996a. Health effects of outdoor air pollution. Part 2. Am. J. Resp. Crit. Care Med. 153: 477-498.
American Thoracic Society, Committee of the Environmental and Occupational Health Assembly: R. Bascom, P.A. Bromberg, D.A. Costa, R. Devlin, D.W. Dockery, M.W. Frampton, W. Lambert, J. M. Samet, F. E. Speizer, and M. Utell. 1996b. Health effects of outdoor air pollution. Part 1. Am. J. Resp. Crit. Care Med. 153: 3-50.
Ashford, N.A., C.J. Spadafor, D.B. Hattis, and C.C. Caldart. 1990. Monitoring the Worker for Exposure and Disease: Scientific, Legal, and Ethical Considerations in the Use of Biomarkers. Baltimore: Johns Hopkins University Press. 224 pp.
Archer, V.E. 1990. Air pollution and fatal lung disease in three Utah counties. Arch. Environ. Health 45:325-334.
Axelson, O. 1991. Cancer and combined exposures to occupational and environmental factors. Rec. Res. Cancer Res. 122:60-70.
Bates, D.V., and R. Sizto. 1987. Hospital admissions and air pollutants in Southern Ontario: the acid summer haze effect. Environ. Res. 43:317-331.
Bates, D.V., M. Baker-Anderson, and R. Sizto. 1990. Asthma attack periodicity: a study of hospital emergency visits in Vancouver. Environ. Res. 51:51-70.
Bellinger, D., A. Leviton, C. Waternaux, H. Needleman, and M. Rabinowitz. 1987. Longitudinal analyses of prenatal leads exposure and early cognitive development. N. Engl. J. Med. 316:1037-1043.
Bernard A., J.P. Buchet, H. Roels, P. Masson, and R. Lauwerys. 1979. Renal excretion of proteins and enzymes in workers exposed to cadmium. Eur. J. Clin. Invest. 9:11-22.
Biagini, R.E., W.J. Moorman, T.R. Lewis, and I.L. Bernstein. 1986. Ozone enhancement of platinum asthma in a primate model. Am. Rev. Respir. Dis. 134:719-725.
Bick, P.H. 1985. The immune system: organization and function. Pp. 1-10 in J. Dean, M.I. Luster, A.E. Munson, and H. Amos, eds. Immunotoxicology and Immunopharmacology. New York: Raven.
Bleeker, M.L., and J. Agnew. 1987. The assessment of the dose-response relationship for low-level exposure to neurotoxicants in man. Pp. 508-515 in V. Foa, E.A. Emmett, M. Maroni, and A. Colombi, eds. Occupational and Environmental Chemical Hazards: Cellular and Biochemical Indices for Monitoring Toxicity. Chichester: Ellis Horwood.
Bos, P.M., G. de Mik, and P.C. Bragt. 1991. Critical review of the toxicity of methyl n-butyl ketone: risk from occupational exposure. Am. J. Ind. Med. 20:175-194.
Braun-Fahrlander, C., U. Ackermann-Liebrich, J. Schwartz, H.P. Gnehm, M. Rutishauser, and H.U. Wanner. 1992. Air pollution and respiratory symptoms in preschool children. Am. Rev. Respir. Dis. 145:42-47.
Brunekreef, B., P.L. Kinney, J.H. Ware, D. Dockery, F.E. Speizer, J.D. Spengler, and B.G. Ferris, Jr. 1991. Sensitive subgroups and normal variation in pulmonary function response to air pollution episodes. Environ. Health Perspect. 90:189-193.
Byers, R.K., and E.E. Lord. 1943. Late effects of lead-poisoning on mental development. Am. J. Dis. Child. 66:471
Chappie, M., and L. Lave. 1982. The health effects of air pollution: a reanalysis. J. Urban Econ. 12:346-376.
Chestnut, L.G., J. Schwartz, D.A. Savitz, and C.M. Burchfiel. 1991. Pulmonary function and ambient particulate matter: epidemiological evidence from NHANES I. Arch. Environ. Health 46:135-144.
Cooper, R.L., J.M. Goldman, and T.J. Harbin. 1991. Introduction: assessing environmental influences on the aging process. Pp. 1-6 in R.L. Cooper, J.M. Goldman, and T.J. Harbin, eds. The Johns Hopkins Series in Environmental Toxicology: Aging and Environmental Toxicology: Biological Perspectives . Baltimore: Johns Hopkins University Press.
Cory-Schlecta, D.A., S.T. Bissen, A.M. Young, and T. Thompson. 1981. Chronic post-weaning lead exposure and response during performance. Toxicol. Appl. Pharmacol. 60:78-84.
Dassen, W., B. Brunekreef, G. Hoek, P. Hofschreuder, B. Staatsen, H. de Groot, E. Schouten, and K. Biersteker. 1986. Decline in children's pulmonary function during an air pollution episode. J. Air Pollut. Control Assoc. 36:1223-1227.
Davis, D.L., G. Friedler, D. Mattison, and R. Morris. 1992. Male-mediated teratogenesis and other reproductive effects: biologic and epidemiologic findings and a plea for clinical research. Reprod. Toxicol. 6:289-292.
Deane, M., S.H. Swan, J.A. Harris, D.M. Epstein, and R. R. Neutra. 1992. Adverse pregnancy outcomes in relation to water consumption: a reanalysis of data from the original Santa Clara County study, California, 1980-1981. Epidemiology 3:94-97
Detels, R., D.P. Tashkin, J.W. Sayre, S.N. Rokaw, A.H. Coulson, F.J. Massey, Jr., and D.H. Wegman. 1987. The UCLA population studies of chronic obstructive respiratory disease. 9. Lung function changes associated with chronic exposure to photochemical oxidants; a cohort study among never-smokers. Chest 92:594-603.
Dockery, D.W., and C.A. Pope. 1994. Acute respiratory effects of particulate air pollution. An. Rev. Public Health 15:107-132
Dockery, D.W., F.E. Speizer, D.O. Stram, J.H. Ware, J.D. Spengler, and B.G. Ferris, Jr. 1989. Effects of inhalable particles on respiratory health of children. Am. Rev. Respir. Dis. 139:587-594.
Dorey, F., and J. Zighelboim. 1980. Immunologic variability in a healthy population. Clin. Immunol. Immunopathol. 16:406-415.
Druet P., A. Bernard, F. Hirsch, J.J. Weening, P. Gengoux, P. Mahieu, and S. Birkeland. 1982. Immunologically mediated glomerulonephritis induced by heavy metals. Arch. Toxicol. 50:187-194.
Euler, G.L., D.E. Abbey, A.R. Magie, and J.E. Hodgkin. 1987. Chronic obstructive pulmonary disease symptom effects of long-term cumulative exposure to ambient levels of total suspended particulates and sulfur dioxide in California Seventh-Day Adventist residents. Arch. Environ. Health 42:213-222.
Euler, G.L., D.E. Abbey, J.E. Hodgkin, and A.R. Magie. 1988. Chronic obstructive pulmonary disease symptom effects of long-term cumulative exposure to ambient levels of total oxidants and nitrogen dioxide in California Seventh-Day Adventist residents. Arch. Environ. Health 43:279-285.
Fairley, D. 1990. The relationship of daily mortality to suspended particulates in Santa Clara County, 1980-1986. Environ. Health Perspect. 89:159-168.
Fergusson, D.M., J.E. Fergusson, L.J. Horwood, and N.G. Kinzett. 1988. A longitudinal study of dentine lead levels, intelligence, school performance and behavior. III. Dentine lead levels and cognitive ability. J. Child Psychol. Psychiatry 29:793-809.
Friberg, L., C.G. Elinder, T. Kjellstrom, and G.F. Nordberg. 1985. Cadmium and health: a toxicological and epidemiological appraisal. 1:159-160.
Fulton, M., G. Raab, G. Thomson, D. Laxen, R. Hunter, and W. Hepburn. 1987. Influence of blood level on the ability and attainment of children in Edinburgh. Lancet 1:1221-1226.
Geschwind, S.A., J.A.J. Stolwijk, M. Bracken, E. Fitzgerale, A. Stark, C. Olsen, and J. Melius. 1992. Risk of congenital malformations associated with proximity to hazardous-waste sites. Am. J. Epidemiol. 135:1197-1207.
Gochfeld, M., N. Fiedler, and J. Burger. 1991. Neurobehavioral toxicology. Pp. 325-332 in J.M. Last and R. B. Wallace, eds. Maxcey-Roseneau-Last Public Health and Preventive Medicine. 13th ed. Norwalk, CT: Appleton and Lange.
Goyer, R.A. 1987. Biochemical and cellular indices of renal charges induced by exogenous chemicals. Report of session IV. Pp. 368-371 in Foa V., E.A. Emmett, M. Maroni, and A. Colombi, eds. Occupational and Environmental Chemical Hazards. Cellular and Biochemical Indices for Monitoring Toxicity. Chichester: Ellis Horwood.
Goyer, R.A., and W.J. Rogan. 1986. When is biologic change an indicator of disease? Pp. 17-26 in D.W. Underhill and E.P. Radford, eds. New and Sensitive Indicators of Health Impacts of Environmental Agents. Third Annual Symposium on Environmental Epidemiology, held April 26-28 in Pittsburgh, PA. Pittsburgh, PA: Center for Environmental Epidemiology, Graduate School of Public Health, University of Pittsburgh.
Grammer, L.C., P. Eggum., M. Silverstein, M.A. Shaughnessy, J.L. Liotta, and R. Patterson. 1988. Prospective immunologic and clinical study of a population exposed to hexamethylene diisocyanate. J. Allergy Clin. Immunol. 82:627-633.
Gross, J., J.R. Goldsmith, L. Zangwill, and S. Lerman. 1984. Monitoring of hospital emergency room visits as a method for detecting health effects of environmental exposures. Sci. Total. Environ. 32:289-302.
Harrison D.E., and J. R. Archer. Biomarkers of aging: tissue markers. 1988 Future research needs, strategies, directions, and priorities. Exp. Gerontol. 23: 309-325.
Hasselblad, V., D.M. Eddy, and D.J. Kotchmar. 1992. Synthesis of environmental evidence: nitrogen dioxide epidemiology studies. J. Air Waste Manage. Assoc. 42:662-671.
Hatzakis, A., K. Katsoynanni, A. Kalandidi, N. Day, and D. Trichopoulos. 1986. Short-term effects of air pollution on mortality in Athens. Int. J. Epidemiol. 15:73-81.
Hatzakis A., A. Kokkevi, K. Katsouyanni, K. Maravelias, J.F. Salaminios, A. Kalandidi, and A. Koutselinis. 1987. Psychometric intelligence and attentional performance deficits in lead-exposed children. Pp. 204-209 in S.E. Lindberg and T.C. Hutchinson, eds. Heavy metals in the Environment. Vol. 1. New Orleans/Edinburgh: CEP Consultants.
Hawk, B.A., S. R. Schroeder, G. Robinson, P. Mushak, D. Kleinbaum, and G. Dawson. 1986. Relation of lead and social factors to IQ of low-SES children: a partial replication. Am. J. Ment. Defic. 91: 178-183.
Hudnell, H.K., D.A. Otto, D.E. House, and L. Mohave. 1992. Exposure of humans to a volatile organic mixture. II. Sensory. Arch. Environ. Health 47:31-38.
Hughes, C.L. 1988. Monitoring of ovulation in the assessment of reproductive hazards in the workplace. Reprod. Toxicol. 2:163-170.
Hulka, B.S. 1991. Epidemiological studies using biological markers: issues for epidemiologists. Cancer Epid. Biomarkers Prevention 1:13-19.
Ingram, D.K. 1988. Key questions in developing biomarkers of aging. Exp. Gerontol. 23:429-434.
Kalfakis, N., D. Vassilopoulos, C. Voumvourakis, M. Ndjeveleka, and C. Papageorgiou. 1991. Amyotrophic lateral sclerosis in southern Greece: an epidemiologic study. Neuroepidemiology 10:170-173.
Karol, M.H. 1983. Concentration-dependent immunologic response to toluene diiosocyanate (TDI) following inhalation exposure. Toxicol. Appl. Pharmacol. 68:229-241.
Karol, M.H. 1987. The development of an animal model for TDI asthma. Bull. Eur. Physiopathol. Respir. 23:571-576.
Katsounnayi, K., A. Hatzakis, A. Kalandidi, and D. Trichopoulos. 1990. Short-term effects of atmospheric pollution on mortality in Athens. Arch. Hellenic Med. 7:126-132.
Kinney, P.L., and H. Ozkaynak. 1991. Associations of daily mortality and air pollution in Los Angeles County (California, USA). Environ. Res. 54:99-120.
Kinney, P.L., J.H. Ware, J.D. Spengler, D.W. Dockery, F.E. Speizer, and B.G. Ferris, Jr. 1989. Short-term pulmonary function change in association with ozone levels. Am. Rev. Respir. Dis. 139:56-61.
Koller, L.D. 1987. Immunotoxicology today. Toxicol. Pathol. 15:346-351.
Koos, B.J., and L.D. Longo. 1976. Mercury toxicity in the pregnant woman, fetus, and newborn infant: a review. Am. J. Obstet. Gynecol. 126:390-409.
Lansdown, R., W. Yule, M.A. Urbanowicz, and J. Hunter. 1986. The relationship between blood-lead concentrations, intelligence, attainment and behavior in a school population: the second London study. Int. Arch. Occup. Environ. Health 57: 225-235.
Lave, L.B., and E.P. Seskin. 1977. Air Pollution and Human Health. Baltimore: Johns Hopkins University Press for Resources for the Future. 368 pp.
Lioy, P.J., T.A. Vollmuth, and M. Lippmann. 1985. Persistence of peak flow decrement in children following ozone exposures exceeding the National Ambient Air Quality Standard. J. Air Pollut. Control Assoc. 35:1069-1071.
Lipfert, F.W. 1980. Sulfur oxides, particulates, and human mortality: synopsis of statistical correlations. J. Air Pollut. Control Assoc. 30:366-371.
Littorin, M., M. Welinder, and B. Hultberg. 1984. Kidney function in stainless steel welders. Int. Arch. Occup. Environ. Health 53:279-282.
Luster, M.I., M.F. Ackermann, D.P. Germolec, and G.J. Rosenthal. 1989. Perturbations of the immune system by xenobiotics. Environ. Health Perspect. 81:157-162.
McBride, W.G. 1961. Thalidomide and congenital anomalies. Lancet 2:1358.
McClearn, G.F. 1988. Strategies for biomarker research: experimental and methodological design . Exp. Gerontol. 23:245-255.
McDonnell, W.F., 3rd, R.S. Chapman, M.W. Leigh, G.L. Strope, and A.M. Collier. 1985. Respiratory responses of vigorously exercising children to 0.12 ppm ozone exposure. Am. Rev. Respir. Dis. 132:875-879.
McLachlan, D.R., T.P. Kruck, W.J. Lukiw, and S.S. Krishnan. 1991. Would decreased aluminum ingestion reduce the incidence of Alzeheimer's disease? Can. Med. Assoc. J. 145:793-804.
Melius, J.M., and P.A. Schulte 1981. Epidemiologic design for field studies: occupational neurotoxicity. Scand. J. Work Environ. Health 7 (Suppl. 4): 34-39.
Munoz, C., K. Garbe, H. Lilienthal, and G. Winneke. 1988. Significance of hippocampal dysfunction in low level lead exposure of rats. Neurotoxicol. Teratol. 10:245-253.
Needleman, H.L. 1986. Studies of psychologic performance of children with elevated dentine lead levels. Pp. 139-140 in D.W. Underhill and E.P. Radford, eds. New and Sensitive Indicators of Health Impacts of Environmental Agents. Third Annual Symposium on Environmental Epidemiology, held April 26-28, 1992, in Pittsburgh, PA . Pittsburgh, PA: Center for Environmental Epidemiology, Graduate School of Public Health, University of Pittsburgh.
Needleman, H.L., A. Leviton, and D. Bellinger. 1982. Lead associated intellectual deficit [letter]. N. Engl. J. Med. 306:367.
Needleman, H.L., C. Gunnoe, A. Leviton, R. Reed, H. Peresie, C. Maher, and P. Barrett. 1979. Deficits in psychologic and classroom performance of children with elevated dentine lead levels. N. Engl. J. Med. 300:689-695.
Needleman, H.L., and C.A. Gatsonis. 1990. Low-level lead exposure and the IQ of children. a meta-analysis of modern studies. J. Am. Med. Assoc. 263:673-678.
Ngim, C.N., and G. Devathasan. 1989. Epidemiologic study on the association between body burden, mercury level, and idiopathic Parkinson's disease. Neuroepidemiology 8:128-141.
NRC (National Research Council). 1989a. Biologic Markers in Reproductive Toxicology. Washington, DC: National Academy Press. 395 pp.
NRC (National Research Council). 1989b. Biologic Markers in Pulmonary Toxicology. Washington, DC: National Academy Press.
NRC (National Research Council.) 1991. Environmental Epidemiology. Vol. 1: Public Health and Hazardous Wastes. Washington, DC: National Academy Press.
NRC (National Research Council.) 1992a. Biologic Markers in Immunotoxicology. Washington, DC: National Academy Press.
NRC (National Research Council). 1992b. Multiple Chemical Sensitivities: Addendum to Biologic Markers in Immunotoxicology. Washington, DC: National Academy Press.
NRC (National Research Council). 1992c. Environmental Neurotoxicology. Washington, DC: National Academy Press.
NRC (National Research Council). 1993. Measuring Lead Exposure in Infants, Children, and Other Sensitive Populations. Washington, DC: National Academy Press
Ostro, B., and S. Rothschild. 1989. Air pollution and acute respiratory morbidity: an observational study of multiple pollutants. Environ. Res. 50:238-247.
Ostro, B.D., M.J. Lipsett, M.B. Wiener, and J.C. Selner. 1991. Asthmatic responses to airborne acid aerosols. Am. J. Pub. Health 81:694-702.
Otto, D.A., H.K. Hudnell, D.E. House, L. Mohave, and W. Counts. 1992. Exposure of humans to a volatile organic mixture. I. Behavioral assessment. Arch. Environ. Health 47:23-30.
Ozkaynak, H., J. Spengler, and A. Garzd, et al. 1986. Assessment of population health risks resulting from exposure to airborne particles. In S. D. Lee, ed. Aerosols: Research, Risk Assessment, and Control Strategies. Chelsea, MI: Lewis Publishers.
Pope, C.A. 1989. Respiratory disease associated with community air pollution and a steel mill, Utah Valley. Am. J. Pub. Health 79:623-628.
Pope, C.A. 1991. Respiratory hospital admissions associated with PM10 pollution in Utah, Salt Lake, and Cache valleys. Arch. Environ. Health 46:90-97.
Pope, C.A., and D.W. Dockery. 1992. Acute health effects of PM10 pollution on symptomatic and asymptomatic children. Am. Rev. Respir. Dis. 145:1123-1128.
Pope, C.A., D.W. Dockery, J.D. Spengler, and M.E. Raizenne. 1991. Respiratory health and PM10 pollution. A daily time series analysis. Am. Rev. Respir. Dis. 144:668-674.
Reff, M.F., and E.L. Schneider. 1982. Biological markers of aging. NIH Pub. 82-2221. Washington, DC: US Government Printing Office.
Rice, D.C., and R. F. Wiles. 1979. Neonatal low level exposure in monkeys (macac fasciculosis): effect on two-choice non-spatial form discrimination. J. Environ. Pathol. Toxicol. 2:1195-1203.
Rubin, R.J. 1987. Biological indices of enzyme induction as markers of hepatic alterations. Pp. 127-136 in Occupational and Environmental Chemical Hazards: V. Foa, Cellular and Biochemical Indices for Monitoring Toxicity. E.A. Emmett, M. Maroni, and A. Colombi, eds. Chichester: Ellis Horwood.
Schrader, S.M., J.M. Ratcliffe, T.W. Turner, and R.W. Hornung. 1987. The use of new field methods of semen analysis in the study of occupational hazards to reproduction: the example of ethylene dibromide. J. Occup. Med. 29:963-966.
Schrader, S.M., R.E. Chapin, E.D. Clegg, R.O. Davis, J.L. Fourcroy, D.F. Katz, S.A. Rothmann, G. Toth, T.W. Turner, and M. Zinaman. 1992. Laboratory Methods for Assessing Human Semen in Epidemiologic Studies: A Consensus Report. Reproductive Toxicology
Schroeder, S.R., B. Hawk, D.A. Otto, P. Mushak, and R. E. Hicks. 1985. Separating the effects of lead and social factors on IQ. Environ. Res. 91:178-183.
Schulte, P.A. 1991. New opportunities for interdisciplinary cancer communication. Cancer Epid. Biomarkers Prevention 1:3-4.
Schwartz, J. 1989. Lung function and chronic exposure to air pollution: a cross-sectional analysis of NHANES II. 1989. Environ. Res. 50:309-321.
Schwartz, J. 1991. Particulate air pollution and daily mortality in Detroit. Environ. Res. 56:204-213.
Schwartz, J. 1992. Low level health effects of lead: growth, developmental, and neurological disturbances. Pp. 233-242 in H.L. Needleman, ed. Human Lead Exposure. Boca Raton: CRC Press.
Schwartz, J., and A. Marcus. 1990. Mortality and air pollution in London: a time series analysis. Am. J. Epidemiol. 131:185-194.
Schwartz, J., H. Pitcher, R. Levin, B. Ostro, and A.L. Nicholas. 1985. Cost and Benefits of Reducing Lead in Gasoline: Final Regulatory Impact Analysis. Washington, DC: Office of Policy Analysis, US Environmental Protection Agency.
Schwartz, J., D.W. Dockery, J.H. Ware, et al. 1989. Acute effects of acid aerosols on respiratory symptiom reporting in children. Air Pollut. Control Assoc. Preprint 89-92.1. Effects of inhalable particles on respiratory health of children. Am. Rev. Respir. Dis. 139:587-594.
Schwartz, J., D. Slader, T. V. Larson, W. E. Pierson, and J. J. Koenig. 1993. Particulate air pollution and hospital emergency risks for asthma in Seattle. Am. Rev. Respir. Dis. 147:826-831.
Spektor, D.M., M. Lippmann, P.J. Lioy, G.D. Thurston, K. Citak, D.J. James, N. Bock, F.E. Speizer, and C. Hayes. 1988. Effects of ambient ozone on respiratory function in active, normal children. Am. Rev. Respir. Dis. 137:313-320.
Spektor, D.M., V.A. Hofmeister, P. Artaxo, J.A.P. Brague, F. Echelar, D.P. Nogueira, C. Hayes, G.D. Thurston, and M. Lippmann. 1991. Effects of heavy industrial pollution on respiratory function in the children of Cubatao, Brazil: a preliminary report. Environ. Health Perspect. 94:51-54.
Stejskal, V. D., R.G. Olin, and M. Forsbeck. 1986. The lymphocyte transformation test for diagnosis of drug-induced occupational allergy. J. Allergy Clin. Immunol. 77:411-426.
Sunyer, J., J.M. Anto, C. Murillo, and M. Saez. 1991. Effects of urban air pollution on emergency room admissions for chronic obstructive pulmonary disease. Am. J. Epidemiol. 134:277-286.
Swan, S.H., R.R. Neutra, M. Wrensch, I. Hertz-Picciotto, G.C. Windham, L. Fenster, D.M. Epstein, and M. Deane. 1992. Is drinking water related to spontaneous abortion? Reviewing the evidence from the California Department of Health Services studies. Epidemiology 3:83-93.
Tamburro, C.H., and G.M. Liss. 1986. Tests for hepatotoxicity: usefulness in screening workers. J. Occup. Med. 28:1034-1044.
Tanner, C.M., B. Chen, W.Z. Wang, M.L. Peng, Z.L. Liu, X.L. Liang, L.C. Kao, D.W. Gilley, and B.S. Schoenberg. 1987. Environmental factors in the etiology of Parkinson's disease. Can. J. Neurol. Sci. 14 (Supp.3):419-423.
Tilson, H.A., and C.L. Mitchell. 1983. Neurotoxicants and adaptive responses of the nervous system: introductory remarks. Fed. Proc. 42:3189-3190.
Tilson, H.A., and C.L. Mitchell. 1992. Neurotoxicology. New York: Raven Press.
Tilson, H.A., J.L. Jacobson, and W.J. Rogan. 1990. Polychlorinated biphenyls and the developing nervous system: cross-species comparisons. Neurotoxicol. Teratol. 12:239-248.
Valciukas, J.A., and R. Lilis. 1980. Psychometric techniques in environmental research. Environ. Res. 21:275-297.
Van Thiel, D.H. 1986. New approaches for identifying hepatotoxins. Pp. 87-98 in D.W. Underhill and E.P. Radford, eds. New and Sensitive Indicators of Health Impacts of Environmental Agents. Pittsburgh, PA: University of Pittsburgh.
Whorton, D., R.M. Krauss, S. Marshall, and T.H. Milby. 1977. Infertility in male pesticide workers. Lancet 2:1259-1261.
Wichmann, H.E., W. Mueller, P. Allhoff, M. Beckmann, N. Bocter, M.J. Csicsaky, M. Jung, B. Molik, and G. Schoeneberg. 1989. Health effects during a smog episode in West Germany in 1985. Environ. Health Perspect. 79:89-99.
Williams, J.R., P.S. Spencer, S.M. Stahl, J.F. Borzelleca, W. Nichols, E. Pfitzer, E.J. Yunis, R. Carchman, J.W. Opishinski, and R.A. Walford. 1987. Interactions of aging and environmental agents: the toxicological perspective. Prog. Clin. Biol. Res. 228:81-135.
Wright, D.M., J.S. Kessner, S.M. Schrader, N.W. Chin, V.E. Wells, and E.P. Krieg. 1992. Methods to monitor menstrual function in field studies: attitudes of working women. Reprod. Toxicol. 6.
Xintaras, C., J.R. Burg, B.L. Johnson, S. Tanaka, S.T. Lee, and J.H. Bender. 1979. Neurotoxic effects of exposed chemical workers. Ann. NY Acad. Sci. 329:30-38.
Yule, W., R. Lansdown, I. Millar, and M. Urbanowicz. 1981. The relationship between blood lead concentration, intelligence, and attainment in a school population: a pilot study. Dev. Med. Child. Neurol. 23:567
Zeiss, C.R., R. Patterson, J.J. Rouzansky, M.M. Miller, M. Rosenberg, and D. Levitz. 1977. Trimellitic anhydride-induced airway syndromes: clinical and immunologic studies. J. Allergy Clin. Immunol. 60:96-103.
Zierler, S. 1992. Drinking water and reproductive health. Epidemiology 3:77-78.