Use of Quantitative Exposure and Dose Reconstruction in Litigation Involving Human Health Issues

Brent Finley, PhD, DABT
Exponent

Several recent court decisions suggest that the scientific benchmark continues to be raised with respect to admissibility of expert testimony in cases involving chemical exposure and alleged health effects. In particular, failure to provide adequate evidence of a quantitative exposure analysis has been cited in numerous recent cases in which an expert's testimony has been dismissed. For example, in Current v. Atochem North America, Inc., No. W-00-332 (W.D. Tex. Dec. 17, 2001), the plaintiff's expert witness testimony regarding arsenic exposure as a cause of plaintiff's cancer was recently ruled unreliable and barred in federal court because the expert's opinion was not based on reliable evidence of Current's degree of exposure. In that case, Dr. Michael Gochfeld (the expert witness) did not provide a quantitative estimate of plaintiff's exposure but instead relied on literature reports that arsenic has been reported to induce specific types of cancers under certain exposure conditions. The court ruled that the minimal facts necessary to sustain the plaintiff's burden in a toxic tort case include "scientific knowledge of the harmful level of exposure to a chemical plus knowledge that the plaintiff was exposed to such quantities." These and other decisions are evidence of the increasing expectations regarding exposure assessment testimony - expectations that have arguably not been enforced consistently, even in the recent past. For example, in the 1996 "Erin Brockovich" case (Anderson, et al. v. PG&E), which ended in a substantial settlement for the plaintiffs, exposure experts for the plaintiffs opined primarily on causation and did not present quantitative exposure or dose estimates for any of the 600 plaintiffs.

Quantitative exposure assessments conducted in support of litigation may assume many forms. Most attempt to follow the "classical" methodologies used in the environmental enforcement programs (CERCLA, RCRA, etc.), in which published exposure factors are used in conjunction with measured or estimated chemical concentrations to derive an estimate of the plaintiffs' exposure. The estimated doses are then compared to chemical exposures that are thought or known to be associated with adverse health effects, as determined from animal toxicity data or occupational and epidemiological studies. However, while environmental risk assessments are based on current data collected "in the field" and prospective estimates of exposure (there is virtually no consideration of historical exposures), in the most prevalent forms of litigation involving human health issues (e.g., toxic tort cases) there is often little or no information available regarding the degree of chemical exposure that the plaintiff may have experienced. Specifically, there are rarely any useful measurements of chemical concentrations in the home, environment or workplace where the exposures are alleged to have occurred. Indeed, the alleged chromium exposures in Anderson occurred decades ago, and there was very little reliable groundwater data from that time period.

In such instances where quantitative information is lacking, many experts rely on published literature values (particularly in occupational settings for which hygiene data exist) or mathematical models to estimate chemical concentrations. However, the degree of uncertainty associated with these approaches can be difficult to quantify, much less convey to a judge and jury. Exposure assessment experts are turning increasingly towards the use of "exposure simulation" or "dose reconstruction" techniques in which the plaintiff's exposure setting is recreated and quantitative exposure information is collected. The "information" may be chemical concentrations in exposure media such food, air, soil, water, etc, and/or actual levels of chemicals or metabolites in biological tissues. This approach has two distinct advantages: 1) it can provide a "bound" or ceiling on the chemical dose that the plaintiffs could have plausibly experienced and therefore provides an effective rebuttal to unreasonable claims of exposure, and 2) it contains far less uncertainty and usually has more courtroom appeal than classical modeling or other estimation techniques.

In the past two years, data from exposure reconstruction studies conducted strictly for litigation purposes have successfully been used to settle or try numerous cases involving consumer products, soil and/or groundwater contamination, and occupational exposures to airborne chemicals. Two "case studies" which describe research conducted as part of litigation support and which are applicable to a wide range of toxic tort claims are summarized below.

Railroad Workers and Diesel Exhaust Exposure

Over the past five years, current and former employees of various railroad companies have filed an increasing number of claims based on alleged harm to health as a result of exposure to diesel exhaust. Many of these claims involve specific types of cancers that have been reported to be associated with diesel exhaust exposure in published epidemiological studies. It should be noted that although the conclusions of some epidemiological studies might suggest a possible association between diesel exhaust and certain cancers, the weight of evidence regarding causation and any particular type of cancer is far from compelling.

A number of cases have specifically identified benzene as the purported cancer-causing agent. Diesel exhaust consists of a complex mixture of chemicals, however, and while benzene may appear to be an obvious target in a claim involving cancer and diesel exposure, in fact there is very little known regarding benzene levels in diesel exhaust, particularly in occupational exposure settings. Accordingly, the exposure/dose simulation study described below was designed to generate the data necessary to quantify and characterize benzene exposures in a variety of occupational conditions relevant to the railroad industry.

Methods

Claims involving diesel exhaust exposure in the railroad industry involve a wide variety of job titles, including engineers, rail-yard workers, and maintenance personnel. Arguably, one of the occupations with the highest degree of potential exposure involves the workers who maintain and repair the diesel engines inside an enclosed shop or "roundhouse." The diesel engines often run during maintenance activities and, given the "indoor" nature of the setting, it is reasonable to expect that these conditions represent one of the "upper bound" levels of exposure in the railroad industry.

In order to assess the range of plausible benzene exposures experienced under roundhouse conditions, locomotive diesel exhaust emissions were generated in a roundhouse under simulated worst-case exposure conditions. A locomotive was run for four 30-minute intervals during an 8-hour work shift; full-shift and 1-hour airborne concentrations of benzene were measured in the breathing zone over the work shift between and during the emission episodes on two consecutive days. Day 1 simulated a "winter condition" with the roundhouse doors closed; Day 2 simulated a "summer condition" with the doors open. Background benzene levels were also collected several miles away from any appreciable source of airborne chemicals. Benzene levels were measured using NIOSH Method 1501 in conjunction with sorbent tubes (1-hour samples) and passive sampling badges (full shift).

Results

The results can be summarized as follows:

• the 8-hour time-weighted average (TWA) was 0.0049 ppm on Day 1 and 0.0039 ppm on Day 2
  
• the overall TWA assuming "winter conditions" for approximately 4 months per year and "summer conditions" for the remainder of the year was 0.0046 ppm
  
• with the exception of one concentration at 0.00048 ppm, benzene concentrations in all background samples were below the limit of detection

The lifetime benzene dose associated with 5 years of exposure in this setting (40-hour work weeks) would be approximately 0.19 grams.

Dose comparison

In order to characterize roundhouse benzene exposures, plausible benzene doses associated with different occupational and environmental benzene sources were derived from literature sources. For example:

• the lifetime benzene dose associated with occupational exposure to the Occupational Safety and Health Administration (OSHA) permissible exposure limit (PEL) of 1 ppm benzene for 8 hours per day over 40 years is 340 grams
  
• the lifetime benzene dose associated with filling vehicle gasoline tanks at service stations is 0.55 grams (assuming an average airborne concentration of 1 ppm and exposures of 5 minutes per week for 50 years)
  
• the lifetime benzene dose associated with cigarette smoking is 33 grams, (assuming an average benzene content of 57 micrograms of benzene per cigarette, an average smoking frequency of 32 cigarettes per day over a period of 50 years, and nine puffs [35 ml per puff] per cigarette)
• the lifetime benzene dose to a resident living in Los Angeles for an entire lifetime (70 years) is 22 grams (assuming an average ambient benzene concentration of 0.015 ppm)

The histogram (Figure 1) compares these estimated lifetime benzene doses to those associated with 5 years of occupational exposure in a roundhouse (as measured under the conditions of the above study). The exposure duration (5 years) is consistent with that of a recent case involving a plaintiff with acute myeloid leukemia (Hirsch v. Union Pacific Railroad). These comparisons make it clear, in a manner easily understood by the layman, that the benzene doses permitted under the occupational standards (i.e., the PEL) and the benzene doses associated with "other" sources will likely be much greater than those associated with typical roundhouse exposures.

Figure 1: Comparison of lifetime benzene dose (g) due to diesel locomotive exhaust in roundhouse to the OSHA PEL and selected environmental exposures.

Conclusions

If roundhouse exposures are truly representative of a "worst-case" setting in the railroad industry, then the results of this comparison suggest that it should be difficult to demonstrate (for a majority of plaintiffs) that benzene exposures during employment in the railroad industry contributed substantially to cancer (especially for smokers). Obviously, the proof problems are even greater if a clear foundation for disease causation is lacking.

Residential Exposure to Cr(VI) in Tapwater

Toxic tort claims involving current or historical exposure to hexavalent chromium [(Cr(VI)] in tapwater are increasingly common due to the notoriety of the "Erin Brockovich" movie and the aforementioned settlement in the associated case. Direct ingestion is the primary route of exposure to Cr(VI) in potable tapwater, and it is well known that the acidic secretions of the stomach quickly reduce Cr(VI) to the non-toxic Cr(III) (trivalent chromium) under laboratory in vitro conditions. However, in order to assess whether this reductive capacity is sufficient to protect individuals against systemic absorption of Cr(VI) in the event of Cr(VI) ingestion, it would clearly be ideal to have in vivo data in humans. From a scientific perspective, this information would reduce uncertainty; from a litigation perspective, it is not only more relevant and compelling but also easier to interpret and convey to a judge and jury.

Obviously, the amount of alleged exposure will vary case to case. However, as in the diesel exhaust case study, it is possible to construct a "worst case" scenario that should be applicable in most if not all cases. Cr(VI) in water imparts a distinct yellowish tinge at concentrations of approximately 2 ppm and higher. The color is clearly obvious at 5 ppm and higher, and therefore it is reasonable to conclude that chronic or even acute exposures to concentrations of 5 ppm and above are implausible. As described below, human exposure studies have been conducted to determine whether ingestion of tapwater Cr(VI) concentrations ranging from 0.1-10 ppm could result in systemic absorption of Cr(VI).

Methods

When Cr(VI) is absorbed into the bloodstream, it enters the red blood cells (RBCs), binds to intracellular material, and remains there for the lifetime of the cell (up to 90 days). Cr(III) can enter the RBC but does not bind to the cell, and is therefore cleared from the bloodstream much more quickly than Cr(VI). Hence, a sustained plateau of chromium in RBCs is indicative of absorbed Cr(VI), while a transient "spike" in RBC chromium is suggestive of Cr(III) absorption. In this study, 5 healthy male Caucasian volunteers consumed Cr(VI) in water under a variety of dosing regimens (single day, multiple day, single doses per day, multiple doses per day, etc.) in order to simulate the range of ingested doses that might be associated with tapwater ingestion. RBC and plasma samples were collected and measured for chromium content before, during, and after the dosing periods.

Results

Figure 2 illustrates the RBC and plasma chromium concentrations measured in 3 individuals who consumed 1 liter of water containing 5 ppm Cr(VI) for 3 days (first dosing period) followed by a 3 day dosing of 1 liter per day of 10 ppm Cr(VI). These results, which are consistent with the results obtained with all other dosing regimens, demonstrate a transient "spike" in RBC chromium, followed by an approximate 50% decrease within 7 days of post-dosing. The apparent equilibrium between the plasma and RBC compartments is further evidence that the chromium was absorbed in the trivalent state. Clearly, these results may "resonate" in a courtroom in a manner that could be difficult to achieve with a complex extrapolation from in vitro data.

Figure 2: Plasma RBCs of volunteers in concentrations of chromium following oral ingestion of 1 L/d of water containing either 5 or 10 mg Cr(VI). At each concentration, the volunteers drank a total of 1 liter of water per day for 3 consecutive days.

These results indicate that, even upon bolus ingestion of large amounts of obviously colored water, ingested Cr(VI) is reduced to Cr(III) in the stomach. Given that this particular dosing regimen represents a "worst case" scenario, this clearly has relevance to any claims of health harm associated with Cr(VI) ingestion. Specifically, it is highly unlikely that any individual could systemically absorb Cr(VI) as a result of tapwater exposure, regardless of the dose.

Conclusion

Per the current rules regarding the admissibility of scientific evidence, the plaintiff is expected to quantitate chemical exposure and demonstrate that the degree of exposure was sufficient to cause the alleged health effect. As a result, chemical exposure/dose reconstruction is becoming a standard technique to evaluate plaintiff claims in toxic tort litigation. Such analyses are of most use when there is significant uncertainty regarding the chemical concentrations that may have existed during plaintiffs' exposure. In some cases, such as those described here, it is possible to design a reconstruction analysis that is applicable to a wide variety of possible claims. The use of this technique in the courtroom will likely continue to gain increasing acceptance as a means for assessing the merits of plaintiff claims.

Brent Finley is the director of the human health practice at Exponent (www.exponent.com), an engineering and scientific consulting firm.