Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
5 Process for Establishing and Applying Military Exposure Guidelines The health-protective nature of the current military exposure guidelines (MEGs) makes them most appropriate for use as part of the Armyâs force health protection initiative. In this chapter, the subcommittee reviews how MEGs were derived in Reference Document 230 (RD-230) (USACHPPM 2002) and provides comments and recommendations on their application. In addition, the subcommittee considers the need to address risks from multiple exposure pathways, multiple chemicals, and repeated deployments. AIR EXPOSURE GUIDELINES This section provides a summary of the U.S. Army Center for Health Promotion and Preventive Medicineâs (USACHPPMâs) approaches to de- riving its current air MEGs. The basic approach to determining air MEGs was to review the exposure guidelines of other agencies, use a hierarchical scheme to select the most appropriate guideline for the exposure duration of interest, and then adjust that guideline to meet the militaryâs needs, if necessary. Air MEGs were developed for exposures of 1 hour, 8 hours, 14 days, and 1 year. In the sections below, the process used to derive the duration-specific MEGs is described and evaluated, followed by a review of some chemical-specific MEGs and criteria air pollutants. 91
92 TECHNICAL GUIDES ON ASSESSING AND MANAGING CHEMICAL HAZARDS 1-Hour MEGs Derivation One-hour air MEGs were developed to consider three levels of health effects: â¢ Minimal effects. Above this level, individuals could begin to experience mild, transient effects that should not impair performance. â¢ Significant effects. Above this level, individuals could begin to experience irreversible or serious effects that might degrade performance and incapacitate a small portion of the people exposed. â¢ Severe effects. Above this level, some within an exposed popula- tion could begin to experience life-threatening or lethal effects. The hierarchy used to select source material was (1) acute exposure guide- line levels (AEGLs), (2) emergency response planning guidelines (ERPGs), (3) temporary emergency exposure limits (TEELs), and (4) other. AEGLs are developed by the U.S. Environmental Protection Agency (EPA) and reviewed by a National Advisory Committee and by the National Research Council (NRC). AEGLs are developed for three severity levels, and all of the values are intended to protect the general public, including sensitive and susceptible subpopulations. Above AEGL-1 concentrations, the general population could experience discomfort and irritation effects that are not disabling and are reversible upon cessation of exposure. Above AEGL-2 concentrations, the general population could experience irrevers- ible or serious health effects or impaired ability to escape. Above AEGL-3, the general population could experience life-threatening health effects or death. These levels were developed for exposure durations of 10 minutes (min), 30 min, 1 hour, 4 hours, and 8 hours. All AEGL-1 and AEGL-2 exposures were reviewed by EPA to ensure that they do not pose an excess cancer risk greater than 1 Ã 10-4. ERPGs are developed by the American Industrial Hygiene Association (AIHA) and are intended for emergency planning and response operations. They also have three levels of health effects that are quite similar to those of the AEGLs. They were created to target the general population, but not particularly susceptible individuals. TEELs are developed by the U.S. Department of Energy and are essentially interim ERPGs.
MILITARY EXPOSURE GUIDELINES 93 Evaluation The hierarchy for selecting sources for the 1-hour MEGs is based on a logical argument and is consistent with the NRC (2000) recommendations for developing standards. The NRC (2000) noted that in the development of guidelines, different kinds of guidelines are appropriate for different settings. The report also stated that it is useful to allow for guidelines that permit some degree of toxic response but protect against incapacitation or irreversible injury for use in decision making during emergencies or when important risk trade-off decisions must be made quickly, such as in combat. The quality of the 1-hour MEGs is limited to the quality of the source assessments. Those assessments are, in turn, limited by the quality of the database and how recently the assessments were performed. AEGLs can be assumed to be of higher quality because they were developed recently, had more data to consider, and are extensively peer-reviewed. However, they are few in number. That makes it important for TG-230 and RD-230 to be âlivingâ documents that incorporate new values as soon as they become available. For example, RD-230 mentions that interim and proposed AEGLs were used. It is necessary to track the final versions and to ensure that the final AEGL values are incorporated into the documents. As chemicals are selected for AEGLs development, USACHPPM should give priority consid- eration to those chemicals likely to be found in major theaters of operations. The 1-hour air MEGs are based on different sources. To facilitate making updates, it would be useful to document the existing guideline used and the date it was established in the supporting reference tables. In addi- tion, the entries in air MEG tables provided in TG-230 should be reviewed to ensure that they describe specific end points of interest. Currently, some entries in the tables provide only a broad description of the critical-study end points (e.g., systemic, irritation). Sometimes the entries do not describe the end points at all (e.g., âbased on a slightly higher incidence of nasal tumors in rats,â âbased on extrapolation of acute animal data and limited evidence in humansâ). 8-Hour Air MEGs Derivation RD-230 indicates that the exposure duration of 8 hours was selected to
94 TECHNICAL GUIDES ON ASSESSING AND MANAGING CHEMICAL HAZARDS be consistent with brief exposures. The corresponding MEGs represent levels below which no significant adverse health effects are expected and above which the probability of adverse health effects is increased. The 8- hour MEGs incorporate the assumption that exposures will be continuous. The hierarchy used to select sources to establish the 8-hour MEGs was (1) AEGL-1 values and (2) Threshold Limit Values (TLVs), which are devel- oped by the American Conference of Governmental Industrial Hygienists (ACGIH). The TLVs are developed to protect workers against the effects of a working lifetime of exposure (8 hours/day, 5 days/week, and 50 weeks/year for a working lifetime). RD-230 used the TLVs for 8-hour exposures when no AEGLs were available. In a number of cases, the 8- hour air MEGs are the same as the 1-hour air MEGs for minimal effects. Evaluation Because relatively few 8-hour AEGLs have been derived, TLVs are the preferred sources for the 8-hour MEGs. ACGIH criticizes the direct use of TLVs for purposes other than those intended. However, feasible alterna- tives will not exist until more AEGLs are established. TLVs are concentra- tions expected to be relatively safe for worker populations exposed intermit- tently (8-hour workdays) for a working lifetime. Thus, their direct applica- tion to a single 8-hour period is likely to be protective. Whether that ap- proach is overly protective depends on the database and the calculations that were used to set the particular TLV. TLVs do not use standardized formu- las, so it would be difficult to determine the likely margins of conservatism that were used to establish them. Determining what modifications are nec- essary to create an 8-hour MEG for continuous exposure from a TLV de- signed to be protective for intermittent exposure over a working lifetime is even more difficult. But, until revisions can be made, this approach is the most feasible, however overprotective. It appears that no consideration was given to making adjustments for the higher inhalation rate of deployed personnel; it might be appropriate to make those adjustments for 8-hour exposures. In the future, it would be useful to examine the data underlying the values selected for MEGs to determine whether the original data in- cluded exposure durations closer to 8 hours and therefore might be more appropriate than the calculations used by other agencies for other purposes.
MILITARY EXPOSURE GUIDELINES 95 14-Day MEGs Derivation The 14-day air MEGs incorporate the assumption that exposures will be continuous, recognizing the limited likelihood of that in the real world and that simplifications are essential to create workable guidelines. The hierarchy for selecting the sources of the 14-day MEGs was (1) continuous exposure guidance levels (CEGLs), (2) minimal risk levels (MRLs), (3) TLVs, and (4) special considerations. CEGLs were developed by the NRC (1986) for exposures to military personnel lasting up to 90 days. They are intended to prevent serious or permanent effects in a healthy male popula- tion and do not include consideration of susceptible subpopulations. The 14-day MEGs consider the possibility that increasing concentration or duration could increase the potential for delayed or permanent disease (e.g., kidney disease or cancer). MRLs are developed by the Agency for Toxic Substances and Disease Registry (ATSDR) for noncancer effects. An MRL is an estimate of the daily human exposure to a hazardous substance that is likely to be without appreciable risk of adverse health effects over a specified duration of expo- sure. These estimates are intended to serve as screening levels to identify contaminants and potential health effects that might be of concern at haz- ardous waste sites. ATSDR creates MRLs for acute (1-14 days), intermedi- ate (14-364 days), and chronic (365 days or longer) exposures. Although RD-230 states that the TLVs were not considered protective for continuous exposures of over 24 hours to 14 days, the TLVs were ex- trapolated down from working lifetime values to 14-day continuous expo- sure values. TLVs for âsystemicâ or âmixed-actingâ substances were ad- justed by a factor of 5 days/7 days, a ventilation factor of 10 m3/20 m3 (10m3 is the worker 8-hour default factor) with another calculation of 20 m3/29.2 m3 to account for the military personâs increased ventilation rate (see Chapter 3) (equaling 10 m3/29.2 m3) , and an uncertainty factor of 10 to account for the uncertainty of extrapolation from intermittent to continu- ous exposure (see Equation 5-1). 14-day MEG = (TLV Ã 5 days/7 days Ã 10 m3/29.2 m3) Ã 0.1 (5-1) The TLVs for irritants were not adjusted because they are assumed to be mostly concentration-dependent.
96 TECHNICAL GUIDES ON ASSESSING AND MANAGING CHEMICAL HAZARDS USACHPPM determined that more than 24 hours of continuous expo- sure to chemical warfare agents (CWAs) is unlikely. Therefore, no MEGs were established for CWAs for periods greater than 1 day. Twenty-four- hour MEGs for CWAs were derived by linear extrapolation from the 8-hour MEGs. Evaluation The CEGLs were derived assuming 90 days of continuous exposure, so it is likely that they are conservative. Because many were published in the late 1980s, some of them could be out-of-date. Furthermore, CEGLs were developed for use by the Navy on submarines and, therefore, the target population was assumed to be exclusively male. Thus, female reproductive end points and developmental toxicity were not considered in setting the CEGLs. The acute MRLs were calculated on the basis of exposure durations of 1-14 days. Thus, they are reasonably targeted for duration; however, they include UFs for susceptible groups. The MRL-based calculations for MEGs do not appear to include adjustments for military ventilation rates. USACHPPM should make those adjustments. For the 14-day values based on TLVs, adjusting the TLVs for the change from 8 hour/day, 5 day/week to 24 hour/day, 7 day/week and for the higher breathing rates of military personnel (i.e., 14-day air MEGs = TLVs Ã 5 days/7 days Ã 10 m3/29.2 m3) is reasonable for systemic chemicals when dose rate is not the determining factor and only total dose dictates effects (Gaylor 2000). The calculations assume that a C (concentration) Ã t (time) = k (total exposure) relationship holds for systemic effects. The basis for the 29.2 m3/day ventilation rate is reasonable, although it contains several assumptions. Because TLVs are intended to be protective over a workerâs lifetime, extrapolating 14-day continuous exposures from 8-hour TLVs introduces significant uncertainty. A UF of 10 was used to extrapolate from intermit- tent to continuous exposure. EPA and ATSDR make similar extrapolations for RfCs and MRLs, respectively, but do not use a UF. A weak justification is offered in RD-230, which says that some health effects have been ob- served in some workers at the TLV levels, without further specification. However, the UF of 10 is unduly conservative. Typically, concentration is more important than duration in the C Ã t equation (beyond acute lethality). When a guideline for intermittent exposure is converted to one for continu- ous exposure, it becomes more conservative. For example, consider an
MILITARY EXPOSURE GUIDELINES 97 intermittent TLV of 0.5 mg/m3 for 8 hour/day, 5 days/week for 260 days/ year, or 2,080 hours of exposure. That is equivalent to a total exposure (k) of 1,040 mg-h/m3. When that value is converted to a continuous exposure (24 hours/day for 365 days, or 6,360 hours), the comparable C would be 0.16 mg/m3 (i.e., C = k Ã· t). In other words, if C Ã t = k operates, an inter- mittent inhalation of 0.5 mg/m3 is equivalent to a continuous inhalation of 0.16 mg/m3. Therefore, the 0.16 mg/m3 has built-in conservatism that is appropriate. Applying an additional UF of 10 would result in a guideline of 0.016 mg/m3, which is overly conservative. The 14-day air MEGs are difficult to develop because most of the source materials have different exposure durations and use different assess- ment methodologies. Table 5-1 summarizes the sources and highlights the differences in the portions of lifetime protected and the adjustments that were made to the source material. It would be advisable to check the refer- ence data of the sources to determine to what degree the databases were founded on studies approximating the duration of military interest. If ap- propriate, other data might be used to derive MEGs. Also, it would be best to use a standard approach to applying adjustments across all values. For example, adjustments for military ventilation rates should be used in all the MEGs. 1-Year Air MEGs Derivation The one-year air MEG is defined by USACHPPM (2002) as âThe air- borne concentration for a continuous exposure up to 1 year (365 days, 24 TABLE 5-1 Sources for 14-day Air MEGs Exposure Source Duration and Portion of Lifetime USACHPPM Adjustments in Reference Frequency Protected 14-Day Air MEGs CEGL 90 days, 90 days No adjustments were made continuous MRL 1-14 days, daily 1-14 days No duration or ventilation rate adjustments were made TLVs 8 hours/day, 5 Working lifetime Adjustments to continuous days/ week 14-day and military ventilation rates
98 TECHNICAL GUIDES ON ASSESSING AND MANAGING CHEMICAL HAZARDS hours/day) that is considered protective against all health effects including chronic disease and increased risk to cancer (i.e., cancer risk greater than 1 Ã 10-4). No performance degradation or long-term health consequences are expected with exposure at or below this level. Increasing concentration and/or duration could increase the potential for delayed/permanent disease (e.g., kidney disease or cancer).â The 1-year MEGs were not designed to address continuous exposure exceeding 1 year. Inhalation reference concentrations (RfCs) for noncarcinogenic effects, air unit risks, or inhalation cancer slope factors (CSFi) from the EPAâs Integrated Risk Information System (IRIS) and the Health Effects Assess- ment Summary Tables (HEAST) were selected to derive preliminary long- term MEGs (PMEG-L). When those EPA sources were not available, addi- tional sources, including TLVs and MRLs, were used with additional ad- justments (see discussion below). For carcinogenic polycyclic aromatic hydrocarbons (PAHs), provisional EPA values were used; they included toxicity equivalence factors relative to benzo(a)pyrene. The air-MEG selec- tion was based on the following hierarchy: (1) PMEG-L, (2) TLV-adjusted, and (3) MRL-adjusted. If significant (more than an order of magnitude) discrepancies between those values existed, USACHPPM reviewed the data and selected the final 1-year air MEGs. Derivation of PMEG-Ls USACHPPM developed military ânoncancerâ risk concentrations (MRCs) and cancer risk concentrations (MCRCs) using a method similar to that used in the derivation of EPAâs Region III risk-based concentration values, which are consistent with risk-assessment guidance for Superfund. The cancer and noncancer values were compared, and the lower one (i.e., the more protective one) was identified as the PMEG-L. Because a 1-year exposure duration is of interest for noncancer risks, the first-choice sources were the subchronic RfCs in HEAST; if those were not available, chronic values were used. Subchronic is defined as one-tenth of the average lifespan, or 2 weeks to 7 years, and chronic is defined as more than 7 years. To derive MRCs, RfCs (in units of milligrams per cubic meter [mg/m3]) were converted to reference doses (RfDs, in milligrams per kilogram per day [mg/kg/day]) by multiplying the inhalation rate of 20 m3/day and dividing by 70 kg, the average weight for adults. With the target hazard quotient (THQ) set at 1, a backward calculation was per- formed to derive the MRCs using the following assumptions: (1) body weight (BW) of 70 kg, (2) military inhalation rate (IRA) of 29.2 m3/day, (3)
MILITARY EXPOSURE GUIDELINES 99 exposure duration (ED) of 1 year, (4) exposure frequency (EF) of 365 days/year, and (5) average time (AT) of 365 days (see Equation 5-2). THQ Ã RfD Ã BW Ã AT MRC = (5-2) EF Ã ED Ã IRA EPAâs CSFs were used as a basis for deriving the MCRCs. Those unit cancer risks also were converted from risk per microgram per cubic meter to risk per milligram per kilogram per day, assuming a body weight of 70 kg and an inhalation rate of 20 m3/day. To calculate the MCRC, the target cancer risk (TCR) was set at 1 Ã 10-4 and the following assumptions were made: (1) BW = 70 kg, (2) IRA = 29.2 m3/day, (3) ED = 1 year, (4) EF = 365 days/year, and (5) AT = 25,550 days (70 years Ã 365 days) (see Equa- tion 5-3). TCR Ã BW Ã AT MCRC = (5-3) EF Ã ED Ã IRA Ã CSF Adjustments of TLVs and MRLs When TLVs were used as the sources for the 1-year MEGs, they were adjusted to account for the military personâs assumed respiratory rate and for uncertainties associated with extrapolating values for intermittent expo- sures to continuous exposures. For extrapolation, a UF of 10 was applied. However, the TLVs for irritants were not duration-adjusted. Intermediate MRLs (15-364 days) were given preference over chronic MRLs. The MRLs were adjusted to account for the assumed military inha- lation rate. Evaluation Quality of the Source References EPA, ACGIH, and ATSDR sources are appropriate. IRIS is the only fully official set of EPA assessments. HEAST includes some values that have not been agreed on, some that have not been peer-reviewed, and some that have been removed from IRIS because of quality problems. Also, the
100 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS use of provisional values for PAHs that are close to 10 years old suggests that some significant uncertainties in the available data have not been ad- dressed. Furthermore, many of the current IRIS values, TLVs, and MRLs are out-of-date, and some of them are obsolete because of newer informa- tion. Although it is not feasible for DOD to revise all the source data, po- tential problems associated with using those sources should be recognized and stated. EPA Region IX values were used as sources for many of the 1-year MEGs; however, those EPA values were created using a complex process and they have little apparent worth for the MEGs. The rationale offered by USACHPPM for the MRC and MCRC adjustments from milligrams per cubic meter to milligrams per kilogram per day appears to be numerically driven, and ultimately the additional conversion factors cancel themselves out. That conversion, however, was not applied to MRLs and TLVs. In developing the RfC and cancer unit risks, EPA made decisions to use milli- grams per cubic meter or micrograms per cubic meter as the units for the durations of interest (typically continuous exposure for 70 years). Those units are used in the underlying research studies and are the units that would eventually be used in regulations. Although cubic meters of air breathed per day has a relationship to body weight, the convention of measuring exposure in milligrams per cubic meter is more accurate than the milligrams per kilogram per day. Conversion from an RfC or a cancer unit risk to milligrams per kilogram per day at the level of the individual studies would introduce unnecessary uncertainties. The subcommittee recommends that USACHPPM consult additional sources of guideline values, such as the World Health Organization (WHO 2001) and the State of California. WHO (2001) has health-based guidelines for 35 air pollutants, and most were derived using expert judgment and include consideration of susceptible populations. The State of California has hundreds of values that were derived following a standard procedure similar, but not identical to that used by EPA. The subcommittee recommends that HEAST values not be used to derive long-term air MEGs, because the quality of those assessments is not as strong as that of the other guidelines. For the other sources, the date of the original assessment should be provided in the tables in RD-230 to indi- cate the degree of potential obsolescence of the source material. Inhalation Adjustment Factor Most of the starting values were adjusted from EPA, ACGIH, or MRL
MILITARY EXPOSURE GUIDELINES 101 ventilation defaults of 20 m3/day or 10 m3/day to the military ventilation rate of 29.2 m3/day. The military default rate is based on a series of mea- surements, scenario estimations, and judgments. No default rate is or will ever be perfect, so the rate should be judged relative to its purpose of pro- viding an appropriate level of protection for the population of concern. USACHPPM evaluated all the components of the military inhalation rate assumption, and on the basis of limited information for the types of activity likely to be performed, concluded that it is reasonable (see Chapter 3). However, the RfC methodology used to derive several of the MEGs in- cludes a dosimetric extrapolation from animals to humans that considers ventilation rate. The implications of that are likely to be greater for reactive gases and some particles. The dosimetric model is based on a ventilation rate of 20 m3/day, and a rate of 29.2 m3/day would alter the pattern of respi- ratory tract deposition. RfCs are based on regional deposited dose when that method is supported by the data. Thus, ventilation can influence the RfC, depending on the specific circumstances. For example, a reactive gas with an assessment based on nasopharyngeal doses might be impacted. That possible impact is unlikely to have a major influence on the MEGs, but it bears consideration if there is an attempt to go back to the original data and recalculate the MEGs. Applying a UF of 10 to the TLVs Extrapolating the TLVs from intermittent to continuous exposures is acceptable for nonirritants because incorporating the area under the expo- sure curve is scientifically appropriate and is routine practice in most as- sessments (e.g., RfC, MRLs). However, applying a UF of 10 is not support- able, as discussed earlier (see â14-Day Air MEGsâ). The MEGs should therefore be revised. Varying Exposure Durations One overarching problem is that the 1-year air MEGs are based on source references with varying exposure durations. Table 5-2 summarizes the exposure durations and frequencies associated with the sources used by USACHPPM in developing the 1-year air MEGs. USACHPPM mentioned that they used subchronic RfCs when they are available. However, the chronic RfCs and unit cancer risks, which are both based on lifetime exposure, were also used. For carcinogenic agents, the MCRCs are derived by averaging the 70-year cumulative lifetime dose limit
102 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS TABLE 5-2 Exposure Durations in Source References Source References Duration Frequency PMEG Noncancer (RfCs) 1/7 of lifetime or Daily lifetime Cancer (unit cancer risks) Lifetime Daily MRL Noncancer 15-364 days Daily TLV Noncancer and cancer Working lifetime 8-hour day, 40-hour workweek (given a unit cancer risk and lifetime risk of 1 Ã 10-4) over a 1-year exposure duration. That derivation is appropriate and is protective for exposure dura- tions up to 1 year. One problem with using the chronic RfCs as starting points for 1-year values is that in some cases those RfCs are based on subchronic effects (e.g., a 90-day exposure study of laboratory rats), and a UF of 10 was ap- plied in the original derivation to extrapolate from subchronic to chronic exposure. Because subchronic exposures are of direct interest to MEGs, there is no need for that particular UF. For example, the IRIS RfC for acetaldehyde has a composite UF of 1,000, including a factor of 10 for subchronic to chronic extrapolation. Thus, that starting point would be over-protective for a 1-year exposure scenario. In other cases, the original assessment, following the established meth- odology, would have used lifetime exposure studies, if available and of quality, to derive an RfC, and would only have used subchronic studies to enhance understanding of the chemical. If the goal was a 1-year exposure guideline, high-quality subchronic studies likely would be used for a deriva- tion. Some RfCs might rely on robust chronic studies when subchronic data are inadequate. Those RfCs would be used appropriately as input for a 1- year MEG. An added complication is that the assessment methods of other agencies are changing. For example, benchmark dose methods are becom- ing more common. Thus, approaches that were adequate for the underlying methods might need to be revisited when USACHPPM revises the MEGs. For 1-year air MEGs that are based on adjusted TLVs, there is an addi- tional interpretative issue, because the TLVs address worker-lifetime expo- sure. Therefore, the adjusted TLVs would be protective for exposure dura- tions much longer than 1 year. Although that added level of protectiveness might be appropriate in light of the potential for repeated and multiple deployments, the lack of consistency in adjusting source reference exposure
MILITARY EXPOSURE GUIDELINES 103 durations to fit the MEGs poses interpretation challenges. Efforts should be made to increase consistency in future revisions of MEGs. Specific Chemical MEGs A few 8-hour, 14-day, and 1-year air MEGs were selected outside of the established hierarchy. The MEGs for benzene and toluene were based on the TLV-adjusted rather than the MRLs and PMEGs. The ethyl benzene MEG was based on the MRL-adjusted rather than the PMEG. All of those decisions were made to avoid relying on conservative UFs. Related issues arose for styrene, n-hexane, and xylene. Several PAHs were considered, but inhalation toxicity data were lacking for seven of them. Oral RfD data were extrapolated, and quantitative structure-activity relationship methods were used to derive inhalation values for those seven PAHs. The oral-to- inhalation extrapolation for PAHs is fraught with uncertainty, but there are no reasonable alternatives. The MEG for naphthalene was based on the MRL-adjusted because that source addressed subpopulations with glucose-6-phosphate dehydrogenase (G-6-PD) deficiencies that did not appear to be addressed by the TLV. That subpopulation could be more vulnerable to oxidant exposures. There are two issues: (1) do people with G-6-PD deficiency truly need more protec- tion (they would in theory, but how good is the evidence for likely impact), and (2) if they do, protection against more than naphthalene (e.g., other oxidants) might be indicated. USACHPPM needs to analyze that further. Ambient Air Quality Criteria Pollutants EPA health-based national ambient air quality standards (NAAQS) exist for six pollutants: ozone (O3), nitrogen dioxide (NO2), sulfur dioxide (SO2), lead, particulate matter (PM10 and PM2.5), and carbon monoxide (CO). These pollutants are important because they are ubiquitous and capa- ble of causing adverse health effects in susceptible individuals at high ambi- ent levels. The NAAQS have a variety of durations (some are for 24 hours, some are annual) and are set to protect susceptible subpopulations. Each of the standards also has a descriptive category of air quality (the pollutant standard index [PSI]) developed by EPA to provide precautionary summa- ries about health effects to nontechnical audiences.
104 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS One-year MEGs were developed for the six criteria air pollutants. Annual mean, quarterly averages, and 24-hour NAAQS were considered when available. Linear extrapolation was used for substances that only had 8-hour average standards. TLVs also exist for these criteria pollutants, so the TLV-adjusted was compared with the NAAQS. The rationale for choosing one versus the other was not described. NAAQS are based on extensive databases and are routinely updated by EPA. It would be relatively easy to re-evaluate these databases with an eye towards identifying subpopulations that might be represented in the de- ployed population, thereby getting a more accurate indication of potential risks from these ubiquitous pollutants. EPA has several experts on criteria pollutants who could be approached for assistance. The rationales for choosing the long-term MEGs for the criteria pollut- ants need further discussion. For example, why was the NAAQS for NO2 (0.1 mg/m3) chosen over the TLV-adjusted (0.14 mg/m3) when the NAAQS is set to protect children and the TLV is set to protect adults? For O3, the MEG is 0.052 mg/m3, the 8-hour NAAQS is 0.157 mg/m3, and the TLV- adjusted is 0.004 mg/m3. Background levels for O3 in many areas of the world are above the long-term MEG. Thus, the MEG would likely be ex- ceeded frequently in those areas that have a lot of sunlight and have precur- sors (NOx and VOCs). This issue is complicated by the fact that low levels of O3 (as low as 0.12 mg/m3 under heavy exercise conditions for about a half hour [EPA 1996a]) can affect the pulmonary function and exercise performance of young healthy athletic adults. It would be advisable to re- evaluate the MEGs for O3 to ensure that they appropriately protect the forces. The SO2 annual NAAQS is 0.365 mg/m3, which includes consider- ation of susceptible subpopulations, but the 1-year MEG is 0.13 mg/m3 and is based on an adjusted TLV. The unadjusted TLV for SO2 is 5.24 mg/m3. From a mathematical and procedural viewpoint, the rationale for that choice is clear. However, selecting a military value lower than the NAAQS should be re-evaluated, especially considering the more robust health of the mili- tary population. As mentioned above, the most straightforward approach to dealing with the criteria air pollutants might be to consult with EPA and address the military population explicitly. The subcommittee recommends that the MEGs for criteria pollutants be reconsidered relative to their applicability to deployed forces. This will require evaluating the chemical database for data relevant to deployed forces and deriving MEGs from those data. The following comments show why certain NAAQS are inappropriate bases for the MEGs (Graham et al. 1999). Some specific comments are offered in Appendix A.
MILITARY EXPOSURE GUIDELINES 105 â¢ As mentioned above, the NAAQS are set to protect susceptible subpopulations. In the case of O3, the susceptible populations include healthy young adults exercising outdoors, which makes the NAAQS rele- vant to the troops. In contrast, other NAAQS are not directly relevant to deployed forces, even assuming some degree of health impairments among the military population (e.g., mild asthma). â¢ The NAAQS for CO are set to protect angina patients (or others with forms of coronary artery disease) who exercise; therefore, they are not applicable to deployed troops. The goal of the current CO standards is to maintain blood carboxyhemoglobin (COHb), a biomarker of effects, below 2.1%. Average COHb levels in smokers who smoke 1-2 packs per day are about 4%. CO can have effects on healthy individuals and fetuses, but at higher levels than the NAAQS. Those effects include reduced maximal exercise duration and central nervous system effects (e.g., decrements in hand-eye coordination) that could be of concern during deployments. â¢ The NO2 annual-average NAAQS is based on effects on 5- to 12- year-old children. EPA does not have a short-term NAAQS for NO2 be- cause if the annual NAAQS is met, there is almost no likelihood of short- term concentrations rising to levels of concern for other healthy or suscepti- ble populations. However, that analysis is based on ambient air quality patterns in the United States. The potential exists that other countries with different air quality patterns might have short-term excursions that would affect asthmatic individuals. WHO (2002) set a short-term air quality guideline that might be valuable in this instance. â¢ The NAAQS for SO2 (24-hour and annual average) are set to pro- tect children and people with pre-existing lung disease (especially asthma). Short-term (e.g., 10-min) exposures in exercising asthmatic individuals could cause bronchoconstriction. However, a significant prevalence of asthma among deployed troops is unlikely. â¢ The NAAQS for PM are set to avoid increased mortality, predomi- nantly from cardiovascular and respiratory causes, although children are also at risk for respiratory morbidity. Healthy young adults are not likely affected, even at high ambient levels. DRINKING WATER GUIDELINES This section reviews USACHPPMâs considerations for contaminated drinking water. Packaged and treated water is not always available during deployments. Therefore, guidelines are necessary to evaluate the quality of local water sources.
106 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS How Water MEGs Were Derived USACHPPM reviewed drinking water standards from EPA and other organizations to identify starting points for calculating 5-day, 2-week, and 1-year water MEGs. Table 5-3 provides the health-effects definitions used for determining the water MEGs. As with the air MEGs, certain adjust- ments were performed to make the values more relevant to the military population. For example, the daily water consumption rates of deployed personnel are much higher (between 5 and 15 liters [L]) than those of the general population (which are assumed to be about 2 L). The hierarchy used to choose references on which to base the water MEGs was set as follows: 1. DOD tri-service military field drinking water standards. Field drinking water standards (FDWS) were developed for military personnel primarily to prevent performance degradation on the battlefield. The mili- tary water consumption rate was assumed, and no uncertainty factors were used to protect susceptible subpopulations. FDWS were used to derive water MEGs for arsenic, chloride, cyanide, lindane, magnesium, and sulfate as well as the CWAs sulfur mustard, lewisite, nerve agents (GA, GB, GD, and VX), BZ, and T-2 toxins. For CWAs, only 24-hour MEGs were de- rived, because the likelihood of CWA exposures extending beyond a 24- hour period is small. 2. EPA health advisories (HAs). HAs were used to derive about 50% of the 1-year water MEGs. HAs are guidelines provided for exposure dura- tions of 1 day, 10 days, long-term, or lifetime. Long-term is defined as less than 7 years, and those values are based on a 70-kg adult consuming 2 L of water each day. Adjustments were made to compensate for the higher drinking water consumption rate of military personnel. 3. MRLs. MRLs were used in the absence of drinking water guide- lines (like MCLs or HAs). Intermediate levels were chosen (15-364 day potential exposure). Because MRLs are expressed as daily human doses in milligrams per kilogram per day, it was necessary to convert the values to corresponding water concentrations by assuming a 5 L/day consumption rate and an adult body weight of 70 kg. 4. Reference doses (RfDs). Subchronic or chronic RfDs were used as the basis of water MEGs when no other existing long-term health guidelines were available. About 20% of the long-term water MEGs were calculated using the RfDs. The RfDs were obtained from EPAâs HEAST, IRIS, or Region III risk-based concentration (RBC) tables. Because RfDs are ex- pressed as daily human doses in units of milligrams per kilogram per day,
TABLE 5-3 Definitions of Health Effects Associated with Water MEGs Exposure Duration Health Effect Health Effects and Performance Degradationa 5 days, 5 or 15 L/day Minimal to Nonsignificant The drinking water concentration for a continuous daily consumption of either 5 L/day or 15 L/day for up to 5 days that should not impair performance and is considered protective against significant noncancer effects. Increasing concentration and/or duration could result in performance degradation, need for medical intervention, or increase the potential for delayed or permanent disease (e.g., kidney disease or cancer) 14 days, 5 or 15 L/day Minimal to Nonsignificant The drinking water concentration for a continuous daily consumption of either 5 L/day or 15 L/day for up to 14 days that should not impair performance and is considered protective against significant noncancer effects. Increasing concentration and/or duration could result in performance degradation, need for medical intervention, or increase the potential for delayed or permanent disease (e.g., kidney disease or cancer) 1 year, 5 or 15 L/day Nonsignificant to None The drinking water concentration for a continuous daily consumption of either 5 L/day or 15 L/day for up to 1 year that should not impair performance and is considered protective against health effects including chronic disease and increased risk to cancer (i.e., cancer risk greater than 1 x 10-4). Increasing concentration and/or duration could increase the potential for delayed or permanent disease (e.g., kidney disease or cancer) a Sensitive individuals could be predisposed to toxic effects and, therefore, could be more susceptible. If available scientific evidence regarding susceptible subpopulations exists for a particular chemical, then that information is provided in the guideline tables. Source: USACHPPM 2000a. 107
108 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS it was necessary to convert the values to corresponding water concentra- tions by assuming a 5 L/day consumption rate and an adult body weight of 70 kg. To assess whether the long-term MEGs were protective against cancer, USACHPPM compared risk-specific (1 Ã 10-4) concentrations of carcino- genic chemicals with the corresponding long-term water MEGs. The risk- specific concentrations were obtained from EPAâs drinking water regula- tions and HAs or from IRIS, and they all assumed a lifetime of exposure. Those values were adjusted to estimate concentrations in water that would pose the same cancer risk for an exposure of 1 year and to account for the military water consumption rate. If the adjusted cancer-risk concentration was equal to or greater than the corresponding long-term MEG, the MEG was considered to be protective against cancer. When the adjusted risk- specific value was less than the long-term MEG, the MEG was replaced with that value. Evaluation of and Recommendations on the Derivation of Water MEGs Selection of Chemicals and Existing Standards The chemicals for which water MEGs were developed were selected on the basis of USACHPPMâs review of existing water contaminants, includ- ing chemicals listed in Technical Bulletin, Medical 577 (U.S. Department of the Army 1999); chemicals detected during water sampling in Bosnia; and compounds listed as high priority in RD-230. Although that is a good start, USACHPPM should consider periodically updating the list of chemi- cals with other compounds that are likely to be present, such as industrial mixtures, like gasoline and diesel fuel, and newly identified contaminants. USACHPPM used relevant existing guidelines set by other agencies as starting points for deriving water MEGs. The major difficulty with that approach is that the target populations for the existing guidelines were usually different from the deployed population, which means that some of the assumptions and considerations used might not be relevant to military guidelines. This is discussed in detail in Chapters 2 and 3 and above in the discussion of air MEGs. Other problems include ensuring that the values are properly recorded and updated and that they gain acceptance within the scientific community. For example, the water MEG for lead is based on a WHO guideline that was incorrectly recorded by USACHPPM as 0.05
MILITARY EXPOSURE GUIDELINES 109 mg/L instead of 0.01 mg/L. In addition, several exposure values taken from EPAâs HEAST database were not peer-reviewed and are not as accepted as other guideline values. Time Frames Water MEGs are set for short-term exposures of 5 days and 2 weeks and for a longer period of 1 year. Long-term MEGs were not set for CWAs because their chemical and physical characteristics make long-term expo- sures implausible. These time frames seem appropriate for the military population, whose exposures might range from a limited number of days up to a year. Route of Exposure The guidelines reflect drinking water exposures only. No bathing, dishwashing, or other potential nonpotable water exposures were consid- ered. It appears that USACHPPM considered exposure to water contami- nants by inhalation or dermal exposure to be much less than would occur from ingestion of water. In the case of dermal exposures, the subcommittee performed rough risk calculations using data for some volatile and semi- volatile organic compounds, such as trichloroethylene and benzo(a)pyrene, to evaluate dermal exposure during showering and agrees with USACH- PPM that risk from water consumption generally subsumes that from der- mal absorption. This was true even for chemicals known to have high potential for skin penetration (evaluated by comparing penetration constants [Kpâs]). If the Kp was greater than or near unity (as was the case for certain PAHs), those moieties posed dermal risk close to but not greater than drink- ing water risk. However, it was unclear whether inhalation of volatile chemicals during showering would be as minimal. This is an issue that needs further consideration by USACHPPM. It might also be worthwhile for USACHPPM to consider more unusual exposures to water contami- nants, such as those that might occur in water emersion scenarios. Water Consumption Rates Water MEGs are based on specific exposure conditions defined by estimated daily water consumption rates. Five liters per day is often used
110 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS as the default military water consumption rate, but in dry, arid climates that rate could be as high as 15 L/day. These high rates have been validated and established in Army doctrine (U.S. Department of the Army 1999) and are consistent with reports from the Israeli Defense Forces and U.S. Army Medical Services officers in the Mojave Desert (Henry 1985). SOIL EXPOSURE GUIDELINES This section reviews the technical guidance in RD-230 for deriving guidelines for chemical hazards in soil. In contrast with the MEGs for air and water, soil MEGs are derived for only long-term exposures. How Soil MEGs Were Derived Risk-based soil screening levels are derived by using selected target risk levels, assumptions about exposure routes and characteristics, and toxicity values to calculate acceptable exposure concentrations. For soil contami- nants, three exposure routes are typically considered: inhalation of resus- pended soil particulates; ingestion of soil; and dermal absorption of chemi- cals from soil adhered to skin. When all three exposure routes are included, the soil MEG represents a soil concentration that will not cause unaccept- able health risks even when the three exposure routes are combined. It should be noted that soil MEGs are the only MEGs that consider multiple routes of exposure. For VOCs, the soil saturation concentrations were used as the soil MEGs when they exceeded the health-based values. Soil MEGs were derived in a manner consistent with the derivation of risk-based screening levels for the general population with adjustments added to better represent the characteristics of deployment exposures. The chemicals selected for soil MEG development were described by USACH- PPM as âconsistent with those used to develop drinking water guidelines.â The rationale for the selection process is that ingestion is the primary expo- sure route for both soil contaminants and drinking water contaminants. The target risk levels for soil MEGs are consistent with the target risk levels for air and water long-term MEGs (a target cancer risk of 1 Ã 10-4), although a hazard quotient of 1 was used for noncancer end points. For chemical mixtures, carcinogenic effects are assumed to be additive, whereas for noncancer effects, target organs should be evaluated to determine if additivity is likely. Three EPA methods for deriving risk-based soil screening levels were
MILITARY EXPOSURE GUIDELINES 111 considered for use in developing soil MEGs. The EPA methods included the Office of Solid Waste and Emergency Response soil screening levels (SSLs) (EPA 1996b), the EPA Region III RBCs (EPA 1999a), and the EPA Region IX preliminary remediation goals (PRGs) (EPA 1998). These three methods have some differences in assumed exposure scenarios, routes, and exposure characteristics. All three methods use similar approaches to de- velop soil standards for residential land use, but only the RBCs and PRGs included industrial soil goals at the time the MEGs were developed. Expo- sures of deployed personnel will be more similar to industrial exposures than to residential exposures. The EPA Region IX PRG method was se- lected for modification by DOD because âit results in the most conservative soil concentrations since it includes more exposure pathways than either the SSL or the RBC methodology.â Adjustments were made to the selected model to more accurately represent deployment conditions. In choosing inhalation toxicity values, USACHPPM used the same hierarchy as was used to derive the air MEGs. For oral toxicity values, CSFs from IRIS and HEAST were used for carcinogens, and the water MEG values were used to back calculate oral RfDs. Dermal toxicity values were derived by adjusting the oral toxicity values to represent absorbed dose instead of ingested dose whenever gastrointestinal absorption data were available. Critical assumptions used to estimate oral, inhalation, and dermal expo- sures include soil ingestion rate, inhalation rate, skin surface area, skin adherence factors, and dermal absorption. In RD-230, an average soil in- gestion value of 265 mg/day was derived by assuming that soldiers have equal numbers of high ingestion and low ingestion days while deployed. A high ingestion value of 480 mg/day was assumed to be associated with activities such as digging or crawling on the ground, and 50 mg/day was assumed to be the mean ingestion rate on days when troops do not engage in such activities. The daily inhalation rate of 29.2 m3/day that was used for derivation of air MEGs was also used for the soil MEGs. Dermal exposure assumptions were based on a 90th percentile value for adult male skin sur- face area, a soil adherence factor of 1.0 milligram per square centimeter (mg/cm2), and dermal absorption values of 1% for inorganic chemicals and 10% for organic chemicals. The soil MEG for lead was treated as a special case. RD-230 provides a soil MEG for lead derived on the basis of an EPA modification of the Bowers adult lead model (Bowers et al. 1994; EPA 1996c). EPA greatly increased the conservatism of the model by increasing both the biokinetic slope factor (from 0.375 to 0.4 micrograms [:g] per decaliter [dL] per :g/ day) and the absorption factor for lead in water and diet (from 0.08 to 0.20)
112 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS (Bowers and Cohen 1998). DOD has modified the model further by in- creasing the target blood lead concentration from 10 :g/dL for the fetus to 30 :g/dL for an adult worker. The soil ingestion was also increased from 50 mg/ day to 265 mg/day. Evaluation of and Recommendations on the Derivation of Soil MEGs The general approach used to evaluate soil hazards conforms to current risk-assessment practices. However, the accuracy of the resulting MEGs depends on whether and how the guidelines will be kept up to date as EPA guidance and the underlying science continues to evolve. There are already a number of instances in which specific assumptions used to derive the cur- rent MEGs have been (or will soon be) superceded by revised and updated guidance from EPA. The subcommitteeâs specific comments and recommen- dations are described below. Minor errors are noted in Appendix A. Selection of Chemicals As described above, the chemicals for which MEGs were developed were selected with the assumption that ingestion is the primary exposure route for soil contaminants. That rationale should only apply to nonvolatile chemicals, because for volatile chemicals, the primary route of exposure from soil is often inhalation of released vapors. Inhalation exposures are of particular concern for troops in trenches or in tents or buildings. Common volatile chlorinated solvents, such as tetrachloroethylene, trichloroethylene, and carbon tetrachloride, were inappropriately excluded from the soil MEGs. Elemental mercury is another common, highly volatile industrial chemical that should have a soil MEG based on potential inhalation expo- sures. RD-230 states that soil MEGs might be derived for manganese and selenium in the future. Oral RfDs are available on IRIS for both of those chemicals at this time; however, deriving soil MEGs for manganese and selenium should not be a high priority, because their toxicities are generally low. Method Selection DOD should update its references for the three methods considered for
MILITARY EXPOSURE GUIDELINES 113 deriving MEGs. Both the EPA Region III and EPA Region IX databases were updated in October 2002 (EPA 2002a,b), and the newer references should be cited. In addition, EPA has developed supplemental SSL guid- ance (EPA 2002c) that includes an industrial scenario. That guidance should be discussed in the next update of RD-230 and should be considered as a possibly more appropriate method for soil-MEG derivation. Any re- vised assumptions in the updated guidance should be considered for inclu- sion in the soil MEGs. For example, in the derivation of their particulate emission factor (PEF), EPA (2002c) has changed the value of its site-spe- cific dispersion factor (Q/C) from 90.8 (g/m2-s per kg/m3) to 93.77 (g/m2-s per kg/m3). As described above, the EPA Region IX PRG method was selected because it considers more exposure pathways than the other methods. However, DOD has excluded the inhalation route of exposure to volatile chemicals released from soil from the calculations for soil MEGs. Because that exposure route is probably responsible for the relative conservatism of the PRGs, excluding it is inconsistent with the rationale for selecting the PRG method. The reason that vapor inhalation was excluded form the soil MEGs is that field sampling of air concentrations should also capture vapors released from soil. That assumption needs to be examined further. Volatile organic compound (VOC) releases from soil are dependent on temperature and other weather-related factors. It is possible that field sampling would not occur under the conditions most likely to facilitate the release of VOCs. SSLs based on vapor inhalation are also generally lower than SSLs based on soil ingestion. Evaluating VOCs in soils is further complicated by the technical difficulty of obtaining accurate measurements of VOC concentra- tions in soil due to vapor loss from samples. One possible approach might be to follow the SSL method of listing two separate standardsâone for combined ingestion and dermal exposures, and another for inhalation expo- sures. Identifying exposure-route-specific soil MEGs has the added advan- tage of providing information that might be useful in identifying appropriate protective actions in the event of exposure. Actions appropriate for avoid- ing exposures to vapors released from soils will be very different from those appropriate for avoiding soil contact leading to oral and dermal exposures. Toxicity Data As noted above in the recommendations for the derivation of air MEGs, the subcommittee recommends that USACHPPM avoid using toxicity val-
114 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS ues listed in the HEAST tables for the derivation of MEGs. Also, USACH- PPM should note the date of the underlying toxicity assessment used to derive the MEGs in the RD-230 tables. The subcommittee noted several flaws in the adjustments made to the inhalation toxicity values used to derive the air MEGs (see comments on air MEGs, above). The inhalation toxicity values used to derive the soil MEGs should be re-evaluated in light of those comments. Once those flaws are corrected, it is likely that the inhalation pathway will not have a significant influence on the soil MEGs for semivolatile and inorganic chemicals. The adjustments made to oral toxicity values for deriving the water MEGS were judged to be appropriate, and their use in deriving soil MEGs is also appro- priate. The description of the derivation of dermal toxicity values in RD-230 is misleading and includes errors that need to be corrected. The dermal toxicity section of RD-230 cites a 1989 EPA document as a source of guid- ance in the development of dermal toxicity values. The supplemental guid- ance for dermal risk assessment is more current and is available as a public review draft (EPA 2001a); it was scheduled to be finalized in 2003. EPA (2001a) indicates that VOCs can be excluded from evaluations of dermal exposures to soil contaminants because they are released to air before being absorbed. Also, only two inorganic chemicals are currently included in the EPA guidance (arsenic and cadmium). RD-230 should be revised to explain that the evaluation of dermal exposures requires the conversion of oral toxicity values based on intake into toxicity values based on absorbed dose. The text says âIf a chemical- specific GI ABS [gastrointestinal skin absorption factor] is not available, then a default value of 100 percent is recommended.â RD-230âs Table E-4 shows oral absorption values of 100% for all of the chemicals listed. EPA (2001a) should be consulted as a source of oral absorption values. Gener- ally, oral absorption of organics is greater than 50%, and EPA (2001a) has determined that no adjustment of oral toxicity values is necessary when oral absorption is 50% or greater. Many inorganics are much less completely absorbed, so the use of an assumed 100% oral absorption would not yield adequately protective values for those chemicals. For example, oral absorp- tion of cadmium is 5% or less. The proper adjustment for evaluating risk associated with an absorbed dermal dose of cadmium would be to multiply the oral RfD of 1 Ã 10-3 mg/kg-day (based on ingested dose) by a factor of 0.05 to yield an absorbed dose RfD of 5 Ã 10-5 mg/kg-day. Using the prop- erly adjusted RfD to evaluate dermal exposure to cadmium yields a risk estimate 20-fold greater than that yielded using the unadjusted oral RfD. Consequently, use of unadjusted oral toxicity values to assess dermal expo-
MILITARY EXPOSURE GUIDELINES 115 sures could lead to substantially underestimated risk estimates. This section of RD-230 should be revised to reflect EPAâs guidance. The dermal toxicity discussion in RD-230 also notes that the absence of an ACGIH skin notation was used to preclude some chemicals from evaluation for the dermal exposure pathway. That assumption should be examined more carefully. If the ACGIH notations are based on exposures to pure chemicals, they might not be relevant for assessing the potential for dermal absorption from soil mixtures. It would be more consistent with current risk-assessment practice to select chemicals on the basis of EPAâs latest guidance. Exposure Factors Soil contamination is assumed not to pose an immediate or severe haz- ard unless there are obvious, avoidable signs such as odor, discolored vege- tation, or free chemical product. That assumption is appropriate for oral or dermal exposures to chemicals in soil but might not always be applicable to exposures to vapors released from soil. Vapor inhalation is not likely to be an immediate concern for outdoor, ground-surface activities, but some activities in trenches or enclosed areas such as buildings or tents might allow vapors released from soil to build up to concentrations of concern that would not be detected by ambient air monitoring. The potential for that should be addressed in RD-230, and some screening calculations should be performed to determine the potential health threats from volatile chemicals in subsurface soil. As described above, the critical assumptions used to estimate inhalation, oral, and dermal exposures include inhalation rate, soil ingestion rate, skin surface area, skin adherence factors, and dermal absorption. The inhalation assumptions were addressed in the review of air MEGs. For soil ingestion estimates, RD-230 properly notes that there is extreme uncertainty in those values for adults because of an almost complete absence of reliable empiri- cal data. The discussion of dermal exposures in RD-230 needs to be updated with references to EPAâs new dermal risk-assessment guidance (EPA 2001a). DOD used the 90th percentile of adult male body surface area for head, hands, and forearms in calculating dermal exposure. That assumption is in direct conflict with current EPA risk-assessment guidance, which notes that the 50th percentile population values should be used whenever 50th percentile body weight values are used, because body surface area and body weight are correlated. The estimated surface area used by DOD is 4,090
116 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS cm2, whereas EPA used a value of 3,300 cm2. The RD-230 value should be changed to be consistent with the EPA guidance. An adherence factor of 1.0 mg/cm2 was selected to derive soil MEGs. That value is much higher than the EPA value of 0.2 mg/cm2 for industrial exposures (EPA 2001a). A number of studies have increased our database for soil adherence to skin (Kissel et al. 1996, 1998; Holmes et al. 1999), and they were used by EPA in deriving the new adherence factors. Those stud- ies should be considered in re-evaluating adherence factors for soil MEGs. In particular, DOD should consider using high and low soil adherence val- ues with the same frequency that high and low soil ingestion values are used. The adherence factor of 1.0 mg/cm2 should only be applied to the activities that are assumed to have a higher soil ingestion rate (activities such as digging or crawling on the ground). The lower adherence factor of 0.2 mg/cm2 should be applied to other activities. Because DOD assumes that the high-soil-contact activities occur 50% of the time, the average soil adherence would be 0.6 mg/cm2. The discussion of chemical-specific dermal absorption factors should be revised to reflect the most current guidance (see EPA 2001a). As noted earlier, VOCs can be excluded from the dermal pathway, and only a subset of inorganics is included by EPA at this time. DOD also presents specific dermal absorption factors for CWAs that were derived by assuming that the chemicals in soil first dissolve in water and then are absorbed at rates deter- mined by their flux from water (USACHPPM 1999). The partitioning of chemical from soil to water is predicted on the basis of the octanol water coefficient (Kow) and the fraction of organic carbon in the soil. The flux is determined by a model based on water solubility, Kow, and molecular weight. Estimates of the hourly absorption fractions for CWAs in soil are then derived from the flux (e.g., 0.35% for GB). Those hourly fractions can add up to relatively high absorption estimates over time. For example, for a 12-hour exposure, the absorbed fraction would increase to 3.3% of the amount of GB in adhered soil. The estimated absorption fraction for GB derived using the DOD model is at least an order of magnitude greater than values observed for chemicals that have much higher octanol water coefficients. For example, Wester et al. (1992) did a study with chlordane in soil. Chlordane was added to soil to yield a concentration of 67 parts per million (ppm) and was placed on the skin at a density of 40 mg of soil per square centimeter with 2.7 :g of chlor- dane per square centimeter. At 24 hours, 0.34% of the original dose was in the skin and 0.04% had penetrated the skin. Counting both the amount that
MILITARY EXPOSURE GUIDELINES 117 was in the skin and the amount that had penetrated the skin, the calculated flux was 0.01 :g/cm2/day (2.7 Ã 0.0038), which is equivalent to 0.00042 :g/cm2/hour or 0.015% per hour (more than an order of magnitude below the 0.35% estimated for GB using the octanol water partitioning model). Adding the amount of chemical in the skin to the flux is a conservative assumption, because the chemical might evaporate or might be bound in the skin and sloughed with normal skin turnover. If the DOD model had been applied to chlordane, a flux much greater than the estimated flux for GB would have been predicted. Consequently, the discrepancy between the empirical data for chlordane and DODâs predicted values casts doubt on the validity of the DOD model. It appears that there is not any scientific support for the assumption that partitioning to water will be predictive for dermal absorption from soil. EPA does not estimate dermal absorption of other soil contaminants using a flux, but instead uses estimates of the absorbed fraction of applied dose each day. This is a controversial subject at present, and which model is most appropriate might depend on how contact with soil occurs. EPAâs model essentially assumes one soil application per day that stays on the skin all day, whereas other scientists argue that there would be continued soil contact and turnover on skin throughout activities. If soil stays on the skin for a prolonged period, the amount of contaminant absorbed per hour is likely to decrease over time, so assuming a constant flux might not be ap- propriate. It seems likely that there would be some turnover during activi- ties that bring personnel in contact with soil, but it also is likely that a resid- ual amount would stay on the skin for a prolonged period (especially during deployments). With a few exceptions, it is implausible that CWAs would be present in soil for prolonged periods of time. Newer CWA residues would be ex- pected to behave differently than older residues, and absorption of new residues is likely to be substantially different from absorption of old resi- dues. Short-term soil MEGs should be considered for the volatile CWAs, and long-term MEGs should be derived only for the toxic breakdown prod- ucts and for HD (which can be preserved in a coating of polymeric hydroly- sis products). For short-term CWA soil MEGs, the Army will need to break down the screening levels by exposure route to allow for field personnel to decide on appropriate protective actions. It is vital to know if the primary risk is from inhalation, dermal exposures, or soil ingestion. Soil ingestion is relatively easy to avoid, whereas inhalation can not be avoided without a respirator.
118 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS Lead As described above, EPA increased the conservatism of the Bowers adult lead model by increasing both the biokinetic slope factor and the absorption factor (Bowers and Cohen 1998). DOD then increased the target blood lead concentration from 10 :g/dL for the fetus to 30 :g/dL for adult workers, an action that reduced the protectiveness of the model. DOD also increased the soil ingestion from 50 mg/day to 265 mg/day. It is not clear whether these and other assumptions made by DOD are justified. The 30 :g/dL target blood lead might not be protective enough for the embryo and fetus. However, many of the other assumptions are overly conservative and could result in substantially overpredicted blood lead levels. Lower lead absorption rates are supported by a validated pharmacokinetic model (OâFlaherty 1993), and a lower biokinetic slope factor was used to accu- rately predict blood lead levels during development of the original model (Bowers et al. 1994). The background blood lead level used by DOD is based on a 1988-1991 survey of blood lead levels in the U.S. population. Contrary to the errone- ous definition of PbB1 as background blood lead concentrations in adult males, the values in table RD 3-7 are for women 17-45 years old, which is the appropriate receptor demographic to evaluate. More recent data suggest declines in the blood lead levels of women of child-bearing age since 1988- 1991, the range of blood lead levels having fallen from 1.7-2.2 :g/dL to 1.4-1.9 :g/dL (EPA 2002d). Table RD 3-7 is missing a value for geometric standard deviation (GSD), a parameter needed to calculate soil screening levels. EPA (2002d) indicates that as background lead levels have fallen, the GSD has risen. However, the rise in the GSD appears to be an artifact of increasing proportions of nondetected values in the database (up to 25%). Therefore, EPAâs (2002d) conclusions should not be accepted without a detailed critical evaluation. The soil ingestion rate selected by DOD is very high and is highly uncertain. The adult lead model was designed to use a central tendency estimate of soil ingestion. To the extent that the DOD value is an upper-end estimate, its use will lead to substantial overprediction of blood lead levels. The model requires central tendency estimates for all parameters and ap- plies a GSD to derive a population distribution. Use of input parameters that are not central tendency estimates is indefensible given the modelâs structure. The RD-230 analysis yields a soil MEG for lead of 2,200 ppm. On the basis of observations made in communities that have elevated soil lead
MILITARY EXPOSURE GUIDELINES 119 levels, it was concluded that 2,200 ppm is not likely to result in unaccept- able exposures. The overly conservative exposure assumptions might have balanced out the unprotective target blood lead concentration to yield a reasonably protective soil MEG. Given the availability of reliable tests for blood lead levels, it might be feasible to periodically monitor blood lead levels in troops exposed to soils whose lead levels exceed the soil MEG to ensure that target blood lead levels are not also exceeded. Consideration of Acute Toxicity The soil MEGs are set for 1-year exposures. USACHPPM performed an analysis to verify that soil MEGs do not pose an acute risk at higher exposure rates by comparing the soil MEGs to EPAâs 1-day drinking water HAs. RD-230 cites an EPA document from 1996 (1996d) for the HA val- ues, but a 2002 update of the HAs posted on EPAâs website includes some short-term HAs dated 1998 (EPA 2002e). The more recent reference should be used. Also, many of the HAs are based on assessments that are more than 10 years old. Acute toxicity values (MRLs) for many of the chemicals of interest are also available from ATSDR (2003). If those are generally more up-to-date than the HAs, it might be advisable to use them instead. In evaluating potential acute exposures from soil, DOD focused solely on soil ingestion. As described above, the subcommittee does not agree with DODâs rationale for excluding many volatile chemicals from soil- MEG derivation. When more accurate long-term soil MEGs that consider inhalation exposures are derived for volatiles, it will also be necessary to develop short-term MEGs to evaluate potential acute risks from those chem- icals. APPLICATION OF MEGs Deployment conditions are undoubtedly complex. Potential toxic chemical exposure scenarios are expected to be highly variable from one deployment to the next. Personnel with sufficient knowledge and field experience are needed to assess the health risks associated with these com- plex potential exposures by using the information provided through the MEGs. However, more explicit guidance on how to apply the MEGs is necessary to ensure that assessments and the resulting management actions are performed consistently. Particularly, guidance is needed on how to
120 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS develop appropriate risk-management plans for when measured or predicted exposure concentrations exceed the MEGs and how to adequately character- ize the risks in the event of exposures to the same toxic agent through multi- ple routes and pathways or simultaneous exposures to several chemicals. Interpreting MEG Exceedances As noted in Chapter 2, two separate sets of guidelines are necessary to appropriately assess chemical threats to the mission and chemical threats to force health. Under this new scheme, it would not be appropriate for the health-based MEGs to be used in conjunction with the militaryâs mission risk assessment matrix to evaluate mission risks, as is currently recom- mended in TG-230. Rather, the subcommittee envisions that MEGs will be used as guidelines to assess health risks and the potential risk-management options for reducing or eliminating those risks. That information would then be considered by decision makers in conjunction with mission-related risk assessments. For example, in cases where some level of health risk is accepted to complete the military objective, MEGs could be used to deter- mine the medical follow-up responsibilities of DOD. The subcommittee recommends that DOD develop a risk-management framework that focuses on action plans (i.e., responses) for when MEGs are exceeded. Actions plans should include, but should not be limited to the following elements: â¢ Formulating better characterizations of exposures, including identi- fication of the sources and of the contributions from various contaminated media. (More extensive discussion on exposure assessment is provided in Chapter 3.) â¢ Setting limitations on the lengths of deployments. â¢ Identifying remedial options. â¢ Identifying exposed individuals who are at greater risk for adverse effects, triggering one or more of the following actions: âPost-deployment follow-up with exposed individuals. âIdentification of unusually susceptible individuals. âLimitations on multiple deployments. âConsideration of the possibilities of other exposures contributing to the same health outcomes. âProvision of long-term care.
MILITARY EXPOSURE GUIDELINES 121 Assessing Aggregate Exposure As discussed in Chapter 3, aggregate exposure is total exposure to a single chemical by multiple pathways and routes. The paths that chemicals travel to reach the media through which individual exposures occur are referred to as exposure pathways. Most pathways are complex. For exam- ple, lead added to gasoline (medium 1) is emitted to the air (medium 2) when gasoline is burned. Some of the airborne lead is deposited in soil (medium 3), which is used for growing corn. Some of the lead in soil dis- solves in water (medium 4) and moves through the roots of the corn plant, accumulating in the kernels of corn (medium 5), and the corn is fed to dairy cattle, leading some of the lead to be excreted in cowsâ milk (medium 6). In this scenario, milk is the medium through which humans are exposed to the lead. The lead passed through six media before it reached human be- ings. To make matters more complex, humans could have been exposed to lead at several other points along the pathwayâfor example, by breathing the air (medium 2) or coming into contact with the soil (medium 3). Expo- sure routes are the ways that chemicals can move into the body. They include inhalation, ingestion, skin absorption, absorption through the eyes, placental transfer from a pregnant woman to the fetus, and transfer from mother to child through breast-feeding. In many cases, contaminated media (air, soil, or water) can lead to several exposure routes. For example, hu- mans could be exposed to an organic solvent in tap water through drinking the contaminated water or through inhaling chemical vapors during warm showers. All of the exposure possibilities should be considered when as- sessing human health risks. During short-term missions and deployments, the route of exposure most likely to be of relevance to deployed personnel is inhalation. During longer-term deployments, deployed personnel might be exposed to low levels of common contaminants through various environmental media. In those longer-term scenarios, personnel could inhale contaminants in air that were volatilized from soil and/or water; ingest contaminated water; and/or experience dermal exposures from bathing or from direct contact with con- taminated soils. Assessing risk from each exposure route independently might indicate âlow to moderateâ risk categories; however, considering these potential exposures in aggregate could indicate more significant risks. EPAâs hazard index (HI) method is an example of a simple aggregate- exposure assessment framework that could be implemented by DOD. EPA defines the HI method as an aggregation of individual hazard quotients (HQs) for each route of exposure. The HQs are ratios of exposures to refer-
122 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS ence concentrations. For example, the HQ for inhalation is calculated as follows: HQinhalation = Exposure Concentration (mg/m3). (5-4) RfC (mg/m3) An oral RfD, a dermal RfD, and/or an inhalation RfC must be defined for each route of concern. HQs for each route of concern can then be aggre- gated into an HI. HIpathway = HQoral + HQdermal + HQinhalation. (5-5) Risk increases with increasing HQs and HIs. Generally, HQs or HIs of less than or equal to 1 are of little concern, whereas HQs or HIs of greater than 1 are of greater concern. This is a simple summary of the EPA procedure. EPAâs methodology for aggregate exposure (EPA 1999b, 2001b) should be consulted for more details. For the purpose of assessing aggregate expo- sures involving MEG chemicals, the subcommittee recommends DOD adapt EPAâs method for use with MEGs. For example, HQair, water, or soil = Exposure Concentration/MEG; (5-6) HI = HQair + HQwater + HQsoil. (5-7) Assessing Cumulative Risk TG-230 points out that because âcertain contaminants may have similar adverse effects on the human body, it is necessary to consider the total sum of all similar effectsâ (USACHPPM 2000a). The document further indi- cates that in the preliminary threat analysis, when occupational and environ- mental health (OEH) hazards are identified and prioritized, the effects of exposures to the same or similar chemicals through different media should be considered additive. Algorithms have been adopted by federal agencies to address the problem of exposures to multiple chemicals; however, in RD- 230 USACHPPM states that those quantitative approaches are ânot well- suited to the overall qualitative/ranking nature of the TG-230 deployment risk assessment approachâ (USACHPPM 2002b, p. 5). The subcommittee agrees that conventional algorithms used to assess health risks from multi- ple chemicals are not useful for assessing mission risks. However, for the
MILITARY EXPOSURE GUIDELINES 123 purposes of force health protection, those algorithms are appropriate for assessing cumulative risks to the deployed force. The most common proce- dure is discussed below. Cumulative exposures involve exposures to multiple chemicals. EPA defines cumulative risk as the likelihood of occurrence of an adverse health effect from exposure to multiple chemicals that have common modes of toxicity from all routes and pathways. The subcommittee agrees with the Armyâs assumption that the toxicity of a mixture of chemicals that have similar modes of action will be equal to the sum of the weighted dose toxic- ities of the individual chemicals in the mixture. When assuming âaddi- tivity,â the methods for combining component data described in EPAâs Supplementary Guidance for Conducting Health Risk Assessment of Chemi- cal Mixtures (EPA 2000) could be implemented. The primary method for component-based risk assessments of mixtures of chemicals with similar modes of action is the hazard index (HI), which is derived from dose addition. In the EPA guidance, dose addition is inter- preted as simple similar action where the component chemicals act as if they were dilutions or concentrations of each other, differing only in rela- tive toxicity. Dose additivity might not hold for all toxic effects, and the relative toxic potency between chemicals might differ for different types of toxicity or for toxicity by different routes. To reflect those differences, an HI usually is developed for each exposure route of interest and for a single specific toxic effect or for toxicity to a single target organ. A mixture could then be assessed using several HIs, each representing one route and one toxic effect or target organ. EPAâs HI is defined as a weighted sum of the exposure measures for the mixture component chemicals. According to dose addition, the âweightâ factor should be a measure of the relative toxic strength. The guidelines formula for the HI is general. n HI1 = Î£ Ei/ALi, i= (5-8) where Ei is the exposure level to chemical i, ALi is the acceptable level for chemical i, and n is the number of chemicals in the mixture. When an effect-specific HI exceeds 1, potential toxicity is a concern. In practice, EPA usually calculates the HIs by using RfDs or RfCs as the ALs. By modifying the formula, DOD can utilize other expressions for exposure and relative toxicity that might be more appropriate for deployment situations.
124 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS To apply this HI approach to military situations, the relevant MEGs should be used as the ALs. In practice, the HI method could be applied to chemicals that have similar target-organ effects. However, given the range of data sources on which the current MEGs were based, to begin considering cumulative risk the existing information in TG-230 and RD-230 would have to be reorga- nized by target organs. That will require going back to the source data in some cases to identify all of the end points considered in addition to the critical effects on which the source guidelines were based. It might be more practical to use a qualitative assessment scheme as the first stage in integrat- ing cumulative risk considerations into the MEG guidance. Repeated Exposures and Multiple Deployments Many soldiers will participate in multiple deployments during their military career. It is unlikely that multiple short-term deployments (less than or equal to14 days) involving exposures at levels below the MEGs (but not necessarily the CCEGs) will affect the likelihood of toxicity. The im- pacts of multiple long-term deployments are more relevant to force health protection. As described in previous sections, long-term MEGs for noncancer effects were based preferentially on subchronic toxicity values. However, in the many cases where only chronic toxicity values were available, those values were used. When long-term MEGs are based on chronic toxicity data they will be protective for lifetime exposures and will, therefore, also be protective for multiple deployments. When long-term MEGs are based on subchronic toxicity data, they will be protective for up to 7 years (10% of lifetime). Thus, a soldier would need to have more than 7 years of de- ployment exposures to a chemical at concentrations close to the long-term MEG before any concern would arise regarding the health impacts of those multiple deployment exposures. Furthermore, the UFs applied to the noncancer MEGs provide additional protection. For the long-term MEGs based on cancer risks, risks from multiple deployments might be viewed as irreversible over time. In a worst-case scenario with multiple exposures near MEG concentrations based on 1 Ã 10-4 incremental cancer risk, it is unlikely that risks from multiple deployments will contribute to total risk in excess of 1 Ã 10-3, which is a target risk used to develop many occupational standards. At that upper-bound risk level, risks to an individual soldier are still low compared with the background cancer risks of 0.25 (or one in four).
MILITARY EXPOSURE GUIDELINES 125 The subcommittee was informed that DOD is working to record sol- diersâ exposures, to the extent feasible. Those records would be available if the need for retrospective analysis arose. For example, when exposures in excess of long-term MEGs occur during a deployment, it might be useful to review the records of exposures from previous deployments. However, there is no indication of an imminent need for prospective analysis of these records for the purpose of monitoring future deployments. RECOMMENDATIONS This section summarizes the major recommendations for the develop- ment and application of MEGs. The chapter itself should be consulted for more thorough descriptions and for several other recommendations that are more detailed, are more chemical-specific, or are of secondary importance. Overall, the MEGs must be re-evaluated and revised to make them more relevant to force health protection and more consistent with each other. Ideally, USACHPPM will develop a set of principles, guidelines, and proce- dures for developing MEGs de novo from the primary toxicology data. Those procedures would solidify the purpose and goals of the MEGs and would make explicit the risk-management policy decisions that underpin the removal or modification of uncertainty factors used in the existing guide- lines set by other agencies. However, the subcommittee realizes the immen- sity of that undertaking and suggests that, in the interim, revisions be made to improve the quality of the MEGs. To assist in obtaining and managing resources for this effort, DOD should analyze the resources (staff and fund- ing) needed to accomplish the recommendations, prioritize the tasks, and estimate how much time it will need to complete this work. Near-Term Revisions â¢ Improve the quality of the MEGs by making revisions that require relatively minimal resources. Specifically, USACHPPM has applied some adjustments to the source guidelines to make them relevant to the deployed population but does not appear to have done so consistently. The following are recommended modifications: âWhen using TLVs to derive the 14-day and 1-year MEGs, it is unnecessarily conservative to apply a UF of 10 to account for uncer-
126 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS tainty associated with extrapolation from intermittent to continuous exposure. âAll relevant MEGs should include the military adjustment factors for higher ventilation and water-intake rates. âFor the six criteria air pollutants, ensure that the MEGs are ap- propriate to the military population rather than to susceptible civilian subpopulations. âPeriodically review the guidelines set by other organizations that were used as sources for the MEGs. If those sources have been revised, incorporate the changes into the MEGs. Values reported in HEAST should not be used as bases for MEGs, because they have not been peer-reviewed. Additional exposure guidelines should be consulted, such as the RfCs developed by the State of Californiaâs Office of Envi- ronmental Health Hazard Assessment. âImprove the documentation of the existing exposure guidelines by specifying the date of their establishment, the toxicity end points on which they are based, UFs used, and any special considerations in the supporting reference tables. Adjustments to the values should be made on a case-by-case basis. âDevelop short-term soil MEGs for certain contaminants, particu- larly volatile organic compounds. âRe-evaluate the approach used to assess dermal exposures to CWAs. â¢ Establish risk-management framework that focuses on action plans (i.e., responses) for when the MEGs are exceeded. Appropriate actions would include considering risk-management options for reducing or elimi- nating risks (e.g., using protective gear) and determining the appropriate medical follow-up responsibilities of DOD (e.g., documenting the exposure in medical records, tracking exposed individuals, providing long-term care) when some health risks must be borne. Mid-Term Revisions Steps in this category would result in more relevant and internally con- sistent MEGs that would be less likely to be overly conservative. However, there should be an analysis of the level of effort required for this activity relative to that required for the long-term revisions. The optimal approach to revising the MEGs would be for USACHPPM to consult the original source material (e.g., the critical study selected by EPA for an RfC or can-
MILITARY EXPOSURE GUIDELINES 127 cer unit risk) and perform its own calculations. That would bring more unity to the guidelines. The effort would be time-consuming and would have flaws resulting from out-of-date source materials. However, it would avoid the more time-consuming tasks of literature searches and evaluations of the primary literature while providing a transparent, systematic, and uniform method of applying adjustments to exposure durations, inhalation rates, and water intake rates. It would also standardize the treatment of susceptible subpopulations. In some cases, other agencies could be asked to provide assistance. Discussions with other agencies also might reveal possibilities for accessing professionals already familiar with the assess- ments who could go back and make recalculations. Long-Term Revisions â¢ As discussed, the source material used to derive the MEGs has inherent problems, the primary problem being the obsolescence of many of the values. Thus, simply changing the UFs and other factors will not solve all of the underlying difficulties. However, it is not feasible for USACH- PPM to create MEGs entirely de novo by beginning with literature searches. All of the agencies developing health-based guidelines struggle with the problem of obsolescence. For example, EPA is beginning a major effort to reinvigorate IRIS. That presents an opportunity to explore partnership arrangements. For example, if one agency were doing a de novo assessment on a chemical of interest to the military, it would be relatively easy for that agency to establish one guideline applicable to their interests and another applicable to the military. The word relatively is used because the major effort in assessments is evaluating and interpreting the literature, which would have to be done by the agency as well as the military. â¢ Aggregate exposure and cumulative risk should be addressed, to the extent feasible, in each stage of revisions to the MEGs. â¢ USACHPPM should periodically update the list of chemicals for which MEGs have been derived to include chemicals that were omitted in previous reviews (e.g., gasoline) or that have been newly identified as con- taminants. REFERENCES ATSDR (Agency for Toxic Substances and Disease Registry). 2003. Minimal Risk Levels
128 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS (MRLs). Agency for Toxic Substances and Disease Registry, Atlanta, GA. [Online]. Available: www.atsdr.cdc.gov/mrls.html [accessed Dec. 3, 2003]. Bowers, T.S., and J.S. Cohen. 1998. Blood lead slope factor models for adults: Compari- sons of observations and predictions. Environ. Health Perspect. 106(Suppl. 6):1569- 1576. Bowers, T.S., B.D. Beck, and H.S. Karam. 1994. Assessing the relationship between environmental lead concentrations and adult blood lead levels. Risk Anal. 14(2):183- 189. EPA (U.S. Environmental Protection Agency). 1996a. Air Quality Criteria for Ozone and Related Photochemical Oxidants. EPA/600/P-93/004cF. National Center for Environ- mental Assessment, Office of Research and Development, U.S. Environmental Protec- tion Agency, Washington, DC. EPA (U.S. Environmental Protection Agency). 1996b. Soil Screening Guidance: Userâs Guide, 2nd Ed. Pub. No. 9355.4-23. EPA540/R-96/018. Office of Solid Waste and Emergency Response, U.S. Environmental Protection Agency, Washington, DC. [Online]. Available: http://www.epa.gov/superfund/resources/soil/index. htm#user [accessed Dec. 5, 2003]. EPA (U.S. Environmental Protection Agency). 1996c. Recommendations of the Technical Review Workgroup for Lead for an Interim Approach to Assessing Risks Associated with Adult Exposures to Lead in Soil. Environmental Criteria and Assessment Office, U.S. Environmental Protection Agency, Research Triangle Park, NC. EPA (U.S. Environmental Protection Agency). 1996d. Drinking Water Regulations and Health Advisories. EPA822-R-96-001. Office of Water, U.S. Environmental Protec- tion Agency. October 1996. EPA (U.S. Environmental Protection Agency). 1998. U.S. EPA Region 9 Preliminary Remediation Goal Tables. Office of Solid and Hazardous Waste, U.S. Environmental Protection Agency. EPA (U.S. Environmental Protection Agency). 1999a. EPA Region 3 Risk-Based Concen- tration Table. Risk Assessment, Office of Solid Waste and Emergency Response, U.S. Environmental Protection Agency. EPA (U.S. Environmental Protection Agency). 1999b. Guidance for Performing Aggregate Exposure and Risk Assessments, Draft, February 1, 1999. Office of Pesticide Pro- grams, U.S. Environmental Protection Agency. [Online]. Available: www.epa.gov/ scipoly/sap/1999/february/guidance.pdf [accessed Dec. 5, 2003]. EPA (U.S. Environmental Protection Agency). 2000. Supplementary Guidance for Con- ducting Health Risk Assessment of Chemical Mixtures. EPA/630/R-00/002. Risk Assessment Forum, U.S. Environmental Protection Agency, Washington, DC. [On- line]. Available: http://www.epa.gov/iris/backgr-d.htm [accessed Dec. 3, 2003]. EPA (U.S. Environmental Protection Agency). 2001a. Risk Assessment Guidance for SuperfundâVolume I: Human Health Evaluation Manual. (Part E, Supplemental Guidance for Dermal Risk Assessment), Interim Review Draft. EPA/540/R/99/005. OSWER 9285.7-02EP. PB99-9633112. Office of Emergency and Remedial Response, U.S. Environmental Protection Agency, Washington, DC. [Online]. Available: http://www.epa.gov/oerrpage/superfund/programs/risk/ragse/ [accessed Dec.3, 2003]. EPA (U.S. Environmental Protection Agency). 2001b. General Principles for Performing Aggregate Exposure and Risk Assessments. Office of Pesticide Programs, U.S. Envi- ronmental Protection Agency. [Online]. Available: http://www.epa.gov/pesticides/ trac/science/aggregate.pdf [accessed Dec. 3, 2003].
MILITARY EXPOSURE GUIDELINES 129 EPA (U.S. Environmental Protection Agency). 2002a. U.S. EPA Region 3 Risk-Based Concentration Table. Risk Assessment, Office of Solid Waste and Emergency Re- sponse, U.S. Environmental Protection Agency. [Online]. Available: www.epa.gov/ reg3hwmd/risk/ [updated semiannually]. EPA (U.S. Environmental Protection Agency). 2002b. U.S. EPA Region 9 Preliminary Remediation Goal Tables. Office of Solid and Hazardous Waste, U.S. Environmental Protection Agency. [Online]. Available: http://www.epa.gov/region09/waste/sfund/ prg/index.htm [updated annually]. EPA (U.S. Environmental Protection Agency). 2002c. Supplemental Guidance for Devel- oping Soil Screening Levels for Superfund Sites. OSWER 9355-4-24. Office of Solid Waste and Emergency Response, U.S. Environmental Protection Agency, Washington, DC. [Online]. Available: http://www.epa.gov/ superfund/ programs/ risk/toolthh.htm [accessed Dec. 3, 2003]. EPA (U.S. Environmental Protection Agency). 2002d. Blood Concentrations of U.S. Adult Females: Summary Statistics from Phases 1 and 2 of the National Health and Nutrition Evaluation Survey (NHANES III). OSWER 9285.7-52. Office of Solid Waste and Emergency Response, U.S. Environmental Protection Agency, Washington, DC. [Online]. Available: www.epa.gov/superfund/ programs/lead/products.htm [accessed Dec. 3, 2003]. EPA (U.S. Environmental Protection Agency). 2002e. 2002 Edition of the Drinking Water Standards and Health Advisories. EPA 822-R-02-038. Office of Water, U.S. Environ- mental Protection Agency, Washington, DC. [Online]. Available: http://www.epa. gov/waterscience/drinking/ [accessed Dec. 3, 2003]. Gaylor, D.W. 2000. The use of Haberâs law in standard setting and risk assessment. Toxi- cology 149(1):17-19. Graham, J.A., L. Folinsbee, J.M. Davis, J. Raub, and L.D. Grant. 1999. Critical health issues of criteria air pollutants. Pp. 365-397 in Toxicology of the Lung, 3rd Ed, D.E. Gardner, J.D. Crapo, and R.O. McClellan, eds. Philadelphia, PA: Taylor and Francis. Henry, C.D. 1985. Heat stress and its effects on illness and injury rates. Mil. Med. 150(6):326-329. Holmes, K.K., J.H. Shirai, K.Y. Richter, and J.C. Kissel. 1999. Field measurements of dermal soil loading in occupational and recreational activities. Environ. Res. 80(2Pt1):148-157. Kissel, J.C., K.Y. Richter, and R.A. Fenske. 1996. Field measurement of dermal soil load- ing attributed to various activities: Implications for exposure assessment. Risk Anal. 16(1):115-125. Kissel, J.C., J.H. Shirai, K.Y. Richter, and R.A. Fenske. 1998. Investigation of dermal contact with skin in controlled trials. J. Soil Contam. 7(6):737-752. NRC (National Research Council). 1986. Criteria and Methods of Preparing Emergency Exposure Guidance Level (EEGL), Short-Term Public Emergency Guidance Level (SPEGL), and Continuous Exposure Guidance Level (CEGL) Documents. Washing- ton, DC: National Academy Press. NRC (National Research Council). 2000. Strategies to Protect the Health of Deployed U.S. Forces: Analytical Framework for Assessing Risks. Washington, DC: National Acad- emy Press. OâFlaherty, E.J. 1993. Physiologically based models for bone-seeking elements. IV: Kinetics of lead disposition in humans. Toxicol. Appl. Pharmacol. 118(1):16-29. USACHPPM (U.S. Army Center for Health Promotion and Preventive Medicine). 1999. Derivation of Health-based Environmental Screening Levels for Chemical Warfare
130 TECHNICAL GUIDES FOR ASSESSING AND MANAGING CHEMICAL HAZARDS Agents. U.S. Army Center for Health Promotion and Preventive Medicine, Aberdeen Proving Ground, MD, and Oak Ridge National Laboratory, Oak Ridge, TN. USACHPPM (U.S. Army Center for Health Promotion and Preventive Medicine). 2002a. Chemical Exposure Guidelines for Deployed Military Personnel. Technical Guide 230. U.S. Army Center for Health Promotion and Preventive Medicine. January 2002. [Online]. Available: http://chppm-www.apgea.army.mil/deployment/ [accessed No- vember 25, 2003] USACHPPM (U.S. Army Center for Health Promotion and Preventive Medicine). 2002b. Chemical Exposure Guidelines for Deployed Military Personnel. A Companion Docu- ment to USACHPPM Technical Guide (TG) 230 Chemical Exposure Guidelines for Deployed Military Personnel. Reference Document (RD) 230. U.S. Army Center for Health Promotion and Preventive Medicine January 2002. [Online]. Available: http://chppm-www.apgea.army. mil/deployment/ [accessed November 25, 2003] U.S. Department of the Army. 1999. Sanitary Control and Surveillance of Field Water Supplies, Draft, May 1999. Technical Bulletin, Medical 577. (as cited in USACHPPM 2002a) Wester, R.C., H.I. Maibach, L. Sedik, J. Melendres, C.I. Liao, and S. DiZio. 1992. Percutaneous absorption of [14C]chlordane from soil. J. Toxicol. Environ. Health 35(4):269-277. WHO (World Health Organization). 2000. Air Quality Guidelines for Europe, 2nd Ed. European Series No 91. Copenhagen: World Health Organization.