Potential Radiation Exposure in Military Operations: Protecting the Soldier Before, During, and After (1999)

The previous chapters have described the potential adverse health consequences of radiation exposure, have outlined currently accepted methods for limiting those consequences, and have described current U.S. Army approaches to limiting those consequences. Using that information as background, the committee discusses here how well the Army radiation protection and safety programs are structured to protect soldiers. This report's focus, reflecting the charge to the committee, is radiation doses of 700 millisievert (mSv) or lower that are incurred during military operations.

The Army has published guidance for the control of doses received from routine occupational exposures and those associated with nuclear war. Its work to incorporate concepts of the Allied Command Europe (ACE) Directive (NATO, 1996) is an encouraging step in the development of control measures for other situations. The committee realizes that the Directive was initiated for a specific mission (Bosnia) and that the U.S. Army recognizes its limitations. The comments that follow should be viewed as constructive; in no way does the committee intend them to diminish the significance of the progress that the Army has made toward the control of the complete spectrum of radiation hazards both on the battlefield and in operational situations other than war. In its interim report, the committee recommended that the ACE Directive be revised to ensure completeness and clarity. The U.S. Army has been working in that direction.

The first part of this chapter reprints the interim report's discussion (IOM, 1997) evaluating the ACE Directive in light of standards in the civilian sector. The committee continues this chapter with information and guidance on what it

considers to be three essential components of a radiation protection program: training, recordkeeping, and reporting.

The discussion begins with an assessment of the underlying philosophy of U.S. Army radiation programs. The ACE Directive (NATO, 1996, § 1-2.), which forms the developing basis for U.S. Army policy regarding operations other than war, states that:

1. Deliberate exposure of ACE forces to a radiological hazard shall not be permitted unless it is required by military necessity.
2. All exposures of soldiers to radiological hazards during operations must be kept as low as reasonably achievable consistent with military necessity.

further contamination or to put out a fire in a reactor. In the case of the U.S. Army, the object of the intervention may have nothing to do with the radiation source.

Many situations in military operations resemble practices more than interventions. For example, sending a survey team into an area of unknown radioactive contamination is clearly an intervention, and the ACE Directive is applicable. On the other hand, consider a soldier assigned to guard the entrance of a damaged nuclear plant. The dose rate at the guard station probably would have been measured as the result of a preceding intervention. The provision of routine guard services would no longer be part of the intervention. At that point, exposure levels should be well known and the dose that the soldier receives should therefore be kept not only within accepted dose limits but also as low as reasonably achievable. This activity should be controlled as a practice, not as an intervention.

The committee firmly acknowledges that a military operation is a unique situation in which simple definitions of practices and interventions become complex and conditions may change quickly. In the civilian version of the scenario outlined above, the guard would finish a shift and go home. In the military situation, the plant may suddenly come under attack, resulting in the guard being unable to avoid exceeding occupational limits. Thus, the military situation that began as a practice, subject to dose limits, must now be managed as an intervention.

One could argue that all military operations, since they involve such uncertain situations, should be managed as interventions, without dose limits. However, given the substantial involvement of the military in peacekeeping and humanitarian assistance missions, it is hard to justify not providing soldiers with the level of protection that controlling exposures as a practice would provide. For nonemergency situations, the ACE Directive does not provide guidance that would afford protection to soldiers at a level appropriate for a practice.

The U.S. Army should:

1. Provide soldiers the same level of radiation protection provided to civilians working in similar environments. The ACE Directive appears to manage all military missions involving radiation exposures as interventions. Although this is clearly appropriate for many missions (e.g., emergencies, radiation accidents, and operations involving hostile action), other missions can more properly be treated as routine practices, thereby affording more complete control of the radiation exposure. Missions amenable to control as practices might include security details, decontamination of vehicles, and other scenarios in which hostile action is not expected.

2. Develop and state an explicit radiation protection philosophy that defines missions as falling under the framework of either a practice or an intervention. Practices would be subject to modified requirements of the Army's existing occupational radiation protection program as described previously. It is likely that the situation in Bosnia would fall into this category. Under the committee's recommendations, soldiers would be considered radiation workers if they are assigned military duties that have the potential for radiation exposures that could result in doses in excess of ICRP limits for the public (ICRP, 1991a)—1 mSv per year. A revision of the existing exposure guidance in the ACE Directive would govern those situations that are of an emergency nature and that would be managed as interventions. In both cases, keeping doses as low as reasonably achievable will continue to be of primary importance.
3. Clearly state in the policy paragraph of the subsequent versions of the ACE Directive the definitions adopted for practices and interventions in the necessary military context. The procedures that follow the policy statement should address practices and interventions separately. It would seem reasonable for the commander to have the authority to determine which of these frameworks to follow on the basis of the military mission.

4. Terminology

   The committee considers some terms in the ACE Directive (NATO, 1996) misleading. The first and by far the most serious one is the term low level radiation when it is applied broadly to doses in the range 50 to 700 mSv (5 to 70 rem). Low level may be an appropriate descriptor when comparing these doses to those that could result from the detonation of a nuclear device. In the broader context of radiation protection, however, low level clearly implies much lower doses. Although the terminology may be perfectly clear to those involved in developing the guidance, it probably will be misunderstood by others. The U.S. Army's use of this term to describe doses that approach thresholds for acute effects could easily be misinterpreted as an intent to mislead soldiers on the seriousness of such exposures.

   The committee has concerns about the terms used to describe the effects of dose categories in the table in Annex A of the ACE Directive (NATO, 1996).³ No risk is used to describe the effect of doses of less than 0.5 mGy (0.05 rad). This is inconsistent with international positions on the effects of radiation, specifically, the assumption that even low radiation doses may produce some deleterious effects. Likewise, the term normal risk incorrectly implies that an exposure

   3

   The committee notes that the United States has drafted a revised table for the draft of Annex A to the ACE Directive that replaces the narrative terms for risk (none, normal risk, minimal risk, limited risk, increased risk, and significant risk) with quantitative risk estimates (none, 1:4,000. 1:400, 1:200, 1:80, 1:30).

of 0.5 to 5 mGy (0.05 to 0.5 rad) adds no additional risk to that from exposure to natural background radiation, even though such exposures are considered to contribute very small, possibly negligible, health risks.

Radiological hazard is often used in the ACE Directive to describe any radiation exposure.⁴ Hazard is an ambiguous term. Given the uncertainty as to the magnitude of the health consequences at low levels, the term hazard should not be automatically appended to radiation. Rather, it should be used advisedly to identify the potential for significant health consequences.

The U.S. Army should:

4. Not use the term low level to describe the radiation dose range of 50 to 700 mGy (5 to 70 rad). Low level may be an appropriate descriptor when comparing these doses to those that may be experienced from the detonation of a nuclear weapon. In the broader context of radiation protection, however, low level clearly implies much lower doses.
5. Use terms other than no risk and normal risk for the radiation exposure state (RES) categories labeled RES 0 and RES IA in the table of exposure guidance in Annex A of the ACE Directive. The description of any nonzero dose as no risk is inconsistent with current international positions on the effects of radiation. Likewise, the term normal risk incorrectly implies no additional risk to that from natural background radiation exposures, even though such exposures are considered to contribute very small, possibly negligible, health risks.
6. Avoid the term radiological hazard when describing the exposure of soldiers to radiation unless the hazard refers to a specific detrimental effect. For most cases in the ACE Directive, radiological hazard simply means radiation.

The ACE Directive (NATO, 1996, §1-3.a. and c.(l)) indicates that there is a prospective assessment of risk, high or low. It is unclear whether this refers to (1) intelligence assessments of the likelihood of radiation contamination or (2) the magnitude of measurable levels of radiation contamination.

In principle, the committee agrees with the Directive's requirement for the use of ''dose rate instruments to measure alpha- and beta-emitting particles as well as gamma radiations" (§ 1-3.b.). Instruments sensitive to beta and alpha radiation will be useful in conducting assessments for potential skin contamination and internal deposition and for triggering appropriate protective actions. However, the exact wording of the requirement suggests that the instrumentation will be capable of measuring "dose rate." The committee is not aware of any practical and durable instruments that can directly measure beta- and alpha-radiation dose rates in the field.

The U.S. Army should:

7. Develop requirements for measuring, interpreting, and responding to airborne and surface contamination (particularly that containing alpha and beta emitters). Guidance should define the levels of alpha and beta contamination that would trigger the use of protective equipment and actions. The ACE Directive gives only cursory consideration to this topic, and the terminology used to describe the instrumentation necessary for the detection and measurement of radioactive contamination is not clear.
8. Reconsider its absolute requirement that soldiers wear protective equipment within an exclusion zone as defined in the ACE Directive. The decision to use protective equipment should be based on the potential for personal contamination with radioactive materials, externally or internally. To require respiratory protection regardless of the existence of an airborne hazard may be counterproductive to completing the mission in a timely and effective manner.
9. Make a clear distinction between military intelligence threat estimates and radiation risk estimates. It is unclear in the intelligence procedures section of the ACE Directive (NATO, 1996, §1-3.a.) whether risk (high or low) refers to (1) intelligence assessments of the likelihood of radiation contamination or (2) the magnitude of measurable levels of radiation contamination.
10. Develop explicit requirements to define when individual radiation monitoring is required in the field. The guidance on whether a soldier could enter an area with low level contamination without individual dose monitoring is vague. It would be reasonable to require individual dosimetry for all incursions into an exclusion zone where radioactive contamination is likely.

The ACE Directive (NATO, 1996) defines a radiological hazard area as anywhere that the dose rate is in excess of 0.002 mGy/hour (0.0002 rad/hour). This dose rate is approximately 20 times the background radiation dose rate found in the United States (NCRP, 1987a) and 1/10 the maximum dose rate allowed for uncontrolled areas that members of the public might frequent. If a soldier were to spend a year in an area with such a dose rate—.002 mGy/hour (0.0002 rad/hour), a worst-case scenario—that soldier would accrue a dose of approximately 20 mGy (2 rad). That is equal to the ICRP-recommended annual dose limit for civilian radiation workers (ICRP, 1991a). Continuous exposure at this level would not exceed the current annual exposure limit of 50 mSv for U.S. radiation workers set by the Nuclear Regulatory Commission (CFR, 1991). Given its consistency with these comparison figures, the radiation dose rate at the edge of the exclusion zone is reasonable for defining contaminated areas and instituting radiation protection actions.

Current thinking in the U.S. Army requires that, in deciding to allow a soldier to be put at risk of exposure to radiation, a commander ensure that an accurate radiation dose is recorded to document that soldier's exposure (NATO, 1996). To do that, the commander must be able to determine an accurate dose for each individual soldier. The committee agrees with that requirement but finds its implementation problematic.

The available dosimeters may not be capable of providing adequate dosimetry. The IM-93 pocket dosimeter, currently fielded for individual soldiers, is not issued to all soldiers and is fragile and prone to error during rugged field use. The dosimeter planned for individual issue, the DT-236, is not sensitive below 100 mGy (10 rad). Thus, it cannot be used to differentiate between exposures in the low-dose categories specified in the ACE Directive (NATO, 1996). Dosimeters that can detect thermoluminescence can be used to monitor dose at low dose levels. These are available from the Army Dosimetry Center, but the equipment needed to read these devices is not normally available or issued to combat units. In summary, although individual dosimetry is, appropriately, required by the ACE Directive, it may be difficult to do with currently available hardware.⁵

The level of exposure at which dosimetry is recommended is stated in the Operational Exposure Guidance table of the ACE Directive. At 0.5 mGy (0.05 rad), the beginning of RES category labeled "IA," the commander is advised to "record individual dose readings [and] initiate periodic monitoring" (NATO, 1996, p. A-1). It is not clear what circumstances would lead to the start of dosimetry for individuals. If dosimetry for individuals has not yet begun, how is it determined that the 0.5-mGy (0.05-rad) level has been exceeded, triggering the start of periodic monitoring? One assumes that there are no dose histories, since monitoring has not yet begun. Therefore, the decision to start monitoring must be based upon projected whole-body doses of 0.5 mGy or more. Similarly, in civilian practice, the decision to issue individual dosimeters for monitoring can be made on the basis of projected doses. However, the ACE Directive requirement is considerably more stringent than that commonly followed in occupational programs and the rest of the U.S. Department of Defense (DoD) (DoDI, 1996).

DoD requires the monitoring of individual doses only when doses are above 5 mSv (0.5 rem), which is 10 times the level recommended by the ACE Directive (NATO, 1996).

There is an inconsistency between the text (NATO, 1996, § 1-3.f.(3)) and the table in Annex A. The text states that:

The table of Annex A, however, directs that monitoring begin at 0.5 mGy (0.05 rad). As a result of this ambiguity, it is conceivable that an individual could enter a zone with low level contamination without dosimetry. It is possible that this could result in exposure from unknown, localized hot spots that could cause the individual to receive doses above the monitoring threshold in Annex A.

Although it is understandable that the radiation community within the military might want to retain the familiar unit of absorbed dose, the rad, and rename it the centigray, the practice is not internationally accepted. The same may be said for the unit of effective dose, the centisievert, as a pseudonym for the rem.

Reported doses and particularly dose rates will probably be low. Reporting of doses and survey measurements in centigray, or centigray per hour, will require the use of very small numbers (e.g., the ACE Directive limit on a contaminated area of "0.0002 cGy/hour"). In handwritten transmissions of data, this could lead to errors in transcription (e.g., 0.0002 could be mistaken for 0.00002).

Although the ACE Directive (NATO, 1996) requires determination of external whole-body doses for individuals, there do not appear to be any requirements to identify or evaluate internal deposition of radionuclides or to estimate the radiation dose from such depositions. Neither is there a requirement to determine the potential for internal dose hazards in the area of operations. In fact, Note I of the table in Annex A to the ACE Directive (NATO, 1996, p. A-1) states that:

The ACE Directive recognizes the problem of internalized radioactive materials—soldiers are directed to put on their protective masks when they are in a "radiological hazard area" (NATO, 1996, p. 1-6)—but proceeds under the assumption that no such exposures will occur. The ACE Directive assumes that the respiratory protection is 100 percent effective and is silent on situations in which protective equipment is not worn or is defective. The ACE Directive does not specify, quantitatively, at what level of radiological contamination the protective mask should be worn.

The note in the table cited above implies that protective clothing and respirators are being used whenever any radioactivity above the background level is detected. ACE Directive paragraph 1-3.g.(1) requires that respiratory protection be worn in a "radiological area," but the area is not defined (§1-3.c.(2)(c)) by airborne radioactive contamination levels. It is conceivable that the wearing of the protective mask could be required when the actual concentration of radioactivity in the air is well within acceptable limits. This could happen if the radiological contamination was not easily resuspended or was fixed on the surfaces of military hardware that had been partially decontaminated. On a very hot day, the wearing of the protective mask under these conditions would unnecessarily diminish the performance of the soldier, thereby jeopardizing the mission, while perhaps also increasing the risk of other nonradiation hazards.

In addition to knowing the total dose accumulated by an individual, it is useful to know the time history of that exposure. The ACE Directive (NATO, 1996) enhancements to the Operational Exposure Guidance specify that dose reference levels are to be used with cumulative doses. However, the guidance does not specify whether doses are accumulated over an operation, a year, or a lifetime. It does not

appear that individual dose records indicating prior occupational and other exposures will be available to commanders when they are assigning RES categories.

The U.S. Army should:

11. Review its dosimetry capabilities and determine if they are adequate to support the use of the Operational Exposure Guidance in the ACE Directive. To manage soldier exposures according to the ACE Directive, all soldiers would have to have dosimeters that can measure doses as low as 0.5 mGy (0.05 rad).
12. Increase the specificity of the dosimetry program guidelines in subsequent versions of the ACE Directive (e.g., provide specific guidance on the capabilities of monitoring devices and equipment). The committee considers radiological monitoring and dose estimation for individuals, outside the occupational environment, as areas that require significant attention by the U.S. Army.
13. Not assume, as the ACE Directive does, that internal doses will be zero because respiratory protection will be used. Soldiers may receive an internal dose from inhalation or ingestion of radionuclides. This may occur if they are unaware of the airborne contamination and are not wearing protective equipment or if the equipment fails or is used improperly.
14. Review its capability to measure airborne radioactive contamination. The ability to measure airborne radioactivity and respond accordingly is essential to an adequate radiation protection program. The lack of exposure information for airborne hazards has proven to be a problem, as noted previously for the Atomic Veterans. More recently, potential chemical exposures during the Persian Gulf War at Kamisiyah, Iraq (DoD, 1996; Schafer, 1996), have demonstrated how a lack of airborne exposure data creates problems with health assessment activities.
15. Expand Operational Exposure Guidance to include radiation doses from both internal and external sources of radiation. These should be expressed in terms of effective dose and should be consistent with the requirements of the U.S. Nuclear Regulatory Commission. The lack of consideration of internal dose is a major shortcoming in the ACE Directive.
16. Adopt the millisievert as the standard unit of effective dose and the milligray as the unit of absorbed dose. There are three reasons for this recommendation. First, the units currently used in the ACE Directive—centigray and centisievert—are not internationally accepted scientific units. Second, by using millisievert, all doses to individuals can be compared to I year's nominal U.S. background dose from external sources (1 mSv). This should make it easier for

16. soldiers to understand their exposures.⁶ Third, at low radiation levels, the use of the unit millisievert will reduce, albeit only slightly, the problems of recording doses that are much less than I and that are expressed to several decimal places (e.g., 0.00002).
17. Clearly define the time over which doses are to be accumulated for assignment of radiation exposure state (RES) levels in the Operational Exposure Guidance in Annex A of the Directive. Presumably, doses are cumulative over a career and are not reset to zero after each operation.
18. Review and revise doctrine and procedures on dosimetry to ensure that individual doses are monitored and recorded for all soldiers exposed to radiation, whether from routine occupational exposure or as a consequence of uniquely military missions. Although the ACE Directive requires that records of individual dose be maintained, existing guidance (HQDA, 1994) requires tracking only of unit doses (e.g., average doses for a platoon).

The ACE Directive Operational Exposure Guidance table (Annex A [NATO, 1996]) subdivides the some-exposure category (RES-1; Table 4-1 in this report) of existing Operational Exposure Guidance (HQDA, 1994; NATO, 1986). Each level is accompanied by a narrative description of the risk corresponding to a dose range and by a series of required control actions.

The appropriateness of the dose categories depends largely on the way in which they will be used. These categories could be very useful and appropriate in controlling individual exposures and making future assignments. Such uses assume that individual dosimetry is available with the resolution and sensitivity of better than 0.5 mGy (0.05 rad; the width of the narrowest category). Without that, it will be impossible to resolve exposures into the lower RES categories.

If the table is intended for the planning of interventions in heavily contaminated areas, the fine detail in the lower categories may not be useful. It is not uncommon in nuclear accident areas (e.g., Chernobyl) to find wide variation in dose rates across small distances. Individuals could easily stray into hot spots where dose rates are significantly higher (e.g., by a factor of 10) than initial survey estimates would indicate. Without real-time, self-reading,⁷ individual dosimetry, it would be unreasonable to expect to control doses for all individuals in

the first two RES categories (0 to 0.5 mGy [0 to 0.05 rad] and 0.5 to 5 mGy [0.05 to 0.5 rad]).

The ACE Directive provides no indication of how unknown doses will be handled in the recording of individual doses or in the assignment of RES categories to units. In occupational radiation protection practice, it is normal to assign an administrative dose or to estimate a dose on the basis of the best available data.

At doses ranging from 0.5 to 5 cGy (5 to 50 mGy or 0.5 to 5 rad; RES Category 1B), the Operational Exposure Guidance recommends "establishing dose control measures as part of operations" (NATO, 1996, p. A-1). If one considers that the dose limit for the public used by the U.S. Nuclear Regulatory Commission until 1994 was 5 mSv, this level for beginning dose control might be appropriate. However, the current limit for public exposure is I mSv (CFR, 1991; ICRP, 1991a). In addition, the ACE Directive (NATO, 1996) itself institutes controls of radiation exposure beginning at 0.002 mGy/hour (0.0002 rad/hour). From this it would appear that some measures of control may be appropriate below the RES Category 1B level.

RES Category IC indicates that only priority tasks are to be attempted between 5 and 10 cGy (50 and 100 mGy or 5 and 10 rad). Priority tasks are defined as those required to avert danger to persons or to prevent damage from spreading. This level is comparable to Environmental Protection Agency (EPA, 1991) guidance that allows up to 100 mSv (10 rem) for similar tasks.⁸ It is also within the 500 mSv limit recommended by the National Council on Radiation Protection and Measurements (NCRP, 1993). In the next higher exposure categories—RES Category 1D (10 to 25 cGy [100 to 250 mGy or 10-25 rad]) and RES Category IE (25 to 70 cGy [250 to 700 mGy or 25 to 70 rad])—the ACE Directive limits missions to those that are necessary to save a life. The only difference between these two categories appears to be that the lower category is described as increased risk and the higher category is described as significant risk. The actions associated with them are the same. In emergencies, ICRP (1991b) recommends that every effort be made to keep doses below 1,000 mSv (100 rem) to prevent serious deterministic health effects (e.g., acute radiation sickness). The exposure levels in RES Categories 1D and 1E are in keeping with that guidance.

The U.S. Army should:

19. Include radiation doses from internal sources (e.g., from inhaled airborne radioactivity) in applying reference levels in Operational Exposure Guid-

20. 8	For comparisons in this paragraph the committee is assuming that the exposure is to gamma or x-ray radiation and that 1 mGy is approximately equivalent to 1 mSv ( rad ⋍ 1 rem).

19. ance. The reference levels shown in the ACE Directive Operational Exposure Guidance table (Annex A [NATO, 1996]) appear at least as stringent as those found in current civilian radiation protection recommendations of expert national and international advisory bodies. However, the ACE Directive misapplies the levels by assuming that there will be no internal doses.
20. Clearly specify what actions are recommended at each reference level in the Operational Exposure Guidance. Although the reference levels in the ACE Directive are generally appropriate, the actions recommended at each level lack specificity. Future versions of the ACE Directive or its implementing instructions should specify the details of each action (e.g., when to initiate a monitoring program and what its specific requirements are).
21. Restructure the table of Operational Exposure Guidance to account for the uncertainty of dose estimates in interventions. Because of this uncertainty, the two lowest dose categories in the existing guidance are too narrow to be scientifically justified (in the environment of an intervention) and should be combined.
22. Develop separate Operational Exposure Guidance for managing practices (routine tasks involving radiation exposure) in the context of a military operation. If the U.S. Army adopts the philosophy that soldiers should receive the same level of protection as civilian radiation workers in similar environments and circumstances, the guidance in Annex A (NATO, 1996) should be expanded to include dose limits and reference levels appropriate for a practice as well as an intervention.

The ACE Directive (NATO, 1996, §1-3.f.(3)) requires preparation and maintenance of individual medical records. Again, implementation is in question. Current U.S. Army doctrine for maintaining records during combat operations (HQDA, 1994) specifies that only the unit's radiation exposure state be transferred with the individual soldier. On the other hand, DoD requires that during peacetime individual doses be maintained (DoDI, 1996).

The ACE Directive (NATO, 1996, §1-3.f.(3)) requires commanders to ensure that the dose that a soldier receives is accurately recorded upon each radiological exposure and that the total dose is annotated in his or her individual medical record in accordance with national regulations. Also, the theater commander (NATO, 1996, § 1-3.f.(4)) is charged with ensuring "that the appropriate medical and NBC Cells [consisting of specialists in nuclear, biological, and chemical matters] are tasked to receive, monitor and maintain all radiological data in accordance with national regulations" (emphasis added). For U.S. soldiers, it is not clear whether that means in accordance with U.S. Nuclear Regulatory Commission guidelines or U.S. Army regulations. The committee assumes

that this refers to the Nuclear Regulatory Commission regulations. If that is the case, then internal doses must be documented along with external doses.

The ACE Directive does not specify exactly what dose-related data must be collected (e.g., internal dose, external dose, effective dose, or environmental data). Ultimately, it may be necessary to link this information from its repository to an individual for purposes of compensation determinations or epidemiologic study.

Although the first part of this chapter has focused specifically on the August 1996 draft of the ACE Directive (NATO, 1996), in this second section the committee broadens its discussion. The topic remains exposures of less than 700 mSv; and the task remains the presentation of the committee's evaluative findings on the basis of its integration of (1) information about civilian standard practices, (2) its understanding of current U.S. military practices, and (3) the process of committee deliberation that defines the Institute of Medicine-National Academy of Sciences approach.

Throughout the report, the committee discusses the topics of training, recordkeeping, and reporting in sequence. In a good radiation protection program all three must be intricately interwoven. Training should impart some basic understanding of radiation, communicate the risk, help the soldier to understand the ramifications of risk perception, and then place that knowledge in a context whereby the risks associated with radiation exposure can be compared with other non-radiation-related risks. The soldier then can draw upon this foundation to ( 1 ) protect himself or herself and others during an exposure situation, (2) know which pieces of information are important to obtain and record, (3) act to notify whomever should know about exposure or effects, and (4) use his or her own dose report to help guide his or her own future occupational, avocational, and health care activities. In addition, through training, the military attempts to teach commanders how to decide when it is appropriate to put subordinates at risk (justification) and how to do so to minimize short- and long-term harm while also achieving the military mission (optimization).

Therefore, training content includes conveying the value of information (e.g., records are important and notification of personnel is important) and the lesson that recordkeeping and reporting procedures are valuable only if the soldier knows (through training) what to measure and how to do so, what to record, and what to do with that information once it is recorded.

The common thread is communication. Accurate and appropriate information must be maintained so that it is available to be given to the right people at the right time. Furthermore, this communication must be carried out within an ethical framework in which the government seeks to meet its military objectives,

protect the health of military personnel, and take responsibility for the health consequences resulting from its decisions.

Information is vital to sustaining protection. When existing technology allows detection of radiation exposures, advance notice of radiation exposures is the goal. When feasible, radiation levels should be monitored in settings of suspected exposure. The levels of radiation that may involve short- or long-term risks need to be predetermined. Chains of command should be prepared to disseminate radiation warnings quickly and efficiently. If possible, soldiers should be equipped with devices to detect the levels of radiation in the operational field in cases in which significant radiation exposure is expected. They should not only know how to operate the devices, but should also understand how to interpret the readings that these devices provide.

In the military, choice is inherently constrained, and the nature of volunteering likely varies widely from situation to situation. The nature of military service has been used by U.S. courts as the primary rationale for denying service personnel the right to sue the U.S. military for injuries sustained while on active duty. In the leading case, Feres v. United States (1950), the U.S. Supreme Court ruled that military personnel may not sue the federal government for injuries sustained on active duty because (1) the Federal Torts Claims Act (FTCA; passed in 1948) does not provide for such lawsuits, (2) it would be unreasonable for the military to have to follow the liability laws of the various jurisdictions in which soldiers are posted, and local law would determine liability under FTCA, (3) the relationship between the soldier and the armed forces was "distinctively federal in nature," and (4) the Veterans Benefits Act provides a no-fault-based scheme to compensate veterans for service-connected and non-service-connected disabilities (Dalton, 1996).

In addition to these technical points of FTCA, the U.S. Supreme Court noted 4 years later in United States v. Brown (1954) that the Feres doctrine was based on the peculiar and special relationship of the soldier to his superiors, the effects of the maintenance of such suits on discipline, and the extreme results that might obtain if suits under FTCA were allowed for "negligent orders given or negligent acts committed in the course of military duty." Since Brown involved a claim by an honorably discharged soldier who was injured during knee surgery at a Veterans Administration (VA) hospital 6 years after discharge, the Court held that the Feres doctrine did not apply to him and that he could sue the VA for negligence.

Because contracting with a private military supplier creates a relationship that is as "distinctly federal in character" as the relationship between the government and its soldiers, soldiers may not sue private suppliers for defective products (Stencel Aero Engineering Corp. v. United States, 1977). Also, because of the need for strict military discipline, soldiers are barred from suing superior officers, even for violation of their constitutional rights (Chappell v. Wallace, 1983; Dalton, 1996).

For volunteering to be real, the soldier must be informed of the nature of the task and its risks and have the real option to decline to participate. Because the

U.S. military is also the employer of the soldier, the military has an independent obligation to the volunteer to minimize the risks as much as is reasonably possible. This can be done in a number of ways, including the use of planning, the use of protective equipment, and the exploration of less risky alternatives. Unlike the civilian, the military volunteer incurs no additional obligations by beginning a task: it is the nature of his or her initial agreement to perform the task and the nature of military service itself that would obligate the individual to complete the task to the best of his or her ability. Because the military knows that especially hazardous assignments will predictably occur and that volunteers will be sought for such assignments, the military has an added ethical obligation to plan for such occasions and minimize the risk of harm to the individual volunteers.

The committee emphasizes four overlapping purposes of training:

1. to address and fulfill ethical responsibility,
2. to address and fulfill legal responsibility,
3. to provide knowledge, and
4. to provide understanding.

train the commanders in both the radiation-related information and the risk communication techniques that they would need to inform their troops.

In addition to the requirement that DoD maintain radiation exposure data for all its personnel, the committee strongly recommends that each military member so exposed be provided annually and on termination of his or her service with a written document specifying the magnitude of each exposure (if possible) and the location(s) of such exposure(s) during his or her service. A copy of this information can then be made available to the U.S. Department of Veterans Affairs for future determination of the service connection of the disability and follow-up medical care if required. If possible, the exposure data notification document should include both a list of the agents to which the person was exposed and a general statement of the potential health consequences related to those exposures. The quality of the information provided will vary depending on whether the military operation was during war or peacetime, with more detail expected during peacetime activities.

Adequate recordkeeping of radiation exposures has two important ethical facets. First, recordkeeping requirements should respect, to the extent possible, the privacy of the individual and the confidentiality of that person's data. Individuals are entitled to know the purposes of data collection on radiation exposures, how this information will be used, those who may have access to the data, and the circumstances under which they are stored. Individuals should have access to their medical and exposure records and should be allowed to make corrections if warranted. The reliability of such data should be guaranteed by the military, with updating as necessary. Records should be kept secure from unauthorized users. Authorized access to records with personal identifiers, including individual medical records, should be limited to those who need access in the interests of the patient, certain types of epidemiologic research, or other justifiable uses. Even in such circumstances, the military should follow ethical standards of research by hewing to the federal rules laid out in the ''Common Rule" (CFR, 1993) or by developing its own set of policies and procedures for consent and other research ethics.

Second, recordkeeping requirements should further the interests of military personnel and the military. There are three primary reasons in support of systematic recordkeeping:

• mation may be important for documentation (for example, personal radiation monitoring data) should be based on sound medical and scientific findings.

• 2.	Cumulative recordkeeping: Comprehensive recordkeeping of incidences of radiation exposure over time provides the military with the means to track and reduce or prevent harmful exposures and subsequent health effects in the present and in the future.

• 3.	Population exposure: Information collected through comprehensive recordkeeping for those exposed to radiation could be highly beneficial in assessing potential harm to populations in the event that they are exposed.

Although records of radiation exposures may be kept for a variety of legitimate purposes, information should be collected pursuant to these objectives and not merely for the sake of having the information. No secret databases or uses of information should exist unless they are consistent with sensitive and ethical military objectives that require justified temporary nondisclosure. Declassification of secretly held information must be made as soon as possible. Procedures to determine legitimate uses of information should be standardized prospectively. Users who do not require information with personal identifiers should not have access to such information. Disclosures of such information, when authorized, should follow the least-intrusive disclosure principle. Disclosures must be the narrowest in content, must be the least identifiable and sensitive, and must go to the fewest number of persons as reasonably necessary to achieve a stated and justified objective. Information that has personal identifiers and that has been gathered for one purpose should not be disclosed for another, inconsistent, or secondary use without the consent of the individual. Although the dual goals of maintaining privacy and achieving comprehensive and accurate recordkeeping may seem incongruous, in fact, they can both be accommodated in a properly designed and implemented health information system.

The privacy of health-related information in military settings is in many ways distinct from that in the civilian sector. Military service explicitly and implicitly requires individuals to waive some of the privacy of their health information. Thus, for example, all military personnel can be required to undergo testing for drugs and sexually transmitted diseases as part of their agreement to serve in the military even though civilians cannot constitutionally be required to submit to such tests without some substantial justification. These test results become part of a military member's medical file, which may be circulated among perhaps thousands of people during the course of a career and afterward. Many military veterans use federal health care services through the U.S. Department of Veterans Affairs, to which copies of their military medical records may be forwarded.

Despite these and other exceptional waivers of privacy as part of military service, military personnel are entitled to some expectation of basic levels of accuracy, privacy, confidentiality, and security in the keeping of records of their exposures to radiation. To clarify these expectations, accuracy, privacy, confidentiality, and security in these contexts require definition. The accuracy of records means that the data that are collected should be complete, material, current, and correct. Health information privacy may be defined as an individual's claim to control the circumstances in which personally identifiable (versus anonymous or linkable) information is collected, used, stored, or transmitted. Confidentiality refers to privacy interests arising out of a specific relationship with the person about which information is gathered. In this context, a soldier may expect that a military physician whom he or she has seen for a medical condition will keep that information confidential, despite the dual fiduciary relationship of the physician to the patient and the physician's commanding officer. Security denotes the technological, organizational, or administrative processes designed to protect data systems from unauthorized access or unwarranted disclosures, modification, or destruction (Gostin, 1995, 1997; Gostin et al., 1996).

Consistent with these definitions, even the most secure system of military medical record management cannot maintain the privacy of records because no collection of information is free from unauthorized access. Although privacy expectations arise, in part, from the ethical principle of autonomy, they are not in any sense absolute. Medical records, by their nature, are created to be shared with others. Health information is lawfully exchanged among numerous parties, regardless of an individual's claim to control the circumstances in which it is transmitted.⁹ In the military individual interests in health information privacy must be balanced against the individual's own interests in comprehensive and accurate recordkeeping, as well as the competing interests of the military and clinicians in information concerning radiation exposure. The result in military settings is a privacy trade-off between the privacy of the medical records of military personnel and the communal defense-oriented interests of the military.