synthesis of a variety of information sources, because information on locations, operations, and activities may be combined to define relative differences in exposure.
AVAILABLE DATA
There are sources of publicly available data on the use of herbicides in Vietnam, veterans' service in the Vietnam theater, and characteristics of other populations exposed to dioxin that may be of use to researchers. Some of these sources are mentioned below or are cited in the Appendix. However, their mention here is for information only and is not intended as a recommendation for their use. Neither should this listing be considered comprehensive. Data regarding individuals may be subject to privacy restrictions.
The HERBS and Service HERBS computer tapes contain information regarding aerial and some ground-spraying activities, including date, type of herbicide, number of gallons sprayed, and location of spraying. These tapes and their limitations are discussed in greater detail in the Appendix and in Veterans and Agent Orange (IOM, 1994:96–106). Data from the 1987 Air Force Health Study physical examination of Ranch Hands veterans and a comparison group is available through the National Technical Information Service (NTIS Accession Number A-283-319). These data include date of birth, exposure group, and dioxin levels for approximately 2,500 subjects. The Minnesota Department of Veterans Affairs has compiled service information on approximately 156,000 Vietnam-era veterans—including approximately 68,000 who served in Vietnam—in support of that state's bonus program. The DVA is developing a comprehensive Vietnam veterans roster that will include service branch, rank, occupational codes, and the dates of Vietnam service. This roster is expected to be completed during 1997. The National Institute of Occupational Safety and Health has collected information on 5,172 male workers at 12 plants that produced TCDD-contaminated herbicides. Among these data are vital statistics and duration of exposure; lipid-adjusted, serum TCDD levels are also available for a subset of 253 workers.
POTENTIAL APPROACHES
Exposure reconstruction should consider methods that will identify homogeneously exposed groups, rank-order cohorts by exposure levels, or use some combination of approaches to generate information for epidemiologic studies. This could include, but is not limited to, using biomarkers of exposure; identifying exposure patterns (e.g., from air vs. ground spraying), locations, concentrations, and frequencies; considering types of foliage, weather conditions, troop location and movement in time; date(s) of field service for