criticized conventional categories used to classify populations in the United States. Indeed, even the Office of Management and Budget has noted that

these classifications should not be interpreted as being scientific or anthropological in nature, nor should they be viewed as determinants of eligibility for participation in any Federal program. They have been developed in response to needs expressed by both the executive branch and the Congress to provide for the collection and use of compatible, non-duplicated, exchangeable racial and ethnic data by Federal agencies.2

The workshop attendees agreed that there are multiple subcategories of Hispanics/Latinos. These include countries of origin from Spain to Mexico to Argentina, and length of residency from recent arrivals to multiple generations in the United States These groups and subgroups do not constitute a “racial” or even an “ethnic” group. As described in the sources mentioned above, there are variations in language, behavior, and other cultural aspects within and among these groups, as the groups and the individuals in them have different cultural experiences or cultural processes that shape and influence them (NRC, 2002). The workshop participants noted the change that occurred between the 1990 and 2000 censuses to allow designation of multiple racial groups may affect comparisons with non-Hispanic black and non-Hispanic white populations but should not affect the fraction of Hispanics of any race compared with non-Hispanics of any race.

The workshop participants agreed that a sensible public health agenda would give priority to addressing high-risk groups of workers. While absolute numbers of workers in various occupations or industry sectors should be given due consideration, groups with high fatality rates were deemed the highest-priority target populations. Defining such priorities using existing datasets is challenging. Using the Census of Fatal Occupational Injuries combined with denominator employment information from the Current Population Survey, it is possible to calculate fatality rates of Hispanics and non-Hispanics in various occupations and industry sectors. Non-fatal injury rates among Hispanics by occupation or industry can also be calculated. This provides valuable targeting information for defining high-priority occupations and industries with elevated Hispanic fatality or injury rates. During the workshop several cautions when using such data to target health education initiatives were discussed. This included concerns about the statistical validity of fatality rates in very small populations. If there are only two Hispanic elephant trainers and one dies in a given year, the calculated rate may be very high but may not reflect real relative risk of the occupation or be statistically meaningful. Similarly, if there are no fatalities in a given year, this may not accurately reflect risk. Therefore, strictly targeting the highest-risk or highest-relative-risk occupations or industry sectors could lead to misallocation of effort. The workshop participants agreed that it would be more appropriate to focus priorities on larger high-risk occupations or industry sectors where sample sizes are adequate to provide reliable results.


The self-employed are exempt from most occupational safety and health regulations, including fatality, injury, and illness data collection and reporting requirements. As a result the Occupational Safety and Health Administration (OSHA) reports far fewer occupational fatalities than the Bureau of Labor Statistics’ Census of Fatal Occupational Injuries, since the Census of


See <>. Date accessed November 5, 2002.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement