population, and that the calculations listed below are fairly robust to competing risks. It is possible that the attributable fractions for several risks will sum to more than 1 (i.e., more than 100 percent).

Finally, the risk estimates should not be generalized to the entire population when they are calculated based on studies of a segment of the population, such as children. The reason for this is that measures of risk may systematically change over time. As noted above, the prevalence of skin test reactivity is lower in school children than in young adults.

Studies that are used to calculate attributable fractions should meet several criteria. For example, the exposed and unexposed populations should be alike with respect to demographic variables that could affect disease rates. The study instruments that are used should be the same, and preferably the exposed and unexposed populations should have been studied by the same team.

Finally, calculation of attributable fractions should be specified as referring to the entire population or to only the exposed population.

According to Last (1986) the population attributable risk is calculated as:

where AFp, the attributable fraction (population), also called the population attributable risk, is the proportion of the disease that can be attributed to exposure to the risk factor; Ip is the incidence in the population; and Iu is the incidence in the unexposed population.

The attributable fraction in the exposed population is calculated as:

where AFe is the attributable fraction (exposed), i.e., the proportion of cases among the exposed population that can be attributed to exposure to the risk factor of interest; Ie is the incidence rate among the exposed group; and Iu is the incidence rate among the unexposed group.

Table 2-6 presents some estimates of attributable fraction using data from several published studies.


Recent surveys have found prevalence rates of asthma of 8 to 12 percent in the U.S. population—an estimated 20–30 million people (Burrows et al., 1989; Mak et al., 1982; Pattemore et al., 1990; Weiss et al., 1992a). As yet unpublished data suggest that 10 percent of children had been prescribed an inhaler for the treatment of asthma (Platts-Mills, personal communication; Sporik et al., in press). The National Asthma Education Program

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