consumption of apple juice as reported in the 1977–1978 Nationwide Food Consumption Survey (NFCS) (Murdoch et al., 1992). The average intake of apple juice for survey respondents is indicated by open circles; the solid lines represent smooth curves fitted to the raw data. On a body weight basis, 1 year-old children consume more than 30 times as much apple juice per day as adults.
The data for the three other foods depicted in Figure 8-1 reflect different consumption patterns with age. These particular foods were chosen to illustrate a variety of food consumption patterns. The maximum consumption of potatoes and tomatoes occurs among children 1 to 6 years of age. The consumption of lettuce increases throughout childhood and is higher among adults than among children.
These consumption data have important implications for assessing the risk of carcinogenesis in infants and children. In laboratory studies of carcinogenicity in which the dose is expressed relative to body weight (in mg/kg bw/day), the dose is generally held constant throughout the study. Such studies are used to estimate carcinogenic potency in terms of a unit risk factor (q1*). Multiplication of the anticipated level of exposure by the unit risk factor leads to an estimate of cancer risk.
Methods for estimating lifetime cancer risk when the level exposure varies over time have been investigated by Crump and Howe (1984) and Kodell et al. (1987). One simple approach is to calculate a lifetime average daily dose (LADD) by distributing the cumulative lifetime dose equally over a lifetime. This amortization of total lifetime exposure is suggested by the EPA (1986) in the absence of evidence of dose-rate effects. This method is especially appealing when considering carcinogens that remain in the body permanently or have very long half lives, such as some fat-soluble organic compounds and heavy metals.
Murdoch and Krewski (1988) showed that except in very special cases, the use of the LADD will not lead to the same lifetime risk as the actual time-dependent exposure pattern. These authors defined a lifetime equivalent constant dose (LECD) in terms of the relative effectiveness of dosing at different ages, which leads to the same lifetime risk as the time-dependent dosing pattern. (The LECD is a weighted average of the time-dependent dose; weights are proportional to the relative effectiveness of dosing at each point.) The ratio C = LECD/LADD then provides a measure of error of risk estimates based on the LADD.
Murdoch et al. (1992) used this measure to investigate the extent risk estimates based on average adult levels of exposure may underestimate the risks for infants and children as a consequence of their higher exposures on