Lifetime attributable risk

Einstein et al. (2007) estimate a lifetime attributable risk of cancer incidence (EAR) based on the BEIR VII report (Committee to Assess Health Risks from Exposure to Low Levels of Ionizing Radiation 2006) for 64-sUce CTCA with Monte Carlo simulations of male and female patients. According to these estimations, the LAR of a 20-year old woman equals 1 in 143, while an 80-year-old man only suffers a LAR of 1 in 3,261. The estimated LAR of a 60-year-old woman and man were 1 in 466 and 1 in 1,241, respectively, which resembles that of more typical patients for CTCA. The use of ECTCM reduced this LAR to 1 in 715 and 1 in 1,911, respectively. 

Lifetime excess risks, also referred to as lifetime attributable risks (Kellerer et al. 2001, BEIR VII 2006), LAR, are calculated by means of site-specific risk models, site-specific baseline rates on cancer incidence (or cancer mortality), and life-table data to account for competing risks ... 

Table 7.3. Cmiiulative organ doses due to four CT screening scenarios (see Table 7.2 and text) and corresponding estimates of lifetime attributable risk (LAR) to incur cancer/leukemia for those sites contributing most to total LAR...

Radioactive slope factors calculated by EPA s Office of Radiation and Indoor Air (ORIA). Slope factors are central estimates in a linear model of the age-averaged, lifetime attributable radiation cancer incidence (fatal and nonfatal cancer) risk per unit of activity ingested, expressed as risk per picocurie (pCi). 

Fig. 7.4. Lifetime excess risk at 1 Sv organ dose in dependence on age at exposure for those cancer sites attributing most to the total excess lifetime risk according to BEIR VII risk models and German life tables and German cancer incidence rates...

Only in a few instances have the exposures been sufficiently marked to allow attribution, as in the case of a heavily exposed survivor of the A-bombings in Japan who subsequently developed leukemia. From the magnitude of the survivor s radiation exposme and fiwm epidemiological data on the incidence of leukemia in the population at the same exposure level, we might infer that the survivor s increase in risk corresponded to a 2 percent lifetime incidence, in contrast to the normal lifetime leukemia incidence of about 0.5 percent. The attribution would then be 2/2.5, or 80 percent, leaving 20 percent for background radiation and other causes. 

If the exposure had been much smaller, the risk calculation would have been less direct and less certain. For purposes of risk reduction in public health, we may choose to err on the pessimistic side in risk estimations. For purposes of attribution, however, we want to make best estimates. Most of the numbers in Ikble 8.4 are overestimates of the risks. For radiation-induced leukemia, as described in Section 6.1.2, the best dose-incidence model might be lineai>quadratic and not linear. Thus, someone exposed to 50 mSv (5 rem) might be considered, on a linear extrapolation basis, to have a radiation related lifetime risk of cancer mortality of 10 (2 x 10 Sv 2 x 10 rem ), or a lifetime risk of mortality from leukemia of approximately 1.5 x 10 (0.3 x 10" Sv 0.3 X 10 rem ). The natural lifetime risk of mortality from leukemia other than chronic lymphocytic leukemia is approximately 56 x 10 . Therefore, the percent attribution to radiation according to the linear model would be ... 

The World Health Organization estimates that over 3 million deaths a year can be attributed to overweight and obesity, a figure that is predicted to increase (WHO, 2002). Obesity increased 74% between 1991 and 2001 in the U.S., and steep increases like this are now occurring elsewhere. The rapid rise in obesity has made it one of the greatest risks to human health worldwide. Of particular concern is the obesity rate among children, which increases disease risk throughout their lifetime. 

There is no evidence to date of any increase in the incidence of any malignancies other than thyroid carcinoma or of any hereditary effects attributable to radiation exposure caused by the Chernobyl accident. This conclusion, surprising for some observers, is in accordance with the relatively small whole body doses incurred by the populations exposed to the radioactive material released. The lifetime doses expected to be incurred by these populations are also small. In fact, the risks of radiation-induced malignancies and hereditary effects are extremely small at low radiation doses and, as the normal incidences of these effects in people are relatively high, it is not surprising that no effects could be detected. 

As expected, the above results are significantly higher for the PEL case. The 50-fold factor difference can be primarily attributed to the pollutant concentration (PC) employed in both calculations the OSHA PEL is based, to an extent, on workplace conditions rather than ambient conditions in the atmosphere, which can be orders of magnitude lower. Regarding an individual s lifetime risk, the workman s risk is approximately 1 in 1000 or 10 over his/her lifetime, as compared to approximately 40 in a million or 40 x 10 for an average citizen. 

Calculation of Lifetime Excess Cancer Risk and Hazard Indices for Chronic Exposure. For each exposure estimate, it is extremely easy to generate corresponding risk estimates. Thus, the user can quickly specify the full range of risks that may reasonably be attributed to a site. If reasonable worst-case or worst-case assumptions indicate minimal risk, it may be possible to defer the action until more pressing problems are addressed. Alternately, the Risk Assistant analyses may indicate current or potential risks that should be immediately ameliorated. When a wide range of risks may apply to a site, it may be important to conduct additional studies to reduce the uncertainties associated with a site. 

---