Comments on Epidemiologic Studies that NCRP Commentary 27 Claimed to Provide Moderate, or Weak-to-Moderate, Support for LNT Model
| Study | Comments |
| Mayak nuclear workers (62) | For radiation doses less than 0.5 Gy, ERR is consistent with no increased cancer risk (Fig. 2 of the publication). |
| Chernobyl fallout, Ukraine and Belarus thyroid cancer (63) | Like Massachusetts tuberculosis study (53), this study has flaw of considering thyroid cancer incidence, which is subject to large overdiagnosis, and considering thyroid cancer only. |
| Breast cancer studies, after childhood exposure (64) | According to the publication, 3% of women in whole cohort had breast doses exceeding 1 Gy, and mean dose was 0.18 Gy. Thus, cohort included patients with highly carcinogenic doses, even though average dose was much lower. Increased cancers would occur in patients with high breast doses and should not be attributed to low radiation doses. Dose–response data from earlier publication on same cohort (65) show no significantly increased breast cancer risk for breast doses less than 1 Gy (Table 3 of the publication). |
| In utero exposure, Japan atomic bombs (57) | Data shown in Table 4 of the publication indicate no significant increase in relative risk of cancer for low-dose category (0.005–0.2 Sv) for in utero or childhood exposure. Next dose range (>0.2 Sv) shows significantly increased relative risk for a few categories. Because this dose range includes high doses, increased relative risk for this dose range may not be attributable to low radiation doses. A review of a large number of studies (57,66) indicated that no definitive conclusion can be drawn on carcinogenic effect of in utero exposure. |
| Techa River, nearby residents (67) | Data in Table 3 of the publication indicate no significant increase in cancer risk for any dose range. |
| In utero exposure, medical (68) | This is a review article. A later review of a large number of studies (66) indicated that no definitive conclusion can be drawn on carcinogenic effect of in utero exposure. |
| Japan nuclear workers (69) | The publication stated that data are likely confounded by alcohol drinking. Data in Table 2 of the publication indicate that this study does not show significantly increased cancer risk in radiation workers when alcohol-related cancers are excluded. |
| Chernobyl cleanup workers, Russia (70) | SMR for all cancers is 0.95 (95% CI, 0.92–0.99) (from Fig. 5 of the publication), indicating reduction of cancer in this cohort after radiation exposure. |
| U.S. radiologic technologists (71,72) | Overall cancer mortality rate for radiologic technologists was significantly lower, with SMR of 0.82 (95% CI, 0.80–0.84), as reported in Table A2 of supplementary materials of one publication (71). The other publication (72) studied breast cancer only. Studying single type of cancer does not provide complete information on carcinogenic effect of low radiation doses, as hormetic reductions of other cancers would not be considered. |
| Mound nuclear workers (73) | Overall cancer mortality rate was significantly lower in radiation workers, with SMR of 0.86 (95% CI, 0.79–0.93). |
| Rocketdyne nuclear workers (74) | Overall cancer mortality rate was significantly lower in radiation workers, with SMR of 0.88 (95% CI, 0.81–0.94), and relative risk for all cancer mortality did not significantly increase. |
| French uranium processing workers (75) | Overall cancer mortality rate was significantly lower in radiation workers, with SMR of 0.79 (95% CI, 0.72–0.87). |
| Medical x-ray workers, China (76) | Data below 0.15 Gy are consistent with no increase in cancer risk (Fig. 2 of the publication). Shape of dose–response curve cannot be determined reliably from these data because of large errors. |
| Taiwan radiocontaminated buildings, residents (17) | As explained in the publication (19), Standardized incidence ratios for all cancers, calculated using data in the publication, are significantly reduced in irradiated residents of radiocontaminated buildings. |
| Background radiation levels and childhood leukemia (77) | Breastfeeding and child-care attendance are known to be important factors in childhood leukemia (78,79) but were not considered confounding factors in the study. Because relative risk per millisievert for leukemia was 1.07 (95% CI, 1.01–1.13), small changes in these confounding factors could make increase in risk not significant. We should await better studies that account for such major confounding factors. |