HEALTH CONSULTATION

Review of Health Studies Relevant to Lawrence Livermore National Laboratory and the Surrounding Community

LAWRENCE LIVERMORE NATIONAL LABORATORY, MAIN SITE (USDOE)
LIVERMORE, ALAMEDA COUNTY, CALIFORNIA

APPENDIX A: FIGURES

Figure 1: Map of Lawrence Livermore National Laboratory, Livermore, California

Figure 2: Melanoma cases among LLNL employees (reproduced from journal article, "Case-Control Study of Melanoma at LLNL") (3)

APPENDIX B: TABLES

APPENDIX C: RESPONSE TO COMMENTS ON PUBLIC COMMENT DRAFT OF THE HEALTH CONSULTATION, "REVIEW OF HEALTH STUDIES RELEVANT TO LLNL AND THE SURROUNDING COMMUNITY, LIVERMORE, CALIFORNIA"

Many of the comments fell in several main areas, so we have written a general response to each area that addresses a number of specific points.

1. Methodology of LLNL case control study
2. Evidence for melanoma excess vs. surveillance bias/early detection
3. Evidence for association between melanoma and radiation:

- within LLNL studies
- within the scientific literature.

1) Methodology of the LLNL case-control study

Overmatching

Matching as a technique to control confounding seems appealing, but has a number of scientific disadvantages. These disadvantages, plus the more recent development of alternative analytical techniques to control confounding, have reduced the desirability of matching. A major limitation of matching is the inability to evaluate the effect of a factor that has been matched on the risk of the outcome. In the situation of LLNL, if start date is closely associated with exposure, then matching on it prevents the investigation of the effect of exposure on the outcome. The fact that the exposure factors in the Austin/Reynolds study are all temporally related to LLNL activities is actually an argument against using it as a matching factor, rather than a reason for matching on this. This would make it a factor that should be investigated and explored.

Although laboratory activities may have changed over time, that does not necessarily introduce a bias. The controls are not being selected from the "later" employees, but they would be randomly selected from all time periods covered by case accrual, so there would be no reason for this selection to produce bias. This would allow investigators to evaluate whether activities that were associated with greater exposures (and may also have tended to take place earlier), were more prevalent than those that occurred later. This cannot be investigated if start date is used as a match.

[Moreover, if the exposure that caused melanoma stems in part of primarily from contaminant releases to the community that occurred at specific points in time, then matching on start date would over-control for this effect in a way that would prevent it from being discovered. For example, if the melanoma were caused by tritium releases to the community during the late 1960s and 1970s, then this exposure would affect both employees and the community.]

In our view, it seems the "safest" approach is to select a method that allows full exploration of the possible exposures of interest, and that would mean not matching on start date.

Use of radiation measurements

LLNL of course had access to the radiation data, although there are questions regarding its accuracy. However, actual radiation doses were never directly used in the analysis. Dosimetry records also included recorded doses of ionizing radiation received by employees prior to employment at LLNL. Doses were then coded on a scale of 0-5 and, and statistical tests were conducted "to remove the effects of skewing" (presumably the distribution of the radiation levels did not follow the normal, or "bell" curve). They presumably conducted a non-parametric statistical test, used for non-normally distributed data. The authors calculated the average dose (based on this categorical scale) for cases vs. controls, finding the two to be similar. They did not describe the distribution of radiation dose levels.

Also, the authors do not include radiation dose in any multivariable analysis, which would allow for simultaneous control of other factors. Although it seems that the cases and controls in this study group are quite comparable in terms of their exposures, a multivariable analysis would be necessary to answer this question more accurately.

LLNL comments that "subsequent analysis of the radiation badge data from the 1984 cases did demonstrate recall bias." It is unclear what this statement is based on. However, the Schwartzbaum re-analysis of the data, which included evaluation of the dosimetry data, specifically examined the possibility of whether recall bias accounted for the findings, finding no evidence of bias.

Matching algorithm

We contacted the current LLNL Medical Director, Dr. James Seward, to find out what the original algorithm was and how and why it was changed. Subsequently, Dr. Seward replied that he contacted as many authors of the report as possible, but none could recall the original algorithm nor why it was changed, but that all persons queried agreed it was a minor change. Presently, we are unaware if a written record exists documenting any of these decisions.

2) Evidence for melanoma excess vs. surveillance bias/early detection

Melanoma rates were consistently elevated, across different time periods, using different reference groups, and among males and females. The consistency of this result supports the likelihood that there was a genuine excess of melanoma among the LLNL workforce and Livermore community that was not merely an artifact of early detection. (See Tables 3 and 4 of melanoma rates among different study populations and Figure 2, which demonstrates an excess of melanoma compared to expected rates, which is not limited to thinner tumors.)

The articles reviewed offer mixed and qualified evidence, but tend to support a limited role for surveillance bias, rather than the main reason for the excess. The findings are summarized below.

Hiatt 1986: there could be either physician diagnosis bias, or an environmental agent (or agents) responsible.
Hiatt 1993: there could be surveillance bias up to 1976, but data do not support this after that time.
Schneider 1990: thinner tumors at LLNL than a community laboratory, but no difference in the earlier period around 1976.
Schneider 1987: rate of melanoma increased at LLNL increased, as did the proportion of thin tumors; authors also note that a high percentage of in-situ cases is consistent with a high overall incidence.
Gong 1992: excess thin but not thick tumors; promoting the idea that many early-detected melanomas would have spontaneously regressed.
Joint Legislative Audit Committee: high melanoma rate appeared to be correct.
DOE Panel: rate of melanoma appeared to be 3-4 times higher than general population.
LLNL-commissioned panel: early detection was an unlikely explanation for excess, as there was no decrease in later time periods.

Early detection usually would result in more tumors in the immediate time period in which they were detected, but fewer tumors in later time periods, as they would have been discovered and counted while they were thinner. However, there was no decrease in the numbers of tumors detected in subsequent study periods. In order for this second hypothesis still to be true, it would mean that the excess number of thin tumors would have disappeared on their own. Spontaneous regression is a rare phenomenon, and thus early detection seems unlikely to explain the excess.

[Even if surveillance bias were a reason for the elevation among LLNL workers, that would be unlikely to explain the excess melanoma which was seen among children and young adults in the Livermore community for the three decades from 1960-1990.]

3a) Evidence for association between melanoma and radiation at LLNL

Our findings that melanoma may be associated with LLNL-related exposures such as ionizing radiation are well-supported by the studies of LLNL.

The association between occupational risk factors and melanoma is based on many analyses, in many models. Risk ratios varied, with most showing statistical significance:

1. work with ionizing radiation: Odds Ration (OR) = 3.7 (95% CI = 1.6 - 8.6) (50);
2. radioactive materials (ever vs. never exposed): OR = 2.8 (1.1-7.1) (50);
3. radiation-occupations: summary OR = 10.8 (95% CI: 1.4-85.1) (52).

Schwartzbaum et al recalculated the odds ratio estimates for reported employment in proximity to radiation for 588 models containing different sets of confounding variables, finding the odds ratios ranged from 3.0 to 46.4 (all statistically significant). The median odds ratio and odds ratio adjusted for basic confounders were both 5.4.

Furthermore, the Schwartzbaum paper was based on an extensive review and re-analysis of the data that specifically attempted to determine whether there were sources of error or bias that could have falsely led to the association between radiation exposure and melanoma. They concluded: "No definitive evidence for influential matched sets, confounding, recall bias, or any other sources of bias was detected. These results suggest that the odds ratio for reported employment in proximity to radiation may be valid."

3b) Evidence for association between melanoma and ionizing radiation in the scientific literature and biological plausibility

Most cancers can be caused by ionizing radiation. The body of epidemiology between ionizing radiation and melanoma is mixed, but there continues to be new data supporting a possible link between radiation and melanoma. We support a careful and specific investigation of this potential association, including development of new studies that are better designed to address this question, as well as consideration of what previous studies have had null findings and which have had positive findings, and why they may differ. (Biological plausibility and specific exposure pathways in the Livermore community are further addressed in the report.)

The relative rareness of melanoma may have hindered its identification as a radiation-related cancer. In 2002, approximately 900,000 Americans were diagnosed with basal cell skin cancer, another 300,000 with squamous cell skin cancer, but only 53,000 with melanoma (26). Studies finding a small increase in skin cancers may not be able to detect an increase in melanoma specifically. Because of this, fairly large study populations will be required to investigate this issue adequately.

Some cancers will be caused more easily by radiation than others. Also, some cancers have a high natural incidence but low relative risks per radiation dose (e.g., colon); some have a low natural incidence but a high relative risk per radiation dose (e.g., thyroid); and other cancers of high natural incidence also have a high relative risk per radiation dose (e.g., breast). Thus, it is possible that melanoma could be caused by ionizing radiation, but perhaps it was not noted originally because it may not be as radiosensitive as other cancers.

In addition to the issue of melanoma's relative rareness (and probably often as a result of this), studies of nuclear workers have often combined melanoma and non-melanoma, limiting the available information on melanoma specifically. Studies with negative results could be partly due to the healthy worker effect (persons who work tend to be healthier than the general population). However, it may be that more sensitive tests such as dose-response trend tests are required to detect a relationship between melanoma and ionizing radiation. For example, a large study of nuclear workers was cited as negative for a relationship between melanoma and radiation in a reference book on cancer (64), but a more sensitive analysis found a dose-response relationship of 1.5 excess relative risk per Sievert of exposure for mortality from skin cancer (types combined).

Also, studies of survivors of the atomic bomb in Japan have generally been seen as supporting a relationship only between radiation and non-melanoma skin cancer. However, Japanese have a lower baseline risk for melanoma because of their skin type, and it is possible that an association exists that was not detected originally. A recent article by the NCI states that the excess relative risk point-estimates for melanoma were large but statistically non-significant, as the number of cases was small (n=10) (65). As mentioned earlier, because melanoma is much less common than basal and squamous cell cancers, it would be predicted that melanoma would be found much less frequently. Detecting a less common event is more difficult statistically.

The data in BEIR V are consistent with this interpretation (32). The BEIR V reports that skin cancers were among the first cancers to be recognized as radiation-induced. In general, however, the studies cited do not provide enough information on melanoma separately to evaluate it. One of three studies cited found excess skin cancer, in a combination of six basal cell carcinomas and two melanomas. Another noted excess skin cancer, reported to be primarily basal cell carcinomas. Thus although there may have been fewer melanomas in absolute numbers, because melanoma incidence was not provided separately (probably due to few cases), it cannot be ruled out that melanoma incidence is not also increased along with other skin cancers.

Some relevant evidence has been noted in the following studies. Larger studies appear more likely to detect an effect.

Occupational ionizing radiation exposure:

1. Cohort of 191,333 Canadian radiation workers; statistically significant elevations (66);
2. Cohort of 68,588 American radiologic technologists; Relative Risk (RR)=1.8 (95% CI = 0.6 - 5.5); among those who worked for five or more years before 1950, RR = 2.4 (95% CI = 0.7-8.7) (67);
3. British radiologists (1897-1979); elevated in some periods, but not statistically significant (68);
4. Occupational exposure, significant elevation (24); and
5. Chemists exposed to ionizing radiation, significant elevation (69).

Pilots and flight attendants (gamma ray exposure):

1. Meta-analysis of cancer incidence and flight personnel; statistically significant (RR=1.97) (70);
2. Elevated, statistically significant (71);
3. Elevated, near statistical significance (72);
4. Elevated, not statistically significant (73); and
5. Elevated, near statistically significance (74).

Radon exposure:

1. Review article stating evidence for increased risk for cancers other than lung from radon exposure, including melanoma (75);
2. Radon levels correlated with melanoma in southwest England; statistically significant correlation (76); and
3. Melanoma significantly correlated with radon exposure in home (77).

APPENDIX D: BACKGROUND INFORMATION ON IONIZING RADIATION

Ionizing radiation means that radioactive particles and rays encounter materials and carry enough energy that they strip away electrons from atoms and molecules that normally are electrically balanced (10). The stripped atom or molecule becomes an ion, and ions formed in living tissue by external causes can be hazardous to that tissue's health (10). The main types of ionizing radiation are alpha, beta, and gamma. Alpha rays are charged particles that have a lot of energy of motion, but do not have as much penetrating power as other types of radiation (10). Alpha particles deposit their energy in a small volume; in biological tissue, the range is approximately 25-80 um (micrometers) (10). Studies of diffusion of plutonium-239 through the various skin layers found that most of the radioactivity was found in the epidermis (about 2-4% penetrated as far as the deeper dermal layer) (35). Once inside the human body, an alpha particle can be much more damaging to tissue than beta or gamma radiation (10). Plutonium is an alpha emitter (plutonium-239 also emits gamma rays). Radioactive materials can also emit beta particles, which are high-energy electrons. Beta rays (these are also particles) are lighter and more penetrating than alpha particles. Gamma rays are a type of light (energy), not a particle. They travel at the speed of light, have no mass but tremendous penetrating power, and can pass through the body. Nuclear explosions are a source of gamma rays.

Tritium is a beta emitter. It is a radioactive isotope of hydrogen, and, once released to the atmosphere, it can be taken up by all hydrogen-containing molecules and distributed widely through the environment (78). In the air, it binds with oxygen to form radioactive water molecules, forming tritiated water. Tritiated water or vapor may readily be absorbed into the body through the skin, inhalation, or ingestion (10,34). Contact between skin and surfaces contaminated with tritium can result in transfer of tritium to cells within the skin (79,80). For example, wearers of watches with tritium as a permanent light source on watch dials had increased tritium skin doses and increased levels of excreted tritium in urine (79). An investigation of tritium levels among residents near a heavy water (tritium) reactor facility found comparable tritium levels in urine to tritium levels in precipitation near these locations (81), indicating how tritium disperses through water inside and outside the body.

APPENDIX E: BACKGROUND INFORMATION ON CANCER

Frequency of Cancer

Although incidence of and mortality from many cancers have been decreasing in California, cancer overall is still the second leading cause of death (13). According to current rates, one of every two men and two of every five women will develop some form of cancer in their lifetimes (13). The older someone becomes, the more likely it is that he or she will eventually get cancer. Breast cancer is the most common cancer among women, and prostate cancer is the most common cancer among men.

Lung cancer is the second most common cancer for both men and women (13). According to current rates, 1 of 12 men will develop lung cancer, but this risk is 10 to 20 times higher among smokers. It is estimated that 85% of lung cancer is caused by cigarette smoking (not including exposure to smoke from others).

Causes of Cancer

When and where cancer occurs depends partly on chance. Information is, however, available about how cancer develops. Cancer is a slow process, generally requiring years or even decades. The process of cancer can be started by a causative agent, which might be one factor among many. These agents could include radiation (28), chemical exposures (82,83), viruses (84), or possibly bacteria (85,86). But there must also be a pathway by which this agent can reach someone. This could be through the air, through water or food, or through sexual transmission. Additionally, the agent has to exert an effect that causes a biological change in the person's body. The exposure must also last long enough for harm to occur.

There are many different types of cancers, and different cancers have different causes. For example, in the body there are natural hormones that play a role in whether someone gets cancer (87,88). For example, breast cancer is related to hormonal functions such as early onset of menstruation and when a woman bears children (89,87,88). Exposure to sunlight is a factor in skin cancer (90). Drinking alcohol can be the cause of some cancers, such as liver cancer (91). Some cancers are caused by viruses (84); cervical cancer is primarily caused by a virus that is sexually transmitted (48). Cancer does not generally result from one exposure or influence (92). Instead, there are usually several factors that work together to cause cancer.

APPENDIX F: COMMENTS RECEIVED FOR THE PUBLIC COMMENT DRAFT HEALTH CONSULTATION, REVIEW OF HEALTH STUDIES RELEVANT TO LLNL AND THE SURROUNDING COMMUNITY, LIVERMORE, CALIFORNIA

Click here to view Appendix F in PDF format [PDF, 560KB]

REFERENCES

(1) Lawrence Livermore National Laboratory. Final environmental impact statement and environmental impact report for the continued operation of Lawrence Livermore National Laboratory and Sandia National Laboratory. Livermore, California: Lawrence Livermore National Laboratory; 1992 Aug.

(2) United States Environmental Protection Agency (USEPA). Confirmatory sampling of plutonium in soil from the southeast quadrant of the Lawrence Livermore National Laboratory. National Air and Radiation Environmental Laboratory, USEPA. 1994 Aug 15.

(3) Moore DH, Patterson HW, Hatch F, Discher D, Schneider JS, Bennett D, Mendelsohn ML. Case-control study of malignant melanoma among employees of the Lawrence Livermore National Laboratory. Am J Ind Med 1997;32:377-91.

(4) Agency for Toxic Substances and Disease Registry. Plutonium contamination in Big Trees Park, Lawrence Livermore National Laboratory health consultation, Alameda County, Livermore, California. Atlanta: US Department of Health and Human Services; 1999 May 17.

(5) Agency for Toxic Substances and Disease Registry. Public Health Assessment: Lawrence Livermore National Laboratory: Plutonium 239 sewage sludge used as a soil amendment in the Livermore community. Atlanta: US Department of Health and Human Services, 2003 April 29.

(6) California Department of Health Services. Report: Proposed process to address the historic distribution of sewage sludge containing plutonium released from the Lawrence Livermore National Laboratory. Oakland, CA: California Department of Health Services, November 2002.

(7) Agency for Toxic Substances and Disease Registry. Health Consultation: Lawrence Livermore National Laboratory, Main Site; Lawrence Livermore National Laboratory (Site 300); Savannah River Site: Tritium releases and potential offsite exposures. Atlanta: US Department of Health and Human Services, 2002 March 11.

(8) Weisman J. Alarm over radioactive rainwater; lab surprised at tritium levels. The Tri-Valley Herald; 1994 Mar 25.

(9) Agency for Toxic Substances and Disease Registry. Public Health Assessment: Lawrence Livermore National Laboratory: Community exposures to the 1965 and 1970 accidental tritium releases. Atlanta: US Department of Health and Human Services, 2002 August 7.

(10) Agency for Toxic Substances and Disease Registry. Toxicological profile for ionizing radiation. Atlanta: US Department of Health and Human Services; 1999 Sep.

(11) United States Environmental Protection Agency. Radiation protection. Date accessed: August 19, 2003. USEPA. http://www.epa.gov/radiation/index.html.

(12) Agency for Toxic Substances and Disease Registry. Health consultation: Lawrence Livermore National Laboratory: community health concerns, Alameda County, Livermore, California. Atlanta: US Department of Health and Human Services, 2003 April 25.

(13) Perkins CI, Cohen R, Morris CR, Allen M, Kwong SL, Schlag R, et al. Cancer in California: 1988-1995. Sacramento, CA: California Department of Health Services, Cancer Surveillance Section; April, 1998.

(14) Java J. Rare cancer probe to include valley. The Tri-Valley Herald; 1978 Feb 2.

(15) Lawrence Livermore National Laboratory. Melanoma at LLNL: an update. Energy and Technology Review. University of California, Lawrence Livermore National Laboratory; 1994 Mar; UCRL-52000-94-3.

(16) Hall HI, Miller DR, Rogers JD, Bewerse B. Update on the incidence and mortality from melanoma in the United States. J Am Acad Dermatol 1999;40:35-42.

(17) Rigel DS, Friedman RJ, Kopf AW. The incidence of malignant melanoma in the United States: issues as we approach the 21^st century. J Am Acad Dermatol 1996;34:839-47.

(18) Loria D, Matos E. Risk factors for cutaneous melanoma: a case-control study in Argentina. Int J Dermatol 2001;40:108-14.

(19) Elwood JM, Jopson J. Melanoma and sun exposure: an overview of published studies. Int J Cancer 1997;73:198-203.

(20) Tenkate TD. Occupational exposure to ultraviolet radiation: a health risk assessment. Rev Environ Health 1999;14:187-209.

(21) Hemminki K, Jiang Y, Steineck G. Skin cancer and non-Hodgkin's lymphoma as second malignancies. markers of impaired immune function? Eur J Cancer 2003 Jan;39:223-9.

(22) Mithoefer AB, Supran S, Freeman RB. Risk factors associated with the development of skin cancer after liver transplantation. Liver Transpl. 2002 Oct;8:939-44.

(23) Elwood JM, Whitehead SM, Davison J, Stewart M, Galt M. Malignant melanoma in England: risks associated with naevi, freckles, social class, hair color, and sunburn. Int J Epidemiol 1990;19:801-10.

(24) Pion IA, Darrell SR, Garfinkel L, Silverman MK, Kopf AW. Occupation and the risk of malignant melanoma. Cancer 1995;75 (Supplement January 15, 1995):637-644.

(25) Austin DF, Reynolds P. Occupation and malignant melanoma of the skin. In: Recent results in cancer research. Vol. 102. Berlin: Springer-Verlag, 1986;98-107.

(26) Geller AC, Annas GD. Epidemiology of melanoma and non-melanoma skin cancer. Seminars in Oncology Nursing 2003;19:2-11.

(27) Koh HK. Cutaneous melanoma. N Eng J Med 1991;325:171-82.

(28) Boice JD Jr., Land CE, Preston DL. Ionizing radiation. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer epidemiology and prevention. 2nd ed. Oxford: Oxford University Press; 1996. p. 319-54.

(29) Shimizu Y, Kato H, Schull WJ. Studies of the mortality of A-bomb survivors: 9. Mortality, 1950-1985: Part 2. Cancer mortality based on the recently revised doses (DS86). Radiat Res 1990;121:120-141.

(30) Beral V, Fraser P, Carpenter L, Booth M, Brown A, Rose G. Mortality of employees of the atomic weapons establishment, 1951-82. BMJ 1988;297:757-70.

(31) Checkoway H, Pearce N, Crawford-Brown DJ, Cragle DL. Radiation doses and cause-specific mortality among workers at a nuclear materials fabrication plant. Am J Epidemiol 1988;127:255-66.

(32) Committee on the Biological Effects of Ionizing Radiations, Board on Radiation Effects Research, National Research Council. Health effects of exposure to low-levels of ionizing radiation, BEIR V. Washington, DC: National Academy Press, 1990.

(33) Boice JD. Cancer following medical irradiation. Cancer 1981;1081-1090.

(34) Okada S, Momoshima N. Overview of tritium: characteristics, sources, and problems. Health Phys 1993;65:595-609.

(35) Tymen H, Gerasimo P, Hoffschir D. Contamination and decontamination of a rat and human skin with plutonium and uranium, studied with a Franz's chamber. Int J Radiat Biol 2000;76:1417-24.

(36) Hennekens CH, Buring JE. Epidemiology in medicine. 1^st ed. Boston/Toronto: Little, Brown and Company, 1986.

(37) Austin DF. A study of cancer incidence in Lawrence Livermore Laboratory employees, report no. 1, malignant melanoma. Sacramento, California: California Department of Health Services, Resource for Cancer Epidemiology Section; 1980.

(38) Austin DF, Reynolds PJ, Snyder MA, Biggs MW, Stubbs HA. Malignant melanoma among employees of Lawrence Livermore National Laboratory. Lancet 1981;2:712-6.

(39) Reynolds P, Austin DF. Cancer incidence among employees of the Lawrence Livermore National Laboratory: 1969-1980. West J Med 1985;142:214-8.

(40) Hiatt RA, Fireman B. The possible effect of increased surveillance on the incidence of malignant melanoma. Prev Med 1986;15:652-60.

(41) Schneider JS, Sagebiel RW, Moore DH, Lawton GM. Melanoma surveillance and earlier diagnosis. Lancet 1987;1:1435.

(42) Schneider JS, Moore DH, Sagebiel RW. Early diagnosis of cutaneous malignant melanoma at Lawrence Livermore National Laboratory. Arch Dermatol 1990;126:767-9.

(43) Gong G, Whittemore AS, West D, Moore DH II. Cutaneous melanoma at Lawrence Livermore National Laboratory: comparison with rates in two San Francisco Bay Area counties. Cancer Causes Control;1992;3:191-7.

(44) King M, Spooner D, Rowlands DC. Spontaneous regression of metastic malignant melanoma of the parotid gland and neck lymph notes: a case report and review of the literature. Clin Oncol (R Coll Radiol) 2001;13:466-9.

(45) Hiatt RA, Krieger N, Sagebiel RW, Clark WH Jr., Mihm MC Jr. Surveillance bias and the excess risk of malignant melanoma among employees of the Lawrence Livermore National Laboratory. Epidemiol 1993;4:43-7.

(46) Roberts G Jr. Lab workers during 1974-1997 have high rate of cancer. Tri-Valley Herald; 2001 Nov 15.

(47) Widener A. Lab's cancer rates below norm: but of 17,785 workers in new study, men had higher levels of two types. Contra Costa Times; 2001 Nov 15.

(48) Schiffman MH, Brinton LA, Devesa SS, Fraumeni JF Jr. Cervical cancer. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer epidemiology and prevention. 2nd ed. Oxford: Oxford University Press; 1996. p. 1090-1116.

(49) Austin DF, Reynolds P. A case-control study of malignant melanoma among Lawrence Livermore National Laboratory employees. Sacramento, California: California Department of Health Services, Resource for Cancer Epidemiology Section; 1984 July 3; Report No. 3.

(50) Austin DF, Reynolds P. Investigation of an excess of melanoma among employees of the Lawrence Livermore National Laboratory. Am J Epidemiol 1997;145:423-31.

(51) Moore DH, Patterson HW, Hatch F, Discher D, Schneider JS, Bennett D. Workplace investigation of increased diagnosis of malignant melanoma among employees of Lawrence Livermore National Laboratory. Livermore, California: Lawrence Livermore National Laboratory, University of California; 1994 Aug; report no.: UCRL-LR-106723.

(52) Schwartzbaum JA, Setzer RW, Kupper LL. An exploratory analysis of the occupational correlates of large pigmented nevi at Lawrence Livermore National Laboratory. J Occup Med 1990;32:605-11.

(53) Morton RF, Hebel JR, McCarter RJ. A study guide to epidemiology and biostatistics. Gaithersburg (MD): Aspen Publishers Inc., 1996.

(54) Joint Legislative Audit Committee. The cancer incidence among workers at Lawrence Livermore Laboratory: a synthesis of expert reviews of the study. Sacramento, California: California Legislature, Joint Legislative Audit Committee; 1980 May 23.

(55) Upton AC, Burr WW, Lee JAH, Long JE, Urbach F, Zelen M. Report on the Ad Hoc Advisory Board on melanoma. Arlington, Virginia: American Institute of Biological Sciences, Ad Hoc Advisory Board on Melanoma; July 1980. US Department of Commerce, National Technical Information Service Report No. DE82018384.

(56) Shy CM, Checkoway H, Marshall EG. Malignant melanoma at a scientific laboratory: a synthesis of reviewer's comments on the Austin and Reynolds' study of employees at the Lawrence Livermore National Laboratory. Chapel Hill, North Carolina: University of North Carolina, School of Public Health, Department of Epidemiology, Occupational Health Studies Program; 1985 Nov 15. Report No. UCRL--15737.

(57) Kupper LL, Setzer RW, Schwartzbaum J, Janis J. A case-control study of malignant melanoma among Lawrence Livermore National Laboratory employees: a critical review. Chapel Hill, North Carolina: The Livermore Melanoma Project Partnership; 1987 Jul 1.

(58) Schwartzbaum JA, Setzer W, Kupper LL. Exposure to ionizing radiation and risk of cutaneous malignant melanoma: search for error and bias. Ann Epidemiol 1994;4:487-96.

(59) Gardner MJ. Father's occupational exposure to radiation and the raised level of childhood leukemia near the Sellafield nuclear plant. Environ Health Perspect Aug 1991;94:5-7.

(60) California Department of Health Services, Environmental Health Investigations Branch. Cancer incidence among children and young adults in Livermore, California: 1960-1991, provisional report. 1995 Sep 6.

(61) California Cancer Registry, California Department of Health Services. Cancer incidence in California: 1988-1993. Sacramento (CA): California Department of Health Services. 1996.

(62) Harris JA. Birth defects around Livermore: 1983-1989. Sacramento (CA): California Department of Health Services, California Birth Defects Monitoring Program; 1996 Mar 15.

(63) California Department of Health Services. Proposed process to address the historic distribution of sewage sludge containing plutonium released from the Lawrence Livermore National Laboratory. Oakland (CA): California Department of Health Services; 2002 Nov.

(64) Armstrong BK, English DR. Cutaneous malignant melanoma. Ionizing radiation. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer Epidemiology and Prevention. 2nd ed. Oxford: Oxford University Press, 1996:1282-1312.

(65) Ron E, Preston DL, Kishikawa M, Kobuke T, Iseki M, Tokuoka S, et al. Skin tumor risk among atomic-bomb survivors in Japan. Cancer Causes Control 1998;9:393-401.

(66) Sont WM, Zielinski JM, Ashmore JP, Jiang H, Krewski D, Fair ME, Band PR, Letourneau EG. First analysis of cancer incidence and occupational radiation exposure based on the National Dose Registry of Canada. Am J Epidemiol 2001;153:309-18.

(67) Freedman DM, Sigurdson A, Rao RS, Hauptmann M, Alexander B, Mohan A, Morin Doody M, Linet MS. Risk of melanoma among radiologic technologists in the United States. Int J Cancer 2003;103:556-562.

(68) Berrington One hundred years of observation of British radiologists: mortality from cancer and other causes 1897-1997. Br J Radiol 2001;74:507-19.

(69) Wright WE, Peters JM, Mack TM. Organic chemicals and malignant melanoma. Am J Ind Med 1983;4:577-81.

(70) Ballard T, Lagorio S, De Angelis G, Verdecchia A. Cancer incidence and mortality among flight personnel: a meta-analysis. Aviat Space Environ Med 2000;71;216-24.

(71) Irvine D, Davies DM. The mortality of British Airways pilots, 1966-1989: a proportional mortality study. Aviat Space Environ Med 1992;63:276-9.

(72) Goodman KJ, Bible ML, London S, Mack TM. Proportional melanoma incidence and occupation among White males in Los Angeles County (California, United States). Cancer Causes Control 1995;6:451-9.

(73) Pukkala E, Auvinen A, Wahlberg G. Incidence of cancer among Finnish airline cabin attendants, 1967-92. BMJ 1995;311:649-52.

(74) Vagero D, Swerdlow AJ, Beral V. Occupation and malignant melanoma: a study based on cancer registration data in England and Wales and in Sweden. Br J Ind Med 1990;46:317-24.

(75) Axelson O. Cancer risks from exposure to radon in homes. Environ Health Perspect 1995;103 Suppl 2:37-43.

(76) Etherington DJ, Pheby DFH, Bray FI. An ecological study of cancer incidence and radon levels in south west England. Eur J Cancer 1996;32A;1189-1197.

(77) Henshaw DL, Eatough JP, Richardson RB. Radon as a causative factor in induction of myeloid leukaemia and other cancers. The Lancet 1990;335:1008-12.

(78) Etnier EL, Travis CC, Hetrick DM. Metabolism of organically bound tritium in man. Radiat Res 1984;100:487-502.

(79) Brunner P, Schneider P, Scheicher H, Seyerl G, Kurnik P, Ennemoser O, et al. Tritium intake by exposure to plastic case watches. Health Phys 1996;70:484-7.

(80) Johnson JR, Dunford DW. Dosimetric models of 3H from skin absorption following contact with T2-contaminated surfaces. Health Phys 1985;48:110-3.

(81) Kotzer T, Trivedi A. Dosimetric implications of atmospheric dispersal of tritium near a heavy-water research reactor facility. Radiat Prot Dosimetry 2001;93:61-6.

(82) Linet MS, Cartwright RA. The leukemias. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer epidemiology and prevention. 2nd ed. Oxford: Oxford University Press; 1996. p. 841-92.

(83) Monson RR. Occupation. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer epidemiology and prevention. 2nd ed. Oxford: Oxford University Press; 1996. p. 373-405.

(84) Mueller NE, Evans AS, London WT. Viruses. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer epidemiology and prevention. 2nd ed. Oxford: Oxford University Press; 1996. p. 502-31.

(85) Nomura A. Stomach cancer. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer epidemiology and prevention. 2nd ed. Oxford: Oxford University Press; 1996. p. 707-24.

(86) Brenner H, Arndt V, Bode G, Stegmaier C, Ziegler H, Stumer T. Risk of gastric cancer among smokers infected with Helicobacter Pylori. Int J Cancer 2002;98:446-9.

(87) Bernstein L, Henderson BE. Exogenous hormones. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer epidemiology and prevention. 2nd ed. Oxford: Oxford University Press; 1996. p. 462-488.

(88) Henderson BE, Pike MC, Bernstein L, Ross RK. Breast cancer. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer epidemiology and prevention. 2nd ed. Oxford: Oxford University Press; 1996. p.1022-39.

(89) California Cancer Registry. Monitoring Cancer in California. Date accessed: December 3, 2002. California Cancer Registry, Cancer Surveillance Section, Department of Health Services. Sacramento, CA. http://www.ccrcal.org/brochure/monitor.pdf.

(90) Armstrong BK, English DR. Cutaneous malignant melanoma. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer epidemiology and prevention. 2nd ed. Oxford: Oxford University Press; 1996. p. 1313-30.

(91) Jensen OM, Paine SL, McMichael AJ, Ewertz M. Alcohol. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer Epidemiology and Prevention. 2nd ed. Oxford: Oxford University Press, 1996:290-318.

(92) Rothman KJ, Poole C. Causation and causal inference. In: Schottenfeld D, Fraumeni JF Jr, editors. Cancer epidemiology and prevention. 2nd ed. Oxford: Oxford University Press; 1996. p. 3-10.

Table of Contents