This publication does not discuss the possible health effects that might be related to radiation exposure. For more information, please refer to HHIN's publications: Health Bulletin and An Overview of Hanford and Radiation Health Effects.

I. BASICS OF HEALTH RISK

Information Needed to Assess Risk

To estimate a person’s risk from exposure to radiation requires finding answers to three questions:

1. How much radiation did the person's body receive?

2. Is there a scientifically based connection between this type of radiation exposure and a specific disease?

3. What are the chances of developing a disease given the radiation dose the person received?

To illustrate these concepts, this publication focuses on Hanford’s releases of iodine-131 and the risk for thyroid cancer. (Iodine-131—a radioactive form of iodine—accounted for more than 98 percent of the radiation doses that most people received from Hanford.) The same concepts also apply to estimating the risk for other thyroid diseases and the health risks from other exposures.

1. Radiation Dose

A radiation dose is the amount of radiation that the body or an organ in the body has absorbed. Units of measure for dose include rad, millirad, Gray and milliGray. To estimate radiation dose, one needs to know about the radiation exposure. To find out about exposures and doses from releases that happened in the past, scientists conduct a study called a dose reconstruction. For example, the Hanford Environmental Dose Reconstruction (HEDR) Project studied Hanford’s past releases. The study worked to find out what radioactive materials Hanford had released, how they spread through the environment, how people might have been exposed and what doses they might have received. Then, using the HEDR study’s work, the Hanford Individual Dose Assessment (IDA) Project is calculating individual dose estimates for people who lived or spent time in the HEDR study area between 1944 and 1957.

2. Health Effects and Dose-Response Relation

The second step in assessing risk is to find out if an exposure is related to cancer or other health effects. This is the basic goal of many health studies (called epidemiologic studies). Epidemiologic studies look for connections (also called associations or relationships) between levels of radiation and disease. If enough information is available on health effects in people who received different radiation doses, then scientists can begin to assess the relationship between the incidence of disease and the size of radiation dose. This relationship between disease incidence and the size of radiation dose is called a dose-response relation. In many cases, the information needed to identify a dose-response relationship is difficult to establish.

Note: This example is for excess absolute risk. If this graph showed relative risk, the dose/risk line would start at a risk of 1.0 rather than 0. A relative risk of 1.0 means that the exposure produced no risk above the background risk. 

For cancers connected with radiation exposure, some scientists think there is a no- threshold, linear dose-response. "No-threshold" means that there is no dose level (threshold) below which a person is considered safe from cancer caused by the radiation exposure. "Linear dose-response" means that as the dose increases by a certain amount, the risk is also increased by a similar amount. See Figure 1 for an example of how this might look in a graph.

There is some controversy about whether there is a linear dose-response relationship at low radiation doses. There are, however, few data to examine dose-response relationships at low doses. Many scientists and radiation safety officials have assumed that there is a linear dose-response relationship. Others believe this practice overstates the risk at low dose levels. Since no one really knows, government agencies usually base radiation protection standards on the linear dose-response relation because it gives the most protection for public health.

In addition to looking for a dose-response relationship, there are other criteria scientists use to decide if there is a connection between an exposure and a disease. For more information, please refer to the HHIN publication, Epidemiology: Understanding Health Studies.

3. Risk Per Unit Dose

The third step in assessing risk is to figure out what a person’s chances are of developing a disease, given the dose he or she received. For many radioactive materials, scientists have estimated how much the risk of cancer (or other diseases) may increase with each unit of dose a person receives. This is called the risk per unit dose. For a dose measured in units of rad, the risk of cancer per unit dose tells how much a person’s risk of getting cancer has increased with each rad of dose he or she received. For example, if the risk is one case of a disease in 10,000 people (1/10,000) per rad, then the risk at a dose of 1 rad will be 1/10,000, at a dose of 2 rad will be 2/10,000 and so on.

Scientists have developed these risk estimates based on epidemiologic studies of people who were exposed to radiation, such as survivors of the atomic bombs dropped on Japan and persons exposed to radiation in medical procedures. Scientists then use the risk information from these studies to estimate risk for other exposed groups.

Unfortunately, the risk of thyroid cancer per unit dose from iodine-131 is not well understood. This makes it hard to estimate the chances that a person exposed to iodine-131 will get a disease of the thyroid gland. The results of the Hanford Thyroid Disease Study (HTDS) may shed more light on this, however. The draft results were released on January 28, 1999.

The Chances of Getting a Disease: Risk Estimates and Risk Factors

Health risk can be estimated for varying lengths of time. This publication focuses on risk over an average lifetime.

Reports in the news sometimes refer to the risk that people in the United States have for different types of cancer. For example, the National Cancer Institute’s data show that men have about 1 chance in 2 of developing cancer at some time in their lives. Women have about 1 chance in 3. These risk estimates are based on statistics for cancer cases in the entire U.S. population.

Risk estimates for a population don't necessarily predict the risk for a specific person.

It can be helpful to know about risk estimates for people in general (called population risk estimates) or for a group of people in a health study. However, there are two cautions about using these risk estimates. First, risk estimates for a population don’t necessarily predict the risk for a specific person. For example, take the statistic that U.S. men have about 1 chance in 2 of developing cancer. This does not mean that if you have two brothers, one of them will get cancer. It means that among all the men in the United States, about half will develop a cancer at some time in their lives.

The second caution about using population risk estimates is that many individual differences can affect risk. A person might have certain genes, a lifestyle, or past exposures that make his or her risk higher or lower than the average for everyone in the country or for everyone in a given health study. (See "Accounting for Individual Factors in Risk Estimates.")

There are five terms you may see that describe different ways of looking at risk. These are (1) background risk, (2) excess absolute risk, (3) relative risk, (4) excess relative risk and (5) combined risk. When reading a study, it is important to know which type of risk it is discussing.

BACKGROUND RISK

What is the usual risk for cancer (or a specific disease)? Or, what would the risk for cancer be without Hanford? Example: A 48 percent lifetime background risk of cancer for men = A 48 percent chance of developing cancer in a man’s lifetime = 48 chances in 100 that a man will develop cancer in his lifetime = About one in two men will get cancer in their lifetimes

Background risk can be estimated for a broad population, such as all residents of the United States, or for specific groups that may differ from the general population. It can be estimated for any period of time—on a yearly basis, for example, or for an average lifetime.

Background risk expressed in absolute units, such as a percentage, is called absolute background risk. For example, statistics show that 48 percent of men and 38 percent of women in the United States may be diagnosed with some type of cancer at some time in their lives.

Note that the same risk estimate can be stated in several different ways. See the box below for examples.

Background Risk for Thyroid Cancer and Benign Thyroid Tumors THYROID CANCER The average lifetime background risk of being diagnosed with thyroid cancer is 0.0066 for U.S. women and 0.0027 for U.S. men. To avoid writing lots of zeros, scientists express these numbers in scientific notation, using the exponent of 10 (preceded by a negative sign) to show the number of places the decimal point has moved to the right. For example, in scientific notation, the lifetime risk estimates in the previous sentence would be 6.6 x 10-3 for women and 2.7 x 10-3 for men. In this example, the "minus 3" after the 10 shows that the decimal point in 0.0066 and 0.0027 moved three places to the right to become 6.6 and 2.7. Here are some other ways to look at the same risk estimates: If randomly selected groups of 10,000 women were followed over their lifetimes, we would expect, on average, to diagnose 66 cases of thyroid cancer in each of the groups. Under similar conditions, we would expect to diagnose 27 cases among each group of 10,000 men. Or, on average, we would expect 66 thyroid cancers to be diagnosed in 10,000 women over their lifetimes, and 27 cases to be diagnosed in 10,000 men. A woman has a 0.66 percent (two-thirds of one percent) chance of being diagnosed with thyroid cancer in her lifetime. A man has a 0.27 percent (just over a quarter of one percent) chance. About 1 in 152 women (100 divided by 0.66 = 152) will have a diagnosis of thyroid cancer in her lifetime, as will about 1 in 370 men (100 divided by 0.27 = 370). BENIGN THYROID TUMORS The lifetime risk of benign (non-cancerous) thyroid tumors is about three times the risk for thyroid cancer. So, to find the lifetime background risk for being diagnosed with benign thyroid tumors, multiply the risk for thyroid cancer by three. For example, the lifetime background risk for being diagnosed with benign thyroid tumors is 1.98 percent (or 0.0198) for women (0.66% x 3; or 0.0066 x 3) and 0.81 percent for men (0.27% x 3; or 0.0027 x 3).

EXCESS ABSOLUTE RISK

What's your chance of developing cancer (or a specific disease), given the radiation dose you received? The excess absolute risk for cancer, given a certain radiation dose, is an estimate of the chance of developing cancer over a specified period of time, such as an average lifetime. Excess absolute risk can be expressed as a number from 0 to 1, or as a percentage from 0 to 100. Example: An excess absolute risk for cancer of 2 percent means that the exposure adds 2 chances in 100 to the background risk.

Excess absolute risk is the additional risk that a radiation dose adds on top of the usual (background) risk for a disease. In the example, if a person has a 2 percent excess absolute risk (2 more chances in 100 of getting the disease than he or she would have without the exposure) and the backgound risk is 1 percent (1 chance in 100), then the person's risk is 3 percent (1 from background + 2 from the radiation dose = 3) or 3 chances in 100.

Note that excess absolute risk sounds similar to excess relative risk. However, these terms are different statements about risk. Both are reported in scientific studies.

RELATIVE RISK

How does the added risk for cancer (or a specific disease) from your radiation dose compare with the risk for someone who wasn't exposed? Relative risk is a ratio (the risk of cancer for a given radiation dose, divided by the background risk) A relative risk of 1.0 means the exposure produced no risk above the background risk (the risk to someone who wasn’t exposed) An exposure resulting in a relative risk of 2.0 produced twice the background risk Example: 100 people with the disease in the exposed group¸ 50 people with the disease in an unexposed group (assuming an equal number of people in both groups) = a relative risk of 2.0

Relative risk compares the risk of developing cancer (or any other disease) for exposed persons with the risk for unexposed persons. Results of health studies are often expressed as a relative risk. This is because many harmful agents are thought to increase the risk for disease by multiplying background risk by a certain amount per unit of dose (the relative risk), rather than by adding a fixed excess risk per unit of dose.

While absolute risk is expressed as a number between 0 and 1 or as a percentage, relative risk is a ratio that can be expressed with any real number. Relative risk is usually estimated by taking the risk of the disease in the exposed group and dividing it by the background risk, that is, the risk of the disease in the unexposed group.

A relative risk will be greater than 1.0 if the radiation dose increases the risk of the disease. This is because a relative risk of 1.0 indicates that the risk of disease in the exposed group equals the risk of the unexposed group. For example, if the risk is 2 percent (2 in 100) in the exposed group and 2 percent in the unexposed group, the relative risk is 1.0 (2 divided by 2 = 1).

A relative risk of 2.0 suggests that, compared with the unexposed group, the exposed group has twice the risk of developing the disease in question. A relative risk of 3.0 suggests that the exposed group has three times the risk, and so on.

For example, if the risk is 5 percent (5 in 100) in the exposed group and 2 percent (2 in 100) in the unexposed group, then the relative risk is 2.5 (5 divided by 2 = 2.5). This exposed group has two-and-a-half times the risk as the unexposed group.

EXCESS RELATIVE RISK

How much of the increase in relative risk comes from the radiation dose? Excess relative risk = Relative risk minus 1 (the background risk) Example: An excess relative risk of 2.0 = Relative risk of 3.0 minus 1 (the background risk)

Another way of expressing risk is excess relative risk. This is figured by subtracting 1 (that is, the background risk) from the relative risk. Relative risk and excess relative risk

Another way of expressing risk is excess relative risk. This is figured by subtracting 1 (that is, the background risk) from the relative risk. Relative risk and excess relative risk are both informative and can readily be converted one to another. See the box to the right for an example of excess relative risk.

COMBINING RISKS FROM DIFFERENT SOURCES

If you had an unusual exposure to radiation, what is your total risk for cancer (or a specific disease) in your lifetime? To answer this question, add the risk from the radiation dose to the lifetime background risk. Total lifetime risk = Relative risk x absolute lifetime background risk = (Excess relative risk x absolute lifetime background risk) + absolute lifetime background risk OR (Excess relative risk + 1.0) x absolute lifetime background risk = Excess absolute risk + absolute lifetime background risk

It is possible to estimate the background risk combined with the risk from a particular exposure, to get a person’s combined risk. This risk can be calculated for any time period.

The "total lifetime risk" equations above show how to estimate combined lifetime risk. For example, for a risk expressed as excess absolute risk over a lifetime, add it to the absolute lifetime background risk to estimate the combined lifetime risk. So if an iodine-131 dose results in an excess absolute risk of 2.0 percent over the lifetime of a woman, the woman would have a 2.66 percent combined lifetime risk of developing thyroid cancer (2 percent excess absolute lifetime risk + 0.66 percent lifetime background risk). The box below gives an example of combining risks.

RISK PER UNIT DOSE, EXCESS RELATIVE RISK, COMBINING RISKS A team led by Elaine Ron of the National Cancer Institute analyzed seven studies of children and adults who were exposed to external radiation (that is, sources of radiation from outside the body). They estimated that persons exposed as children had an excess relative risk for thyroid cancer of about 7.7 per Gray of dose to the thyroid gland. (A Gray is an international unit of absorbed radiation dose. One Gray equals 100 rad. The Gray is abbreviated Gy.) Risk per unit dose: The risk per Gray means that at a dose of 1 Gray, the excess relative risk is 7.7. For each added Gray of dose, the excess relative risk would increase by another 7.7 units. Excess relative risk: The excess relative risk of 7.7 per Gray of dose means that a child who received a dose of 1 Gray would have an additional risk for thyroid cancer of 7.7 times the background risk. Using the example on page 7 for background risk, if the background risk for developing thyroid cancer over a lifetime is 0.27 percent for a male, then a boy who received a thyroid dose of 1 Gray would have an excess relative risk of 2.1 percent (7.7 x 0.27%). Combining risks: Since this is an excess risk, the total risk would be the sum of the background risk and 7.7 times this risk, or 8.7 times the background risk. Note: The studies in this analysis were cases where the radiation dose was from a source outside the body (external exposure), such as X-rays. Almost all the thyroid dose from Hanford’s iodine-131 releases was from inside the body (internal exposure). The risks for external radiation do not necessarily apply to internal radiation.

Accounting for Individual factors in Risk Estimates

Each person may have individual differences that may increase or decrease the average risk of developing a disease after a radiation exposure. For example, risk can differ by gender, age, race, diet and lifestyle. Some individual differences cannot be accounted for in making a risk estimate. One example is a person’s genetic heritage or genotype.

Two risk factors that affect the risk for thyroid cancer are gender and age when exposed to radiation. Gender is a risk factor for thyroid cancer in the general population: More women than men get thyroid cancer. Regarding age as a risk factor, an analysis of people exposed to external radiation found that the risk for thyroid cancer decreased the older an individual was when exposed, up to about age 20. For those over 20 when exposed, the risk appears to be very small.

Charts of risk per unit radiation dose that may be included in published studies may show risk by gender or by age at exposure. When comparing your situation to study results, be sure to use the chart appropriate for you.

II. WHY RISK IS NOT EXACT: UNCERTAINTY

Estimating risk is akin to making a prediction. Like most predictions, risk estimates rely on information that is limited in some way. For this reason, risk estimates are uncertain. Uncertainty in a risk estimate may be the result of the amount of data available and/or the quality of the data. For example, in a scientific study of the relation between radiation dose and disease rate, there may not be much information about the radiation doses people received; or the number of people from the exposed population in a study may be quite small. Each of these situations would increase the uncertainty of the estimates of risk per unit dose.

Key Areas of Uncertainty for Risk from Iodine-131

Risk estimates for iodine-131 doses are influenced by two important types of uncertainty. One is simply that the risk estimates are based on uncertain dose estimates. This is the case for dose estimates for people who lived or spent time around Hanford. There is uncertainty about the doses because there is limited information about Hanford’s past releases, how they spread in the environment and the ways people were exposed. The extent to which doses are uncertain also influences risk estimates based on these doses.

The second key area of uncertainty for iodine-131 risk has to do the differences between external exposures to gamma rays and X-rays, and internal exposures to iodine-131. There is still scientific debate over how to estimate the risk per unit dose for internal iodine-131 exposure. Scientists are more certain about the relationship between dose and cancer risk in situations where people received doses from external exposure to gamma rays or X-rays. For example, scientists have extensively studied the relation between radiation dose and disease risk in Japanese atomic bomb survivors and in people who have received certain medical radiation procedures.

In spite of these uncertainties, national and international groups of experts have made estimates of iodine-131’s potency for causing cancer in humans. These expert groups include the U.S. National Academy of Sciences Committee on Biological Effects of Ionizing Radiation (BEIR), the International Commission on Radiological Protection (ICRP) and the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR). The main purpose of their estimates is to provide information for setting safety standards to protect workers and the public.

These experts reviewed the studies of external radiation exposures and compared them with internal iodine-131 exposure. They concluded that internal iodine-131 exposure has less potency to cause cancer than external exposures to gamma or X-radiation. The estimates of the potency of iodine-131 to cause thyroid cancer range from one-fifth (1/5) to two-thirds (2/3) the potency of gamma or X-radiation.

It is important to know that the experts’ risk estimates came out before the recent data from studies of children exposed to iodine-131 from the Chernobyl accident. Also, some scientists disagree with the expert groups and believe instead that a dose to the thyroid from internal iodine-131 has about the same potency to cause thyroid cancer as a dose from external gamma or X-radiation. On the other hand, some believe that iodine-131 has less than one-fifth (1/5) the potency.

The following graph shows what confidence intervals might look like in reports of study results.

Example: Confidence Interval The analysis of seven studies by Ron and colleagues estimated that children exposed to external sources of radiation had an excess relative risk for thyroid cancer of about 7.7 per Gray of dose to the thyroid gland. These researchers estimated a 95 percent confidence interval that ranged from 2.1 to 28.7 per Gray. This means that if the study were repeated many times, 95 percent of the excess relative risk estimates would fall somewhere between 2.1 and 28.7 per Gray of dose to the thyroid.

Quantifying Uncertainty

Confidence Intervals

Because there is uncertainty in risk estimates that are made for different radiation dose levels, scientists often include a confidence interval with a risk estimate. A confidence interval accounts for the possibility that different groups of individuals might have different risk estimates even if they have the same range of dose estimates. The confidence interval is stated as a percentage, such as 90 or 95 percent. A confidence interval of 95 percent means that if scientists repeat a study in many groups of individuals who have the same range of dose estimates, then approximately 95 percent of the risk estimates would fall within the 95 percent confidence interval.

Figure 2 shows what confidence intervals might look like in a graph of dose-response relations.

Combined Uncertainty

"Key Areas of Uncertainty for Risk from Iodine-131" describes some of the uncertainties in estimating the risk for disease from iodine-131 exposure. Each of these uncertainties can be combined into a single uncertainty limit or interval. The National Cancer Institute (NCI) took this approach in describing uncertainty for iodine-131 exposures from the Nevada Test Site. (For information on finding NCI’s dose estimates on its Internet site, contact HHIN.)

Combined uncertainties could be estimated for Hanford doses and would help in using published scientific studies to make estimates of risk. To date, such estimates of risk for Hanford doses have not been made. However, both the individual dose estimates given by the Hanford IDA Project and the dose estimates of the Hanford Thyroid Disease Study (HTDS) are based on the same source: the computer models that the HEDR Project created for iodine-131 dose estimates. For this reason, if you get your individual iodine-131 thyroid dose estimate from the Hanford IDA Project, you can compare it to the HTDS results in order to get an idea of your risk for thyroid disease. The Hanford IDA Project is providing individual thyroid dose estimates as a range and as a central estimate in this range.

III. SUGGESTED READING

HHIN. Epidemiology: Understanding Health Studies.

HHIN. Radioactivity in the Body.

HHIN. Tips for Understanding Health Studies

Makhijani, Arjun. "Risk Analysis: Only One Tool," Science for Democratic Action Vol. 2, No. 2, April 1993. (Institute for Energy and Environmental Research, 6935 Laurel Ave., Takoma Park, MD 10912, (301) 270-5500. Web site: http://www.ieer.org)

Sumner, David, Howard Hu and Alistair Woodward. "Health Risks of Ionizing Radiation," Energy and Security No. 4, 1997. (Institute for Energy and Environmental Research, see contact information above)

References

American Cancer Society. 1997. Cancer Facts & Figures—1997 (citing SEER data). Atlanta, Ga.: American Cancer Society, Inc.

Fernald Health Effects Subcommittee. 1997-98. Glossary of Terms. Atlanta, Ga.: Centers for Disease Control and Prevention.

Hanford Thyroid Disease Study. 1998. Questions & Answers About Radiation & Thyroid Disease. Seattle, Wash.: Fred Hutchinson Cancer Research Center. October, 1997. Questions & Answers About the Study. Seattle, Wash.: Fred Hutchinson Cancer Research Center. February.

National Cancer Institute (NCI). 1997. Backgrounder: Background Information on Thyroid Cancer and Radiation Risk. Bethesda, Md.: NCI. August 1, 1997. Backgrounder: Questions and Answers on the NCI Fallout Report. Bethesda, Md.: NCI. March 7, 1997.

Calculation of the Estimated Lifetime Risk of Radiation-Related Thyroid Cancer in the United States from the Nevada Test Site Fallout (Original Presentation, and Questions and Answers). Bethesda, Md.: NCI. December 19, 1997.

Executive Summary; and Technical Summary. Study Estimating Thyroid Doses of I-131 Received by Americans from Nevada Atmospheric Nuclear Bomb Test. Bethesda, Md.: NCI.

National Council on Radiation Protection and Measurements (NCRP). 1985. Induction of Thyroid Cancer by Ionizing Radiation. NCRP report no. 80. Bethesda, Md.: NCRP.

National Research Council. 1990. Health Effects of Exposure to Low Levels of Ionizing Radiation. Washington, D.C.: National Academy Press.

Ron, E., J. H. Lubin, R. E. Shore, K. Mabuchi, B. Modan, L. M. Pottern, A. B. Schneider, M. A. Tucker, and J. D. Boice. 1995. Thyroid cancer after exposure to external radiation: A pooled analysis of seven studies. Radiation Research 141:259-277.

Spengler, Robert F., ScD. 1997. Hanford Medical Monitoring Program: Background Consideration Document and ATSDR Decision. PB97-193072. Atlanta, Ga.: ATSDR. July.

United Nations Scientific Committee on the Effects of Atomic Radiation. 1994. Sources and Effects of Ionizing Radiation. New York: United Nations.

Published February 1999