A Risk Quiz for the Experts, Part II

Click here to return to Part I
A summary of the Harvard risk center’s ranking of 10 potential hazards found in U.S. homes appears in the table below, which is reprinted with permission from the center’s April 1998 issue of Risk in Perspectives.

"Of the informal feedback we’ve received so far, people have expressed surprise that radon is the top risk out of these 10 hazards," DeAscentis said. "Yet based on the scientific evidence, it by far was the most risky of the 10 we reviewed, causing almost double the number of estimated deaths per year than the second highest ranked hazard."

The risk center’s review of recent radon studies found that the average annual risk of death to the general population from exposure to radon is 5.8 per 100,000 persons, but (article continues below table)

Ranking of Unintentional Hazards

Weight of Evidence

Population Risk Estimate1

Individual Risk Estimate1

Uncertainty interval5

Deaths per year (U.S.)

Death rate per 100,000 persons per year

Population

Deaths per year (U.S.)

Death rate per 100,000 persons per year
1. Radon Gas suggestive 15,000 5.8 Smokers 13,300 20 x/¸ 2
2. Falling definitive 8,200 3.1 Elderly (75+) 5,500 36.2 x/¸ 1.2
3. Poisoning definitive 8,200 3.1 Ages 25-44 5,020 6.0 x/¸ 1.2
4. Fires and burns definitive 2,900 1.1 Elderly (75+) 750 4.9 x/¸ 1.2
5. Suffocation definitive 2,100 0.8 Elderly (75+) 680 4.5 x/¸ 1.2
6. Firearms (accidental) definitive 800 0.3 Teens/young adults 250 0.7 x/¸ 1.2
7. Environmental tobacco smoke suggestive 900 2.42 Never smoker (female) 500 2.4 x/¸ 5
8. Formaldehyde gas plausible 400 1.3 Mobile home residents 100 3.9 x/¸ 10
9. Insulation fibers plausible 20 0.01 High fiber levels NA3 0.02 x/¸ 10
10. Electric and magnetic fields weak NA4 NA4 NA4 NA4 NA4 NA4
Footnotes

a. Bold type indicates that estimate is based on actuarial data; regular type means that true risk could plausibly be zero.

b. Numbered notes: 1. Population and individual risk estimates assume that risk is "real." "Population risk" refers to the number of annual excess deaths per year; "individual risk" quantitates annual risk to a highly exposed and/or susceptible population. 2. Due to limited data on the general population, the population risk estimate was adopted from the individual risk estimate for never smoker females.

3. NA = not available due to insufficient data.

4. NA = not applicable due to weak weight of evidence.

5. E.g., the real population estimate for number of lung cancer deaths attributable to inhaling radon gas in the home may be two times lower or two times higher than 15,000.

SOURCE: Harvard Center for Risk Analysis, Boston, Massachusetts.
most of this average risk is incurred by smokers. "This subset of the population, whose death rate from radon exposure is 20 deaths per 100,000 smokers, is estimated to account for approximately 85% of the 15,000 lung cancer deaths per year," reports the April 1998 issue of Risk in Perspectives.

In regard to the least-ranked risk -- electric and magnetic fields from power lines -- the center noted that "over 10 years of epidemiological and laboratory research throughout the world has provided only a little support for the hypothesis that exposure to electric and magnetic fields from power lines are a risk factor for cancer." The center’s research of this topic included a review of the National Research Council’s report on 500 papers on the health effects of electric and magnetic field. The National Research Council found that the papers taken as a whole neither proved nor disproved the theory that electric and magnetic fields cause cancer. "Although research on this speculative hypothesis continues, our judgment is that this potential hazard is best regarded as a "phantom risk," Risk in Perspectives states.

The Harvard risk center noted that its ranking has several limitations. "Any exercise in risk ranking can be questioned. Value judgments must be made about which potential hazards should be ranked, which criteria should be employed when rankings are made, what types of adverse health effects should be considered, how scientific uncertainties should be taken into account when rankings are made, and how much weight should be given to overall population risk compared to the risks faced by highly exposed or sensitive subpopulations," Risk in Perspectives stated. It also was noted that the center’s ranking bases the risk estimates given on different types of data. "For example, poisoning risk estimates are taken from poison control databases, while formaldehyde risk estimates are based on extrapolations from experimental studies of rodents."

Project’s Next Step

The next phase for the risk center’s study of hazards in the home will be to survey the public’s perceptions, said DeAscentis, a master’s degree candidate at Tufts University’s Graduate School of Arts and Sciences who is basing her thesis on the project. "Our next step will be to poll the public, asking a certain population to rank these 10 risks. Then we’ll compare the public’s perception to the scientific data."

Understanding the public’s perception of risks such as these is a key part of risk communication and management, noted Graham, who is the founding director of the Harvard Center for Risk Analysis and a professor of policy and decision sciences at the Harvard School of Public Health.

"The public tends to receive information about risks one at a time in the news media without the benefit of any comparative perspective," Graham said. "To make strides in improving public health, we have to educate people about which risks are the most hazardous so that they can tackle those first."

Related Links

Harvard Center for Risk Analysis http://www.hsph.harvard.edu/Organizations/hcra/hcra.html

Back issues of Risk in Perspective http://www.hsph.harvard.edu/Organizations/hcra/hcrapub.html#RIPS

John Graham bio http://www.hsph.harvard.edu/Organizations/hcra/jdg.html

Posted June 26, 1998.

Go to:

  • RiskWorld News Department

  • RiskWorld Home Page

  • Tec-Com Inc.

    • Copyright © 1998 by Tec-Com Inc.