When we awaken to meet each day, we face a variety of health and safety risks and benefits in our lives. Life is inherently full of risks, but this should not keep us from "taking on the day" or even unnecessarily limiting our activities. Although we are familiar with many common hazards (driving a car, participating in high-risk sports or even walking across the street), it is difficult to decide which ones to worry about or avoid and which ones to accept.
One key challenge is getting good science-based information about the relative risks of different activities, especially when it comes to a very personal and pleasurable activity: eating and drinking. For example, public concern about foodborne risks often exceeds concerns for other health and safety hazards, despite government assurances that the U.S. food supply is one of the safest in the world. The public also tends to misjudge the relative risks from food safety issues, ranking pesticide residues as posing a much greater threat to human health than not achieving a healthful diet. Persistent focus on food safety issues in the news sparks continued criticism by health professionals that consumers may be getting the false impression that we are amidst a "food safety crisis" (see "Sensible Talk About Food Safety," Food Insight, July/August 1998).
So what are the "real" risks from food and beverages? Although precise estimates of food-related risks are not yet available, we can think rationally about food safety and appropriate health-promoting (or risk-reducing) behaviors. Health professionals, journalists and other communicators can empower consumers as they sort through the many risks they face by helping people think more broadly about risk issues and put food safety risks in the context of other health and safety hazards. Consistency between public and expert opinion is necessary to ensure that the greatest public health risks receive adequate attention and that both science and public values are included in decisions about food safety. This article helps provide a framework for putting food safety risks in perspective and discusses various ways in which risk information is presented. The goal: to help communicators and health professionals illustrate more clearly the relative risks of food safety and other public health issues. Understanding Risk"Most of us have little experience thinking about the size or magnitude of different risks. Although we have an intuitive sense that some risks are smaller or larger than others, research shows that people tend to underestimate relatively large risks such as heart disease and heart attacks and overestimate relatively small risks such as botulism, a foodborne illness caused by clostridium botulinum," said George Gray, Ph.D., deputy director for the Center for Risk Analysis at Harvard School of Public Health. "Public misperceptions may be driven in part by what we hear and read, but a lack of understanding about risk magnitudes and how to interpret risk information also influences them," he continued. Judging the size of a risk can be more relevant if the risk is compared to other, more familiar activities that inherently carry risk. For example, consider the fact that the annual risk of dying in the United States from motor vehicle accidents is about 160 in 1 million. This means that in a population of one million automobile drivers, 160 are "expected" to die from an accident in any given year (keeping in mind it does not mean that 160 people "will" die, since in reality, greater or fewer people may die).
Is this a big risk? Table 1 shows that each year, the average risk of dying from motor vehicle accidents is about 10 times greater than the risk of dying from either fires or drowning, and 100 times greater than the risk of dying from either railway or airplane accidents. In contrast, this risk is 13 and 18 times lower than the annual risk of dying from cancer and heart disease, respectively. Thus, driving a car poses a significant risk when compared to other public health hazards, including some alternate forms of transportation. Yet, it still trails behind several leading causes of death.
Putting Risk into Perspective
While data on foodborne risks are limited, it is estimated that most food-related hazards result from harmful bacteria (such as E.coli or salmonella). This is an approximate annual risk of 36 in 1 million (based on the United States population of 260 million), which is 7 times greater than the risk of dying from tuberculosis (TB) each year, and 100 times greater than the risk of dying from floods. On the other hand, this risk is 78 times lower than the risk of dying from heart disease, 57 times lower than the risk of dying from cancer and 4 times lower than the risk of dying from motor vehicle accidents. Even the mere act of eating food poses an annual risk of dying from choking of 5 in 1 million. Of course, the risk of not eating also can be substantial. While these comparisons help put the magnitude of risk into perspective, they do not convey the level of uncertainty associated with formulating risk estimates or provide information on how the risk of a specific activity is distributed in the population. For example, the risk of dying from a motor vehicle accident is known with greater certainty than the risk of dying from harmful bacteria in food, since the cause of death is easier to identify (and quantify) when it is a car accident versus a foodborne illness death. In addition, the elderly are at much greater risk from cancer or heart disease than are younger persons, while the reverse is true for automobile risks. The acceptability of risk also requires the consideration of other important factors. For example, driving a car or hang gliding are voluntary activities that provide many perceived benefits, thereby increasing the acceptability of these risks for many people.
"Besides thinking about risk in a broader context, it is important to be aware of the various ways in which risk information is presented. Risks can refer to a variety of health outcomes including death, cancer and non-fatal injuries or illness, and risk rankings may differ depending on which outcome is used," said Gray. For example, although the risk of dying from foodborne disease ranks relatively low compared to other public health hazards, the risk of foodborne illness may rank much higher than other sources where the outcome is not death. Indeed, it is estimated that up to 81 million cases of foodborne illness occur in the United States each year. A parallel example is comparing the risk of death from harmful bacteria with the risk of illness from bacteria. According to Gray, risk information may also be framed according to different consequences, such as the number of deaths in a population or the estimated loss in life expectancy (i.e., number of years taken away from a life). Further, risk estimates may be based on exposures that occur in a given year (i.e., annual risk) or over the duration of one's lifetime (i.e., lifetime risk).
Relative or Absolute: A Distinction That Makes a Difference
The distinction between relative risk and absolute risk is also important. "Relative risk puts risk in comparative terms and indicates the ratio of risk among individuals who are exposed versus non-exposed to a particular hazard," explained Gray. Relative risks greater than 1 (>1) mean that exposed persons are more likely to experience a health effect than non-exposed persons, while relative risks less than 1 (<1) suggest that the exposure has a 'protective' effect (e.g., consuming fruits and vegetables has been found to reduce risk of certain cancers).
Absolute risk refers to the actual risk of an occurrence-the chance of a specific outcome occurring. For example, suppose that a study shows that a man who brushes his teeth only once a day is 50 percent more likely to lose all his teeth in the next 10 years than others who brush their teeth twice a day. Yet, the absolute risk of the man losing all his teeth may be only 1 percent. In this case, the relative risk makes the problem seem more important than it may be in reality. On the other hand, relative risk can also make a problem seem less important than it really is, so it is best to consider both types. "In practice, relative risks are more useful for identifying the result of various health outcomes or behaviors, while absolute risks are better suited for evaluating the actual impacts of a risk on a specified population," said Gray.
Food for Thought
Like everything we do, consuming food and beverages poses some degree of risk. Although everyone is entitled to choose which chances they want to accept or avoid, personal actions to reduce foodborne and other risks should be based on informed decision-making and science rather than allowing reports of outbreaks to lead one to believe the risk is greater than it is. Certain behaviors, such as eating fewer fruits and vegetables to avoid potential exposures to pesticides, can also lead to potential negative consequences, such as decreased intake of essential vitamins and nutrients. "While there is no 'right' answer for how to judge food safety risks," said Gray, "thinking about foodborne hazards in the context of other hazards can help shift attention away from the 'wrong' risks and toward those that pose the greatest threats to public health and safety."
Risk Comparisons:
Annual Risk of Dying in the U.S. per Million Persons at Risk
|
|
Cause Deaths
|
Death per
1,000,000/year
|
| Heart disease |
2800 |
| All cancers |
2050 |
| Parachutist |
2000 |
| Fire fighter; Hang glider |
800 |
| Lung cancer |
590 |
| Pneumonia |
320 |
| Diabetes; Police officer |
230 |
| Motor vehicle accidents; Breast cancer |
160 |
| Homicide |
80 |
| Falls |
50 |
| Foodborne bacteria |
36 |
| Accidental poisoning (drugs and medication) |
30 |
| Fires and burns; Drowning |
15 |
| Tuberculosis; Firearms |
5 |
| Choking, inhalation or ingestion of foreign object/food |
4 |
| Electric current; Railway |
2 |
| Airline crash (one trip) |
0.6 |
| Floods |
0.4 |
| Lightning; Insect bite or sting |
0.2 |
| Hit by falling aircraft |
0.06 |
| Hurricane |
0.04 |
|
Sources: 1997 US Statistical Abstract; National Safety Council (1995), Accident Facts; Crouch & Wilson (1982), Risk/Benefit Analysis.
|
Risk analogies
A risk of 1 in 1,000,000 is about equal to:
- the chance of flipping a coin and getting 20 heads in 20 consecutive tries.
- the chance of rolling a dice and getting 7 sixes in 7 consecutive rolls.
- the chance of dying from taking 1 flight on a regularly scheduled jet airliner.
- 30 seconds in a year.
- less than ½ inch in a mile.
A risk of 1 in 100,000 is about equal to:
- the chance of flipping a coin and getting 16 heads in 16 consecutive tries.
- the chance of rolling a dice and getting 6 sixes in 6 consecutive rolls.
- the chance of dying from taking 10 flights on regularly scheduled jet airliners.
- 5 minutes in a year.
- ½ inch in a mile.
A risk of 1 in 10,000 is about equal to:
- the chance of flipping a coin and getting 13 heads in 13 consecutive tries.
- the chance of rolling a dice and getting 5 sixes in 5 consecutive rolls.
- the chance of dying from taking 100 flights on regularly scheduled jet airliners.
- 53 minutes in a year.
- 6 inches in a mile.
