Dihydrogen Monoxide: it’s found everywhere, including in our body, ice caps and the air. There’s no stopping its infiltration into our lives, and if we get too much of it, we’ll die. Sound scary? It might, until we realize that Dihydrogen Monoxide is just the chemical name for water.
Risk analysis is, at its core, a form of structured thinking developed to help grapple with the many risks that humans live with every day. The basic task is to quantify the risk of an action, examine alternatives and determine if the benefits of the action justify its costs. Its practical implications are both important and complex, because context plays an important role in risk analysis. As the quote from Paracelsus goes, “The dose makes the poison.”
Dr. James K. Hammitt, director of the Harvard Center for Risk Analysis and professor of Economics and Decision Sciences, explains. “If you call water by its chemical name, it sounds scary,” he says. “The idea that some things are risky and some things are not is an oversimplification. So, given that everything can be risky, how do we manage living in the world, and how do we set government regulations to assess risk effectively?”
Because the world is so complex, no one person can understand the nuanced details in every decision of daily life: what food to eat, medicine to take, mode of transportation to use and so on. The concept known as “rational ignorance” means that it’s rational to delegate authority to others to help make decisions and minimize the dangers in our lives.
Given that everything can be risky, how do we manage living in the world, and how do we set government regulations to assess risk effectively?
A doctor can provide guidance about medication, and a government body can potentially provide guidelines for exposure to chemical and other public safety risks. Risk analysis helps these experts make recommendations and also help a layperson make informed decisions with that knowledge.
In part, it’s the study of tradeoffs. When we take a flight, for example, there’s always a small but quantifiable risk that the plane will crash. By that logic, the potential of a violent death vastly outweighs the pleasure one might experience at one’s destination. However, the key to risk analysis is that it takes into account both consequence and probability. The probability of a plane crash is very low (much lower, in fact, than that of a car crash). Thus, risk analysis would conclude that the relative risk of taking a flight is low, despite the severity of its consequences.
In the example above, the analysis is somewhat straightforward, because the risk and results are bigger, distinctive and able to be studied separately from other effects. The difficulty of conducting risk analysis with smaller and less quantifiable risks, like coming in contact with very small doses of chemicals like aspartame or radon, is much trickier. It would be unethical to test harmful, long-term chemical impacts on humans, and anecdotal information that comes from real-world data doesn’t necessarily show cause and effect.
Every choice has inherent risks, and replacing one potential action with another also means replacing certain risks with other risks.
By studying epidemiology, researchers can follow individuals who have been exposed to particular chemicals, but sometimes effects can take years or decades to manifest. When it’s a new chemical, there’s no information available yet.
A woman who chewed gum her entire life develops breast cancer, but it’s impossible to extricate what definitively caused the cancer (was it the aspartame? Was it something else? A combination of factors?) to be able to give a formal warning. In essence, Hammitt says, “We care about risks that are too small to measure, and there’s data missing.”
Even attempts to interpret these impacts can be tricky. When a study comes out that shows a small correlation between drinking wine and a longer lifespan, those results can become skewed and overblown in the media and public eye. It’s difficult to prove or disprove that claim as objectively false, and another study might find exactly the opposite reaction. “Trying to identify good chemicals versus. bad chemicals is fundamentally impossible,” says Hammitt.
These overblown fears also don’t effectively make use of rational risk analysis. Bringing in the idea of tradeoffs, Hammitt explains that every choice has inherent risks, and replacing one potential action with another also means replacing certain risks with other risks.
There are small amounts of methylmercury in fish, but eating fish has by and large been proven healthy, so removing fish from one’s diet removes a benefit. Vehicle crashes happen regularly in the U.S., and if the government were to make the speed limit 5 mph it would reduce the likelihood of accidents, but people would no longer be able to travel great distances.
“In the area of pesticides, banning the pesticide doesn’t ban the pest. A lot of organic produce is like this—certain synthetic chemicals cannot be used, so non-synthetic chemicals like sulfur compounds are sometimes used instead,” says Hammitt. In other words, organic doesn’t mean chemical-free, and the choice to eat food means accepting some form of pesticide into one’s diet.
Thus, risk analysis in practice involves recognizing the small risks that are sometimes not measurable, keeping an eye on the bigger risks the can be assessed and thoroughly understanding the tradeoffs. Hammitt teaches experts and administrators how to apply these principles of rigorous risk analysis to their own particular environment as a part of Environmental Health Risk: Analysis and Applications at the Harvard T.H. Chan School of Public Health.
There are small amounts of methylmercury in fish, but eating fish has by and large been proven healthy, so removing fish from one’s diet removes a benefit.
A layperson, while not responsible for setting standards, can also use a more rational and thoughtful kind of risk analysis. Hammitt references Thinking, Fast and Slow by Nobel Prize winner in economics Daniel Kahneman.
“Essentially, we have two decision-making systems. One is System 1, which is fast, intuitive and takes little effort. This is a gut decision. System 2 is cognitive, reflective and analytical. It takes work, so it’s natural not to consciously do it as much. We can be governed by System 1, so the way to improve is to use System 2 as much as you can.”
In other words, automatically rushing to identify “good” versus “bad” decisions is tempting, when the truth is more nuanced and far less easily defined. Since people can’t examine every aspect of their lives, a selective and thoughtful process of risk analysis can be effective when conditions change or new information becomes available.
Ultimately, everyone conducts risk analysis, whether consciously or not. The more that informed, thoughtful decisions can become part of one’s thought process, the more one can integrate effective risk analysis into one’s life.
Harvard T.H. Chan School of Public Health offers Environmental Health Risk: Analysis and Applications, which explores the basics of health risk assessment and applications of risk assessment to diverse problems.