Risk Management

Who Decides?

One of the key differences between the safety assessment and risk assessment paradigms lies in who the decision maker is conceived to be.  In the safety assessment paradigm, the decision of what is safe enough lies with scientific reviewers.  In the risk assessment paradigm, the scientist must inform rather than decide.   That doesn’t mean a scientist cannot participate in the decision, but is does mean that establishing what is known will involve a different group of people than those responsible for resolving the moral implications. 

In regulatory agencies, there is a tendency to equate risk management with agency management.  But, in reality, the correlation is very poor.   The actual decision makers are likely to be a wide variety of different people.  For contaminants in particular, the opinion of the public and the potentially regulated industry are both very important.  When the risk management option being contemplated is consumption advice, then the de facto risk manager is the consumer.

The Toggle Switch

Although it really isn’t quite so simple, a premarket approval decision is often an on/off decision.  An additive is either safe enough to be used or it isn’t.  If it is, then maybe there is still a decision to be made about how much can be used, so the toggle does come with a slider.  But either way, the regulator has nearly complete over how much of a chemical can be in food.

For contaminants, if there is a toggle switch at all, it is about whether the food will be allowed rather than the chemical itself.  Therefore, flipping the switch doesn't just eliminate the chemical; it eliminates the food as well.  Overzealous switch flipping (e.g. trying to regulate contaminants as if they were additives) would easily leave the grocery aisles bare.

Intervention Analysis

So, when a contaminant issue arises, the first question to ask is "What can be done about it?".  The strategies generally fall into three categories:

• Set a Level.  This is the graded version of toggle switch risk management.  There are two problems with it.  First, it requires monitoring to find the food with contaminant concentrations above the specified level.  Second, it potentially requires throwing away food.  But sometimes it works.  In particular, a level may serve to distinguish unusually high concentrations of contamination from those that occur normally, thereby allowing a small portion of the supply that contribute a disproportionate exposure to be excluded.    In addition, setting a level is adept a reducing risk when the health effect occurs as a result of short term exposure that may result from the consumption of a single food item.
• Limit Consumption.  Consumers can be advised to limit consumption, which makes the consumer the de facto risk manager.  This, in turn, brings the decision paradigm issue to the forefront.  Will the consumer be simply told how much is “safe” to eat, or will they be told what the risks are so they can decide for themselves?  This document will pursue the latter option for two major current food safety issues.
• Prevention.  Preventing a contaminant from getting into food in the first place is a very attractive solution.  This isn’t always possible, and it may be expensive, but the dramatic reduction of lead concentrations in food after the elimination of lead soldered cans (Bolger et al, 1996) stands out as a success story.   Changes in agricultural practices can also influence the transfer of contaminants in soil into produce.
• Do Nothing.  Declare victory and move on.  In fact, after some obligatory handwringing, that’s what the FDA usually does.   After all, when it comes to contaminants, the FDA has very little authority.

Pricing Out Lunch

Since it is not just a matter of not adding it, avoiding a contaminant will generally incur other consequences.   There are a number of different strategies that are employed to address this eventuality:

• Cost-Benefit Analysis.  If it is a problem money can solve, then an economist can assign a monetary value to the risk avoided and estimate the cost of doing so.  In fact, the prospect of spending real money is a sign that the safety assessment paradigm is not working.
• Risk-Benefit Analysis.   This type of analysis pits the risk that is avoided against a benefit that may be avoided as well.  This type of analysis is pretty much essential for nutrients that are also toxic like Vitamin A, and metals such as iron and selenium.  A whole food analysis that compares the effects of contaminants and nutrients that co-occur also fall into this category.
• Risk-Risk Analysis.   This type of analysis pits the risk that is avoided against a new risk that arises from avoiding it.  Using fungicides to avoid the risk of mycotoxins (toxins produced by fungi) is a classic example.  Since a foregone benefit can be construed as a risk, virtually any nonmonetary comparative assessment can be put in this category (Graham and Weiner, 1995).

All of these type of analyses turn the decision into an optimization problem.  Instead of “what is safe?” the question becomes “what is best?”   A central problem with all of these analyses is placing the countervailing effects on the same scale.  If the effects are comparable (e.g. USFDA, 2014) then there is no need to weight relative values.  But that is rare.   Cost-benefit analyses use money as the scale, but there is always room for argument about how the effects are valued and the costs estimated (Porter, 1995).  Other analyses often use scales that rate the importance of different health effects.  The Quality Adjusted Life Year (QALY) and Disability Adjusted Life Year (DALY) are the two most common scales. 

References

Bolger PM, Yess, NJ, Gunderson EL, Troxell TC, and Carrington CD (1996).  Identification and reduction of sources of dietary lead in the United States. Food Add Contam 13:53-60.

Graham JD and Weiner JB (1995).  Confronting Risk Tradeoffs.  In: Risk vs. Risk.  Tradeoffs in Protecting Health and the Environment.  Harvard University Press, Cambridge.

Porter TM (1995).  U.S. Army Engineers and the Rise of Cost-Benefit Analysis.  In: Trust in Numbers.  The Pursuit of Objectivity in Science and Public Life.  Princeton University Press, Princeton, pp 148-189.

U.S. Food and Drug Administration (2014). Quantitative Assessment of the Net Effects on Fetal Neurodevelopment from Eating Commercial Fish (As Measured by IQ and also by Early Age Verbal Development in Children).  

Official Post Soundtrack

Dire Straits (1982).  Industrial Disease.  In: Love Over Gold, Track 3.

Post Note

Thesis Post #29.  Part of Risk Assessment Paradigm thread.