Quotes of the Day

In recent years, the last word on the dangers of eating mercury-rich fish seemed to be the government's well-publicized 2004 advisory, which recommended against eating too much higher mercury fish like white tuna, but whose warning applied only to pregnant or nursing women, women of childbearing age and young children. Though mercury overload could damage the still-developing nervous system of a baby, the scientific consensus was that for the average Joe taking in the average amount of fish, heavy with metals or not, it posed no undue threat.

But on Wednesday a New York Times investigation of local restaurants and groceries found that tender slices of tuna sushi being served up all over the city were "tainted" with exceedingly high levels of mercury — so high that eating just six pieces a week would send the average-weight adult over the EPA's acceptable weekly level of mercury intake over a period of several months. All this time, it seems, the average Joe may have been ingesting more harmful mercury than he thought.

So, what is a tuna-lover to do? TIME asked the opinion of Dr. Dariush Mozaffarian, assistant professor of medicine and epidemiology at Harvard Medical School and the Harvard School of Public Health, and co-author of one of the most comprehensive studies to date on the impact of fish consumption on human health.

TIME: Should we stop eating tuna?

Mozaffarian: No. Overall, the dangers of not eating fish [including tuna] outweigh the small possible dangers from mercury. The recommended amount for adults is to eat one or two servings of fish per week — but probably only 10% to 20% of the population in the U.S. eats sufficient fish. The real danger in this country, the real concern, is that we're not eating enough fish. That is very likely increasing our rates of death from heart disease.

What are the dangers of consuming high levels of mercury?

In adults who eat fish with high levels of mercury very frequently — like, every day — there have been case reports that it causes neurologic symptoms, like sensory disturbances and imbalance. Again, that's in people who are eating fish very frequently and eating fish that are generally high in mercury. But the symptoms are temporary; they go away if you stop eating mercury.

Really, the potential concern for long-term exposure at these levels is not with neurologic effects, but with cardiovascular effects. Two of five studies [that looked at the relationship between mercury and cardiovascular disease] suggested that at levels of mercury seen with typical consumption in Western populations, there might be cardiovascular effects long-term [increased risk of heart attack with higher levels of mercury]. But three of the five studies have not shown that.

But both those two studies showed that fish consumption was still protective against cardiovascular disease. Overall, the evidence indicated that people who had higher mercury levels had less protection than people who had lower mercury levels. What those studies suggested is that mercury might lessen the benefit of fish [and omega-3 fatty acid] intake, but they didn't suggest that fish intake was harmful overall.

Is there an optimal risk-benefit ratio for fish? How much omega-3 fatty acids should we get versus mercury?

In commonly eaten fish, if there is any level of mercury where, for an adult [excluding women who are or may become pregnant], the potential risk will be greater than the potential benefit of omega-3s, I haven't seen any study that shows that.... If there is a theoretical limit at which potential harm from mercury might exceed omega-3s, it's probably far higher than what we're seeing now. The idea that you're going to eat a fish meal as an adult and that it's going to give you net cardiovascular harm is just not supported by the current evidence.

I know you want to get to this idea of a balance of risk versus benefit. The evidence suggests that the balance is always toward net benefit. If there were a fish that had almost no omega-3s and had high mercury, then the benefit might be surpassed. But most fish that have mercury tend to be the larger ocean-going fish that also tend to be high in omega-3s.

Still, you can consume too much mercury with ill effects.

If you have high-mercury fish every day for months to years you may get neurologic symptoms. The simplest way to avoid that problem is to eat a variety of fish and seafood. Both our report in 2006 in JAMA and the Institute of Medicine report, which were completely independent and came out at the same time, came to the same conclusion: There's no consistent evidence right now for significant health effects from mercury in adults, and the simplest way to avoid concern is to eat a variety of fish. We went further and recommended, to be prudent, that people who eat fish five or more times a week should choose lower-mercury fish.

For people who eat a few servings per week or less, don't eat the same fish every time, and then you won't have to worry about whether this one is a little bit higher in mercury or this one is a little bit lower.

Just so we know, which fish are higher and which are lower in mercury?

Shellfish are almost all low in mercury because they don't live very long and they're small: shrimp, lobster, crab, scallops. And shellfish have medium levels of omega-3s, similar to other medium-size fish. Salmon are also good. They're high in omega-3s and low in mercury because they're also short-lived.

Light tuna is low in mercury, compared with white (albacore) or red (bluefin) tuna. On average white tuna has three times the mercury as light tuna. But on average white tuna has three times the omega-3s as light tuna — and all the evidence that we can see suggests that omega-3s have more benefit than mercury has harm.

How significant is the EPA's accepted safety level?

The EPA's limit is the acceptable limit of safety, which includes a 10-fold safety factor. That's not a risk level. That's the accepted safety level [0.1 mcg of mercury per kg of body weight per day]. That's 10 times lower than where the EPA determined that risk was occurring — which is a prudent safety limit to be certain that there is no risk. So, for example, if six pieces of tuna sushi a week would put you at the limit, that means you would have to eat 60 pieces to get to the level where the EPA determined risk is occurring.

Also, it's important to note that the EPA set its safety limit based on the potential risks to infants and newborns, not based on the effects in adults.

So, what's the bottom line?

The bottom line is that there's inconclusive evidence that mercury has any long-term effects in adults at the levels that are commonly consumed, and that even if there are effects, studies suggest that they are only to lessen the benefit of the fish. That's important from a public health perspective — we might be getting even more benefit from fish on a population level if we took the mercury out, and that's a very important question that should be answered. But that doesn't mean that the individual person trying to decide on a fish meal should worry about mercury.

I know I sound like I'm trying to downplay the risk but I really think we are experimenting with people's lives when we give recommendations or write stories or reports that make people eat less fish. We know from very good human studies that fish intake reduces the risk of dying from a heart attack by about a third. And heart attack is the number-one cause of death in the U.S. among both women and men. It's the number-one cause of death in almost every country in the world. And eating fish once or twice a week reduces that risk by a third. So if we're causing people not to eat fish or to choose to eat something other than fish because they're worried that the fish has some mercury in it, they're increasing their risk of dying from a heart attack for a concern that has not been established.

• Sora Song