Last updated on April 18th, 2017
Study: Baked, Broiled — But Not Fried — Fish Is Good for the Heart – TIME Healthland, May 26, 2011.
Want a healthier heart? Try adding fish to your diet. But be careful how it’s cooked, a new study warns: baked or broiled fish will boost heart health, but fried fish is probably better left uneaten.
Heart failure risk lower in women who often eat baked/broiled fish–American Heart Association, Press Release, May 24, 2011.
This study showed that the type of fish and cooking method may affect heart failure risk. The researchers found that dark fish (salmon, mackerel and bluefish) were associated with a significantly greater risk reduction than either tuna or white fish (sole, snapper and cod).
In a similar analysis, eating fried fish was associated with increased heart failure risk. Even one serving a week was associated with a 48 percent higher heart failure risk.
“Not all fish are equal, and how you prepare it really matters,” said Donald Lloyd-Jones, M.D., Sc.M., senior author of the study. “When you fry fish, you not only lose a lot of the benefits, you likely add some things related to the cooking process that are harmful.”
Science can be a little fishy. A good case in point is the fish consumption study from Circulation: Heart Failure, announced in an American Heart Association press release on May 24, and then picked up and repackaged by Time and other publications. The gist of the stories is that eating lots of baked or broiled fish, especially dark fish, can “boost heart health” but eating lots of fried fish may “increase heart failure risk.”
I am perfectly willing to believe — indeed, I want to believe — that eating fish is good for my heart, and I am OK with believing that baking or broiling the fish is a healthier method of preparation than frying it. But I still have a few questions.
What kind of study was this? How was it conducted?
It was an observational study (as opposed to a clinical trial) of the diets of some 84,493 postmenopausal women who ranged in age from 50 to 79. Eighty-five percent of the women were Caucasian, seven percent African-American and three percent Hispanic. The women completed detailed questionnaires about their food consumption. So, right away, we have to limit the findings to older women and mostly to older white women. Note that the lead of the Time Healthland story quoted above isn’t doing that, unless the audience of the publication consists of white women of a certain age.
Were variables controlled? If so, how and how well?
Consider this statement from the AHA press release:
Participants whose diets included more baked/broiled fish tended to be healthier and younger than their counterparts who ate fried fish. They were more physically active and fit, more educated and less likely to smoke, have diabetes, high blood pressure and heart disease (irregular heartbeat and coronary artery disease). Furthermore, their diets contained more fruits and vegetables, less unhealthy, saturated and trans fatty acids and more beneficial fatty acids, which are found in fish and in non-marine foods such as nuts, seeds and certain vegetable oils. Consumption of fried fish was associated with higher body mass index (a weight-to-height ratio), higher energy intakes (calories) and lower fiber consumption. Consumption of other fried foods besides fish was adjusted in the analysis.
Now, it would appear that there is a problem here with variables. Is (A) higher consumption of baked/broiled fish and lower consumption of fried fish causing (B) a lower incidence of heart failure? Or is there a (C) factor to consider? Such as, I don’t know, general life-style and eating patterns? Pre-existing medical conditions, maybe?
To be fair, we are told in the press release that the researchers adjusted for the “consumption of other fried foods besides fish.” The abstract of the original article further states that “associations between fish or fatty acid intake and incident HF [heart failure] were determined using Cox models adjusting for HF risk factors and dietary factors.” Additional detail about the models used for the adjustments is provided in the full text of the article. What it comes down to is that the researchers, in analyzing the data, attempted to filter out the effects of age, smoking, diabetes, high blood pressure and general diet, etc., on the heart, and thereby isolate the effects of fish consumption on the heart. Or, more accurately, to isolate the associations between fish consumption and heart failure incidence rates, which are something very different from effects. The mere existence of an association does not prove that a cause-effect relationship exists between the associated items.
The scientists themselves are careful to stop short of claiming causality. In the press release, one of the researchers is quoted as saying that baking or broiling of fish is “associated with heart failure prevention [italics mine].” The article abstract concludes, “Increased baked/broiled fish intake may lower HF risk, while increased fried fish intake may increase HF risk in postmenopausal women.” (I can only add, the moon may be made of green cheese.) The press and the public are prone to miss the nuances and weasel words like “may.” Hence, Time Healthland headline flatly proclaims that “Fish is good for the heart.”
Just how strong is the association between eating baked/broiled dark fish or eating fried fish, and HF (heart failure) incidents? The full text of the article provides the numbers. HF incidents in women who, on average, ate fewer than one servings per week of baked/broiled fish was 26.7 per 10,000 person-years compared to 19.4 per 10,000 person-years in women who ate five or more servings/week. HF incidents in women who, on average, ate one or fewer servings of fried fish per month was 20.8 per 10,000 person-years compared to 39.4 per 10,000 person-years in women who ate one or more servings of fried fish per week. No doubt the associations are statistically significant.
But is it fair to claim, as the AHA press release does, that “even one serving a week [of fried fish] was associated with a 48 percent higher heart failure risk”?
The women who ate (and probably enjoyed) frequent meals of fried fish faced 18.6 additional chances in 10,000 each year of having some type or degree of heart problem. That’s fewer than two additional chances in a thousand. Stated that way, many of us might decide the risk is worth it.
What does “fried” mean?
This is the variable that matters most to me. If I have a piece of fish and I want to cook it up, exactly what cooking technique should I avoid? The press release says this: “Two groups of fish intake were defined: baked/broiled fish or fried fish. The baked/broiled fish group consisted of canned tuna, tuna salad, tuna casserole, white fish (broiled or baked), dark fish (broiled or baked) and shellfish (not fried). The fried fish group consisted of fried fish, fish sandwich and fried shellfish.”
In my experience, the fish in a “fish sandwich” is deep-fried in batter. Usually, thick flour-based batter. The same is true for fish in a restaurant order of fish-and-chips; there is often as much deep-fried batter as there is fish. So, is this what the study means by “fried fish”? Dipped in flour, often heavily coated, and then deep-fried? And then for an added measure of carbs, sometimes squashed between the halves of a bun? When the researchers adjusted for “dietary factors” were they adjusting for the batter and bread, too?
If I were to pan fry the fish in olive oil, with no batter on the fish, would it be healthier than the kind of fried fish usually consumed by women in the study? Or what about if I carefully removed the batter from my deep-fried fish before eating the fish? Would that make a difference? Or is it that the process of frying (or deep-frying) somehow alters the fish itself, as the study seems to suggest?
These are important questions because of this claim in the press release: “. . . fish is a very good source of lean protein that we ought to be increasing as a proportion of our diet and decreasing foods that contain less healthy saturated and trans fats.”
This seems to change the agenda. All of a sudden, the focus is not on the healthy effects (or associations) of consuming Omega 3 oils in dark fish, but on the AHA’s favorite whipping-boy: saturated fat.
But is it the saturated fat that is the problem in fried fish, or is it carb-heavy batter? For that matter, can the two even be separated in an observational study? Did the researchers even try?
As far as that goes, how successful were the researchers in “modeling out” all the other known, unknown and suspected risk factors that could have skewed the heart failure rates of the groups under examination? Yes, they did try, but did it work? Can it ever really work in an observational study?
I don’t know. What I do know is that the average journalist seldom bothers to ask such questions of scientists and scientific studies.
(The inspiration for the above analysis was a viewing of Tom Naughton’s entertaining and educational lecture, “Science for Smart People.” However, any errors in the analysis are entirely my own.)
(This article was originally posted on my JimA’s blog on May 30, 2011.)
I agree – “fried” is often used to mean “heavily battered & deep fried.” Which is why I’ve taken to using the word “sautéed” for pan-fried. Except for hamburgers – they have to be fried. 🙂