Well Noted: Because Nutrition Research Was Always This Way
In previous columns I’ve written about the persistence of certain health and nutrition myths, especially the salt and sodium one, that healthy adults are at risk of developing high blood pressure based on salt intake. But there are others.
By David Feder, R.D., Editor | 10/04/2007
What could sound like heresy to anyone claiming to have human health interests at heart, but which is nevertheless borne out by science, is the connection between total dietary fat intake, blood cholesterol levels, obesity and disease/mortality is not 100-percent certain.
Before you send the lynch mobs with torches, I would like to stress the myth here is not that science has found no connections, it’s that science is neither definitive nor monocular. The inescapable truth is there are hundreds of studies which conclude the connections between dietary fat and disease are not as strong as any self-deputized food police want them to be.
A scan through PubMed, Medline or other nutrition science abstract sites will reveal several things: There is no one single guarantor of heart disease. Epidemiology is a science of statistics, not absolutes. In other words, not everyone who is overweight gets heart disease. Not everyone who eats a high-fat diet gets sick, and certainly high blood cholesterol is not a death sentence. You’ll also find proof of my own favorite adage, used in these pages before: For every Ph.D., there is an equal and opposite Ph.D.
Last spring the New England Journal of Medicine published “Dietary fat and weight gain among women in the Nurses' Health Study.” One of the study’s conclusions read, “Our results show that, overall, percent of calories from fat has only a weak positive association with weight gain.”
Let’s be fair: One of the other conclusions was that “percentage of calories from animal, saturated, and trans fat has stronger associations.” Yet an earlier study, this one published in the American Journal of Clinical Nutrition in November, 1997, determined "The effects of trans fatty acids on high-density lipoprotein cholesterol and lipoprotein(a) concentrations are unclear because of limited and conflicting clinical data.”
We’ve known about the French Paradox for decades — the fact that the French diet, higher in fat and alcohol than many other global diets has not translated to higher death rates from heart disease. Similar paradoxes spring up throughout the Mediterranean.
When World Health Organization data in 2000 questioned the validity of the paradox, more research showed more parity than previously believed between European rates of heart disease and ours. But what had also crept into these cultures was different lifestyles and calorie sources, and decreased physical activity. Back to Square One.
A comment on the aforementioned 1997 study that concluded there is no relation between fat intake and body-mass index included a quote I believe should be the mantra for anyone studying or reporting on nutrition and health — and that means ingredient cause and effect — issues: “The data as presented also support the need for caution in drawing interventional conclusions from cross-sectional observational data.”
Even though the author of the comment (“Dietary Fat Intake and the Risk of Coronary Heart Disease in Women,” New England Journal of Medicine, March 26, 1998), Ira S. Ockene, M.D., University of Massachusetts Medical Center, was suggesting aspects of the study’s data might be flawed, the urge to not jump to conclusions could not have been phrased more succinctly.
Another reason — perhaps the biggest one — for the confusion is that which slips through the cracks of virtually every popular media translation of nutrition science. That is the difference between differences. In a previous column I mentioned the definition of the word “significant.” In everyday use, it means “a whole lot.” In science, it means “measurable.”
This is how salt became demonized. In some studies, although by no means all but especially the infamous DASH II, the difference in blood pressure between a low-salt and a high salt diet was all of a few points — about the same as you’d get if you had your blood pressure taken at one clinic, then sauntered over to another and had it taken there. But the differences were measurable, therefore “significant.” Forget the fact the healthy higher sodium diet showed one of the smallest differences, it showed a measurable one and so the floodgates opened.
And all this is where we in the food processing business have a problem. We are trying to build entire lines of food and beverage products on the inexact science of nutrition. As all the above shows, this ain’t easy. So what can we do about those who make it harder — the self-appointed food police who seem to have more of the popular media’s ear than their counterparts. It’s the squeaky wheel paradigm at work.
It is incumbent upon processors to make better use of the same media. The people demanding change after change in the foods we manufacture based on their credentials as experts (with and without quote marks) and invocations of science deserve a taste of their own hubris. Processors must be prepared to attack bad science first, and with real science.
A solid, proactive approach that reaches out to popular media with the facts is needed. It’s time to give up on the industry standard of a reactive approach that layers a media-manufactured controversy into “we said/they said” sound bites. That latter, typical tactic is one processors lose every time.