"A wise man proportions his belief to the evidence."
-- David Hume
According to the hype generated by the Center for Science in the Public Interest (CSPI) surrounding the Institute of Medicine's report "Food Marketing To Children and Youth: Threat or Opportunity?," the report provides clear and compelling evidence that causally links food advertising with childhood obesity. CSPI has gone so far as to sue Viacom and Kellogg's based mostly on that report.
Read the actual report, though, and you get a different impression than that the committee, IOM and CSPI seem to be pushing in their rush for restrictions on food advertising.
The report correctly states that the "current evidence is not sufficient to arrive at any findings about a causal relationship from television advertising to adiposity among children and youth."
Rather than state the scientific case, though, Michael McGinnis, the report committee's chair, claimed, "There is strong evidence that television advertising influences the diets of children." And this, he avers, puts "kids' long-term health at risk."
But if McGinnis is any kind of scientist, he must know that this is a total red herring that he is trying to get the press and public to swallow.
The Red Herring
Science at bottom is about demonstrating causal connections, connections that are measurable, reproducible and specify their margin of error. The "evidence" assembled by the IOM, by its own admission, fails to meet this standard of legitimate science.
Instead of telling us that, McGinnis tries to cover it up, by claiming that "advertising influences diets." In doing so, Mr. McGinnis and his committee are being either silly or disingenuous.
Anyone who knows anything about advertising would tell you that advertising influences choices. So, of course, food advertising influences children's diets.
This is not the not the same thing, though, as the claim that food advertising causes childhood obesity or even that food advertising causes children's diets. The IOM takes the answer to one trivial and uninteresting question -- does advertising have any effects? -- and uses it to answer another more controversial one -- does advertising cause childhood obesity? This is a disreputable bit of sleight of hand aimed at deceiving the public and press about what its report scientifically shows.
The committee then draws the public and press to conclusions about the need for either self-imposed or governmental restrictions on advertising that are disproportionate to its evidence.
Mostly, the report demonstrates ignorance about the relevant trends in advertising and children's food intake, what the scientific literature says about the link between specific foods and obesity, and, finally, what studies of the effects of food advertising actually show.
Some Disconfirming Trends
Let's begin by looking at the relevant trends. First, according to the Federal Trade Commission, advertising during children's TV programming has declined by 34% since 1977. Second, data from Nielsen surveys shows that food advertising on television has declined by 13% since 1993. Nor are these advertising trends confined to the US. In the United Kingdom, ad spending on food and drink has been falling in real terms since 1999 and is now at roughly 1982 levels. In 1982, food and drink ads constituted 34% of the total TV advertising in the UK whereas in 2002 they were 18%. (Tim Ambler, "Does the UK promotion of food and drink to children contribute to their obesity?" London Business School, 2004.)
So the first relevant trend to note is that food advertising to children is declining at the same time that childhood obesity is increasing -- a peculiar and inconvenient fact for those claiming food advertising causes childhood obesity.
A second relevant trend is total caloric intake in children and young people.
If children are in fact getting fatter from eating advertised foods, then one would expect that their caloric intake would be increasing. Yet, caloric intake for children and young people has not changed significantly over the period in which food advertising has supposedly made them fat. As former FDA Commissioner Mark McClellan observed in 2003: "[I]t's perhaps surprising that, in a debate that has often focused on foods alone, actual levels of caloric intake among the young haven't appreciably changed over the last twenty years."
Or consider a recent article in the Lancet (S.Y. Kimm, et al, 2005) which noted that "The composite findings from NGHS so far indicate that the drastic decline in habitual activity during adolescence might be a major factor in the doubling of the rate of obesity development in the USA in the past decades, since no concomitant increase in energy intake was apparent."
The same finding about no increase in calories is found in a study of energy and fat intakes of children by Richard P. Troiano et al. in The American Journal of Clinical Nutrition (2000) that noted, "The lack of evidence of a general increase in energy intake among youths despite an increase in the prevalence of overweight..."
Putting these two trends together, the IOM report becomes something of an Alice in Wonderland proposition -- Declining food advertising + no increase in kids' caloric intake = fatter kids from food advertising.
Advertised food, diets and obesity
The IOM report also fails to take account of what the science actually says about the connection between specific diets and obesity in children and adolescents.
For example, there is no clear and consistent connection between fat intake and obesity in children. As Professor David Ashton from the Imperial College School of Medicine notes "Epidemiological studies do not show a consistent association between dietary fat and adiposity in children and young adults." (Journal of the Royal Society of Medicine, 2004)
Though there is much speculation about the link between dietary fat -- which is the most energy dense nutrient -- and obesity, the relationship between the two is not scientifically established. Even often uncritical exponents of the childhood obesity epidemic such as David S. Ludwig and Cara B. Ebbeling admit: "Findings of epidemiological studies do not consistently show an association between dietary fat and adiposity in children and young adults. Moreover the prevalence of obesity has increased despite an apparent decrease in proportion of total calories consumed as fat in the diet of US children." (Ebbeling et al. Lancet, 2002) Less fat, more obesity? That is very odd indeed.
Perhaps, the culprit in the obesity equation is not fat at all, but rather carbohydrates -- kids compensating for lower fat intake by taking in more cereal, pasta, bread pastries and soft drinks. One theory is that such high glycemic index foods lead to hormonal changes which in turn cause hunger and lead to overeating.
Only there are no clinical trials that substantiate this thesis. As important, a number of studies work against the claim that such heavily advertised carbohydrates as fast foods and soft drinks lead to fatter children.
A recent study from the University of Alberta (Paul J. Veugelers and Angela L. Fitzgerald, "Prevalence of and risk factors for childhood overweight and obesity," Canadian Medical Association Journal, September 2005) found that eating in a fast food restaurant was not a statistically significant risk factor for childhood obesity. It also reported that there was not a statistically significant association between the availability of soft drinks at school or schools with food vending machines and the risk of children being overweight or obese.
These findings were similar to a Harvard project (A.E. Field et al., "Snack food intake does not predict weight change among children and adolescents," International Journal of Obesity and Related Metabolic Disorders, 2004) that followed nearly 15,000 U. S. schoolchildren over a three-year period. It found that however junk food was defined, with or without soft drinks, it did not lead to obesity in children.
Again, in a just published study by Richard Forshee, deputy director of the University of Maryland's Center for Food, Nutrition and Agriculture Policy, funded by the American Beverage Association, (Risk Analysis, October, 2005), the authors, applying risk analysis to two federally funded data sets -- the National Health and Nutrition examination Survey NHANES and National Family Opinion -- found "no statistically significant association [between soft drink consumption and BMI] and, in fact, regular carbonated soft drinks accounted for less than 1 percent of the variance in BMI."
Another ABA funded study found the "typical high school student purchased at most about one twelve-once can of traditional CSD [carbonated soft drink] per week at school. The typical elementary school student purchased at most only about one can of traditional sugared CSD per school YEAR."
The IOM also leaves unexplained the real world experiences of places such as Quebec that banned all food advertising to children in 1980 and Sweden that banned them in the early 1990s. Neither has obesity rates differing markedly from other similarly situated provinces or countries.
As Nobel Prize winning economist Gary S. Becker of the University of Chicago noted on his blog with Richard Posner on Dec. 11:
"[D]espite the hype the study received, the Institute of Medicine's report on obesity and advertising did not present any convincing evidence that television advertising oriented toward children has been responsible for the increase in children's obesity during the past quarter century."
Red herrings and sleight of hand do not a scientific report make. The IOM and its trumpet CSPI are making nutrition research a scientific joke.