A couple of weeks ago I wrote a post showing how the press fails to mention low-carb diets in weight loss stories, focusing instead on exercise or some other facet of an individual’s quest to lose weight and improve health. Today we’ll look at how the press, in an effort to minimize the untoward effects of carbohydrates on health, sometimes simply misrepresents the true outcome of studies.
A week or so ago a Swedish overfeeding study was released looking at the short term effects on the liver of a diet high in fast food. ABC News reported the study. Let’s first look at what the study was all about and what the data showed, then we’ll see how ABC reported it. By looking carefully at what ABC did to misrepresent the study, we can arm ourselves with the knowledge to identify this kind of press bias in future reporting.
First the study.
The goal of this study was to examine the effect of fast-food hyperalimentation (overeating) on liver enzymes and fat accumulation in liver cells. The specific liver enzyme in question was alanine aminotransferase (ALT), one of the liver enzymes routinely measured on standardize lab panels. Over the past couple of decades the prevalence of elevated ALT in routine labs has about doubled. Elevations of ALT are associated with an increased incidence of metabolic syndrome and all its attendant features, including non-alcoholic fatty liver disease. The Swedish researchers wanted to see if force feeding young, healthy subjects a diet high in fast food over a four week period would increase blood levels of ALT and increase the fatty infiltration of their livers.
Eighteen healthy subjects (12 males, 6 females) with an average age of 26 years were to try to increase their body weight by 5-15 percent in 28 days by adding two large fast-food meals daily in an effort to double their normal caloric intake. Researchers matched a similar control group for comparison.
The subjects took to their assignment with a vengeance. The average intake soared from 2273 kcal/day to 5753 kcal/day, a 153 percent increase! This caloric increase was driven primarily from an increase in fat, the consumption of which skyrocketed from 87 g/day to 261 grams per day (200% increase). Carbohydrate intake increased from about 275 g/day to 644 g/day (134% increase). And protein went up from 89 g/day to 180 g/day (102% increase). Sugar intake increased by 200%.
As a percentage of total calories fat rose from 36% to 43%. Carbs decreased from 48% to 45%. Protein also decreased from 16% to 12% of total caloric intake. So although the actual gram amounts of protein and carb went up as the subjects gorged on fast foods, their fractional percentage of the total calories went down.
Over the four weeks subjects on the high-fast-food diet increased their weight by 9.5 percent, their serum ALT levels by 334 percent, and the content of fat in their livers by 155 percent. The control group showed no changes.
When the researchers ran the standard statistical analysis to determine which – if any – of the specific macronutrients correlated most directly with the increase in ALT, they discovered that
the average consumption of fat or proteins during the 3 days at the end of the first or third week weeks was unrelated to changes in ALT. [Blood was drawn in the non-fasting state at the end of the first and third weeks to more accurately monitor ALT levels.] However, the maximal ALT/baseline ratio correlated with carbohydrate intake during the third week.
One could, I suppose, argue with the standard statistical analysis program the researchers used to determine this correlation, but the fact remains that the researchers did report in the study as quoted above that the increase in carbohydrate intake was the driving force behind the elevation in ALT, not the increase in either protein or fat. You might imagine that those reporting the study would mention this finding.
Before we get to the biased article by ABC, let’s take a look at a more balanced piece of reporting by Yahoo! News [The Yahoo! News link was no longer active; the link above is the same article reprinted in the The Sydney Morning Herald]. Marlow Hood, the Yahoo reporter, not only read the study but took the time to track down the author and interview him as to his findings. (As most of you who are long-time readers of this blog should know, I’m always leery of study authors’ interviews with the press because the conclusions these authors often report haven’t been run through the tempering of the peer-review process and in many cases don’t jibe with what the actual study shows. In this case, however, it did.)
Said Frederik Nystrom, the lead researcher
signs of liver damage were linked to carbohydrates was another key finding…
It was not the fat in the hamburgers, it was rather the sugar in the coke…
Dr. Nystrom also pointed out another surprising finding (surprising only to those, I suppose, who don’t understand lipid biology) that wasn’t included in this study but will be published in a future paper.
We found that healthy HDL cholesterol actually increased over the four-week period — this was very counter-intuitive.
The study showed that the increase in saturated fat correlated with the increase in healthy cholesterol.
Apparently Dr. Nystrom isn’t aware that HDL-cholesterol levels are fat dependent. In other words, increasing fat in the diet, particularly saturated fat, increases HDL-C levels.
Now let’s look at how ABC News reported this study to see a masterpiece of deception.
After a gratuitous opening to the article and the briefest of descriptions of the methods of the study, Radha Chitale, the reporter, pulls out one of the most commonly-used methods of slanting a story: she interviews someone other than the author. In this case she interviewed Dr. Kieth-Thomas Ayoob, an associate professor of pediatrics at the Albert Einstein College of Medicine. Given what Dr. Ayoob says, one really wonders if he even looked at the study or if it were merely synopsized for him by the reporter during a phone call.
Here’s how the reporter led into Dr. Ayoob’s comments.
Studies have shown that a diet high in fat and calories — the magic recipe for delicious, greasy fast food — puts people at greater risk for obesity and type 2 diabetes, both of which can lead to cardiovascular diseases and heart failure.
But the Swedish study, the goal of which was to double calorie intake and increase body weight by about 15 percent, showed that the liver is also at risk when you roll up to the drive-through window.
Here is Dr. Ayoob’s take on the study in his own words.
The extra fat is the big enchilada here, the equivalent of about three sticks of butter daily. The liver is basically using its compensatory mechanism to accommodate all this extra stuff.
Then the reporter follows up with
The liver processes fats in the blood. Excessive calories and fats overload the organ, causing fat to build up in the liver cells and leading to liver damage.
Not no where, not no how in this report does it say anywhere that carbs were the cause of the problem, which is the point the actual authors of the study were at pains to make.
If one were to read only the ABC News report, one would come away with the notion that this study shows in dramatic fashion how an increased fat intake can ruin a liver in only four weeks. But, as we saw above, that isn’t the case at all as shown by the study and confirmed by the study’s lead author. It just one more way that the media disses low-carb diets or the idea that excess carbs may actually be harmful by misrepresenting research reports.
An interesting corroborating side note to this study is found in one of the references. I’m sure these findings were a big surprise to the researchers running the analysis of the data, so they did what all researchers do in the same situation: they looked in the medical literature to try to get a handle on the situation.
One of their references is to a Johns Hopkins study published in Digestive Diseases and Sciences a couple of years ago. In this study researchers found that the fatty infiltration and inflammation of the livers of morbidly obese patients who underwent bariatric surgery correlated with carbohydrate intake.
There were no significant associations between either total caloric intake or protein intake and either steatosis, fibrosis, or inflammation. However, higher CHO [carbohydrate] intake was associated with significantly higher odds of inflammation, while higher fat intake was associated with significantly lower odds of inflammation. In conclusion, present dietary recommendations may worsen NAFLD [non-alcoholic fatty diver disease] histopathology.
On another note, the Swedish fast-food study provides us with much more interesting material than the fact that excess carb intake runs up serum ALT levels. In fact this study gives us a glimpse into the subject of the existence of a metabolic advantage. It will be the subject of the next post.
Note: Here’s another take on this Swedish study from one of the commenters on this blog.