Sweet Drink/Heart Failure Study Has No Pulse

This one will certainly make headlines because of its premise that sweet drinks will kill you in some way. Yawn. Nothing new.

But we expect more from a paper that was just published in the new British Medical Journal, Heart, since it comes from the highly-regarded Karolinska Institute in Sweden. It is entitled "The relationship between sweetened beverage consumption and risk of heart failure in men," so it is not completely absurd to expect some kind of an answer.

Lead author Dr. Susanna C. Larsson claims, "Our study findings suggest that sweetened beverage consumption could contribute to HF [heart failure] development. These findings could have implications for HF prevention strategies."

Don't hold your breath.

The study has a few flaws that are so bad that any of them let alone all of them together relegate this study to the scientific scrapheap. One need only take a look at either the design of the study, or the statistic of the outcome, to see that the whole thing is full of holes.

The study was retrospective (data dredging), which automatically cheapens the integrity of the conclusion. The data were obtained from THE Swedish National Patient Register and the Cause of Death Register, and consisted 42,400 men, all of whom were between the ages of 45 and 79 at the beginning of a 12-year study.

The rates of heart failure in men who drank two or more sweetened beverages per day was compared to that of men who did not. As indicated on the graph below, there appears to be some sort of correlation between the two groups:

Now, the flaws:

• The number of sweetened drinks was determined by a single 1997 questionnaire in which the men were asked: How many soft drinks or sweetened juice drinks do you drink per day or per week?

Self-reporting is a notoriously unreliable method of gathering data. And, the author's conclusions depend on one collection from 1997, which assumes that whatever behavior was reported at that time was assumed to continue for 12 years.

• Not only was there no distinction made between sugar-sweetened and diet drinks, but fruit juice which is every bit as sweet as soda was excluded: "Fruit juice was not included in our definition of sweetened beverages," the study noted.

Why? Is the sugar in fruit juice not sweet? Would the conclusion have been the same had it been included? The opposite? This makes no sense.

• The magnitude of the difference between the groups was 21 percent, which is meaningless in retrospective studies. It is generally accepted that a two-fold difference may be meaningful in determining a trend.

In the absence of a greater difference between the two groups, the chances are high that the data are no more than statistical noise.