Diet drinks may help with weight loss more than water


One problem with this study: it was funded by the American Beverage Association. They claim that there was no influence from that.

Drinking diet beverages helps people lose weight. In fact, it does a better job than water. That’s the conclusion of a groundbreaking study from University of Colorado researchers.
CBS4 Health Specialist Kathy Walsh first reported about the study when it started nearly two years ago, and now it’s got a surprising result. It’s good news for diet soda lovers who also want to drop some pounds.
This study counters the idea that diet drinks stimulate hunger. In fact, it suggests they can play an important role in weight loss as CU researchers concluded drinking diet beverages helps people lose weight.
“What we found, surprisingly, was that diet beverages actually performed better than water,” Dr. Jon Peters of the Anschutz Health and Wellness Center said.
Walsh first reported on the CU study in December 2012 when Kristi Norton was enrolled. She and about 300 others followed a diet and exercise program, but they were divided into two groups — half could drink zero-calorie beverages while the others strictly water. Norton was in the water only group.
The theory was that both groups should lose the same amount of weight since all of the drinks were calorie free. Surprisingly, that wasn’t what happened. The study found those who drank the diet beverages lost an average of 13 pounds in 12 weeks. The water drinkers lost an average of 9 pounds.
The diet drink group felt less hungry and showed greater improvement in “bad” cholesterol.
“Perhaps it helps people not have to give up that little bit of pleasure in their life when they’re otherwise trying to restrict their calories,” Peters said.
But Peters says it’s not healthy to drink a Big Gulp a day to keep the pounds away, however it does counter the notion that diet sodas stimulate hunger and contribute to obesity.
“I’d say diet sodas can be an effective tool in a weight loss program,” Peters said.
Norton said she lost 12 pounds — 3 pounds above the average, which proves there’s an over achiever in every group.
The study was funded by the American Beverage Association, but researchers were quick to point out none of its members were involved.
The takeaway is that willpower is limited because a person is already being denied calories. The finding is people don’t have to also kick the diet soda habit to drop weight.


I am sorry but I am not buying that for one second.


I’d want to know what sweetener was in these drinks. Were they aspartame, Stevia, Splenda… what?


I’m immediately skeptical, since the study was funded by the American Beverage Association.


As someone who drinks what seems like gallons of diet soda every day (I’m not a coffee drinker), and someone who could certainly stand to lose some weight, this sounds like complete horseshit.

Diet sodas, because of their artificial sweeteners, acclimate the body to empty calories (foods which are high in sugar/carbs but are not filling), even though they contain none of their own. Even at it’s most optimistic, this study does very little to challenge that finding as the story claims, since the study didn’t actually measure or investigate that effect in isolation.

There’s not much in the articles on this, but the study appeared to use a crude, simplistic methodology (similar to market research techniques) and the sample size related to the margin of the results seems to render it utterly insignificant (Average weight loss? Could a small group in either group be skewing the results? What were the outliers, the median, etc).

It’s wishful overinterpretation at best, and at worst (and most likely) it’s that patented brand of corporate-sponsored, predetermined-results-oriented research. The conclusion is basically “maybe it’s magic.” :

“Perhaps it helps people not have to give up that little bit of pleasure in their life when they’re otherwise trying to restrict their calories,” Peters said.

What PR hack wrote that line? But it on a bumper sticker!.


My first thought is I wonder how dehydrated the diet soda drinkers were, water wieght could easily account for that difference


I’ve read a lot of articles on weight loss and they all say that calorie intake is the key, not carbs, fats or protein(although proper balance helps), but calories.

So this sounds like complete bullshit.


There’s plenty of evidence to the contrary. The study fails to account for a number of variables.

ADDITIONALLY. If it were a test of will power like the test says, then greater glucose levels in the brain would be better and not worse. However this is not what happens when artificial sweeteners are ingested. There is an insulin response which lowers the amount of blood sugar, so then it would follow that it would cause a decrease in will power rather than an increase.

Not that this is anyway factual of how will power works, since no one knows how it actually works. We just know using it causes blood sugar levels to drop.


Here’s a link to the actual study (PDF) for anyone interested in actually reading it:


Ah, now I see it:

Bogus study said:

Eligible participants had to report […] consuming at least 3 NNS beverages per week.

So basically they took diet soda drinkers, told half of them they had to drink only water instead, and then reported that the soda group “reported significantly greater reductions in subjective feelings of hunger than those in the water group.”

This is, seriously, an insane methodology. Of course, they do “address” this concern:

Bogus study said:

It should be noted that because eligible subjects were already NNS drinkers assignment to the NNS treatment did not require as great a behavior change as the Water group who had to abstain from NNS beverages for the trial. We chose this design rather than admitting all comers in order to ensure that subjects assigned to NNS would adhere to the treatment giving us the ability to see if NNS adversely affected weight loss.

So, they say, it should be noted, but you know, quickly ignored.

The problem here is that with a large enough sample size, they should not have been overly-concerned with maintaining adherence from the NNS group. If some in that group who were not regular NNS drinkers dropped out or had low adherence as a result, those results could either be filtered out, or used to draw secondary conclusions.

They don’t even bother to claim that this decision provided any benefit to the robustness of the results, just that it was a convenient way for them to get enough participants.

The cynic in me says that they also knew that it would weed out people who might complicate or put at risk the result their sponsors were looking for.

This “glossing over” of this element represents a serious intellectual dishonesty on the part of the researchers. Their results didn’t just say that water isn’t better, it explicitly suggests the use of NNS over water for a weight loss plan, and with no caveats. Shouldn’t results from people who do not regularly consume diet drinks be a integral part toward drawing that conclusion?

But wait, there’s more!

Bogus study said:

Participants were given manufacturers coupons weekly (from the three largest beverage manufacturers: The Coca-Cola Company, PepsiCo and Dr Pepper Snapple Group), redeemable for bottled water or NNS beverages at retail stores.

The soda group got their “NNS” drinks paid for, while the water group got free water. Another subtle, non-real-world variable that could certainly alter the results.

I don’t mean these participants were paid off or anything, but this is bad methodology posing as controlling for variables. Simply put, the free drinks introduced a factor that would not exist in a real-world environment or be part of anyone’s actual diet plan, and since the control group by design could not receive an equivalent “perk,” it becomes an uncontrolled variable.

The results actually point out that the study makes no claims to the literal power of diet drinks to reduce weight, so variables like pleasure, cost, reward, effort, etc, are very important to the validity of these results.

Bogus study said:

Study dropouts were similar across […] the treatment groups (5.8% for NNS, 10% for water).

What? Nearly twice as many in the water group dropped out of the study. This not only meant that the NNS group had a larger sample size (which reduces the effect of outliers), but it introduces a question of why the water group were more likely to drop out. Since everyone in both groups was a regular diet soda drinker, it could be that denying something they usually have was a factor in the higher dropout rate, and would also effect the remaining participants to a degree. You could also make the argument that the dropout rate is a plus for the effectiveness of drinking diet sodas to keeping on a weight loss plan, but only if you can positively correlate dropping out of the study with dropping out of a weight loss plan (which seems impossible to prove).

And there’s this nugget, when comparing the larger percentage of people in the NNS group who lost at least 5% body weight:

Bogus study said:

Analysis includes those participants who dropped out of the study in the analysis, using the baseline observation carried forward.

This means that the larger number of discouraged water group dropouts were given a score of 0 and were included in the results, even though they did not actually participate. That is an incredible level of bullshit. These numbers are basically lies.

In the end, they come right out and say it themselves:

Bogus study said:

Based on the design of this study we are unable to say, what is the mechanism for the greater weight loss in the NNS group compared to the water group.

Of course, my original complaint remains. Where is the x/y plot of participant results? How did the groups compare in terms of distribution of results? Where are the outliers (if there were any), and what were the medians? Which participants lost the most weight, and the least? Did anyone gain weight? Of those, which group did they belong in?

The study does not include the complete data set, which is very telling.


Excellent detailed analysis. Bravo!