Diet drinks and stroke

From

Soda Consumption and Risk of Vascular Events in the Northern Manhattan Study
Hannah Gardener, Tatjana Rundek, Clinton Wright, Univ of Miami Miller Sch of Medicine, Miami, FL;
Julio Vieira, Mitchell S Elkind, Columbia Univ, New York, NY;
Ralph L Sacco; Univ of Miami Miller Sch of Medicine, Miami, FL

Previous studies have shown a positive association between diet soda consumption and risk of the metabolic syndrome and diabetes, which are important vascular risk factors. The goal of this study is to examine the association between soda consumption and risk of stroke, myocardial infarction (MI), or vascular death. The study population included participants from the Northern Manhattan Study, a multi-ethnic population-based cohort study to examine the incidence and risk factors for stroke, with data on soda consumption available (N=2564, 36% men, mean age 69 +/- 10 years, 20% white non-Hispanic, 23% black non-Hispanic, 53% Hispanic). Diet and regular soda consumption were assessed by self-report using a semi-quantitative food frequency questionnaire at baseline, and were examined categorically: no soda (<1/mo) (N=901, referent), moderate regular only (1/mo-6/wk) (N=761), daily regular only (N=286), moderate diet only (N=214), daily diet only (N=116), moderate diet and any(1/mo) regular (N=239), daily diet and any regular (N=47). Cox models were used to examine the association between soda consumption and incident vascular events (stroke, MI, or vascular death) controlling for sociodemographics and vascular risk factors. During a mean of 9.3 years of follow-up, 559 vascular events accrued. After controlling for age, sex, race/ethnicity, smoking, physical activity, alcohol consumption, and calories consumed per day, a 61% increased risk of vascular events was observed among those who drank daily diet soda only as compared to those who drank no soda (95% CI 1.13-2.30), and this association persisted after additionally controlling for the metabolic syndrome, peripheral vascular disease, and cardiac disease history (RR=1.48, 95% CI 1.03-2.12). A marginally significant increased risk of vascular events was also observed among those who consumed diet soda daily and regular soda once or more per month (adjusted RR=1.74, 95% CI 0.96-3.16). There was no significant increased risk of vascular events among those who consumed moderate or daily regular soda only (vs. no soda: moderate regular only adjusted RR=0.93, 95% CI 0.74-1.16; daily regular only adjusted RR=1.20, 95% CI 0.91-1.59). This study suggests that diet soda is not an optimal substitute for sugar-sweetened beverages, and may be associated with a greater risk of stroke, MI or vascular death than regular soda. Further studies are needed to explore potential mechanisms for the association between diet soda consumption and vascular events.

Author Disclosures: H. Gardener: None. T. Rundek: None. C. Wright: None. J. Vieira: None. M.S.V. Elkind: None. R.L. Sacco: None.

Debunking

This study was merely presented at a conference. It has not been published in a journal and there is no indication that it has been peer-reviewed.
The reported 61% increase in relative risk for vascular events among regular diet soda drinkers is a weak statistical association (i.e., less than 100%) and so is inherently unreliable and meaningless.
The claimed consumption of diet soda by study subjects occurred once at the beginning of the study and was self-reported; so the researchers really have no idea how much diet soda any study subject actually consumed.
The study subjects are elderly (mean age 69 years) and already at increased risk of stroke.
The study subjects are from Northern Manhattan, which indicates they were likely poor and not well-educated — both risk factors for the reported health effects.
The researchers report that the statistical association weakened as more confounding risk factors were considered — meaning that when all confounders are considered, the association will likely disappear.
Contrary to the researchers' claim that previous studies link diet soda consumption with diabetes, the authors of an April 2009 study in Diabetes Care concluded that their data, "cannot establish causality" between consumption of diet soda and metabolic syndrome and type 2 diabetes.

This study is so bad that even the Los Angeles Times debunked it.

Here's the American Beverage Association statement:

In response to an abstract presented today at the American Stroke Association's International Stroke Conference 2011, Dr. Maureen Storey, senior vice president of science policy for the American Beverage Association said:

"There is no scientific evidence to support the idea that diet soda uniquely causes increased risk of vascular events or stroke. It is important to recognize that this is an abstract - not a peer-reviewed, published study - that was presented at a conference; however, even the abstract itself does not support a link between diet soda consumption and risk of stroke. In fact, the partial statistical analysis shown in the abstract does not show significant stroke risk between those who drank diet soda daily and those who drank diet soda less often.¹ Furthermore, it appears that the investigators failed to control for two important variables - family history of stroke and weight gain - in their analysis.

According to the National Heart Lung and Blood Institute of the National Institutes of Health, heart disease and stroke risk factors that can be controlled are high blood pressure or hypertension; abnormal blood cholesterol levels; tobacco use; diabetes; overweight; and physical inactivity. Risk factors beyond our control include age, that is being 55 years and older for men and 65 years or older for women; family history of early heart disease; or family history of stroke.

The body of scientific evidence does show that diet soft drinks can be a useful weight management tool, a position supported by the American Dietetic Association. Thus, to suggest that they are harmful with no credible evidence does a disservice to those trying to lose weight or maintain a healthy weight. Importantly, our industry supports and encourages healthy lifestyles by providing consumers with myriad beverage choices in a wide range of calories so they can choose the beverage that is right for them."

¹Stroke 2011, 42(3):e273; pg. 163; "A marginally significant increased risk of vascular events was also observed among those who consumed diet soda daily and regular soda once or more per month (adjusted RR=1.74, 95% CI 0.96-3.16)."

Here's the Calorie Control Council statement:

The Calorie Control Council stated today that research findings presented during a poster session at the International Stroke Conference claiming an association between diet soft drink consumption and increased risk of stroke and heart attack are critically flawed.

"The findings are so speculative and preliminary at this point that they should be considered with extreme caution. In fact, the study has not been peer reviewed by any independent scientists and has not been published in a scientific journal," stated Beth Hubrich, a registered dietitian with the Council.

The research, as well as the publicity regarding the study abstract, is drawing a growing body of criticism and skepticism from experts in the field of nutrition and science.

"I have to say this is one of the worst studies I've seen capturing headlines in a long time," said Dr. Richard Besser, Chief Health and Medical Editor at ABC News, commenting during a February 10 segment on Good Morning America. "It's bad because of the science, but it's also bad because of the behavior that it can induce and the fear that people have. I don't think people should change behavior based on this study."

Pointing out some of the flaws in the study, Besser added, "They didn't look at how much salt they took in, they didn't look at what other foods they ate. Those things we know are associated with stroke and heart attack. They didn't even look at obesity over time. And so to conclude from this, that it's all from the diet soda, just makes no sense whatsoever."

Dr. Walter Willett, chair of the Department of Nutrition at Harvard School of Public Health, echoed this criticism in comments presented on ABCnews.com (http://abcnews.go.com/Health/video/taking-diet-soda-study-grain-salt-12870775): "It's important to keep in mind that his was really a preliminary report. It's not even published yet, and the study was fairly small. I think we have to interpret the findings about diet soda very carefully, in almost any first report, we shouldn't really change our behavior, because it could easily have occurred by chance."

The poster presentation is also at odds with statements on the website of the American Heart Association, sponsor of the conference where this poster was presented. Regarding the low-calorie sweeteners used in diet soft drinks, AHA states: "Try non-nutritive sweeteners such as aspartame, sucralose or saccharin in moderation. Non-nutritive sweeteners may be a way to satisfy your sweet tooth without adding more calories to your diet. The FDA has determined that non-nutritive sweeteners are safe."

Among the weaknesses and criticisms of the poster:

The poster is not a published study and has not been peer-reviewed for a scientific journal. As a result, it has not been examined by experts as to design, reasonableness of conclusions, and other methodological issues. The academic standards for poster presentations are lower than for peer-reviewed scientific journals, calling the conclusions into greater question.
The soda intake data used for this presentation are of questionable reliability. They are based on self-reported consumption figures, which are grouped into general categories, such as "moderate diet only" soda consumers. This imprecision makes it very difficult to draw any meaningful conclusions from the study.
This study is observational and does not show cause and effect. As a result, the presenters' conclusion that diet soda "may be associated with a greater risk of stroke, MI or vascular death than regular soda" is not well founded. For example, the authors do not appear to have controlled for such critical factors as family history of heart disease or stroke, or for weight gain. Also, stroke is more common in men 55 years of age and older and women 65 years or older. The average age of those in the study was 69.
The sample size of those who claimed they drank diet soda daily was small – only 4.5 percent of the total study sample.
The overall study population was not representative of the U.S. population, either in age (average age was 69), or race (only 20 percent non-Hispanic white).
Cardiovascular events are very complex statistically. Many people who have had a cardiovascular event (and therefore are a much higher risk for a second, statistically) are switched to a low calorie diet which will often include diet drinks as opposed to high calorie drinks.
The authors offer no explanation or theories for why diet soft drinks might be related to an increased risk of vascular events.
Low-calorie sweeteners are some of the most thoroughly studied food ingredients in the food supply. The safety of low-calorie sweeteners has been reaffirmed time and again by leading health and regulatory groups worldwide.

Diet drinks and stroke

From

Contents

Summary

Study Description

Debunking

Related Topics

Views

Personal tools

Navigation

Search

Toolbox