Austrian Vegetarian Study Making Waves

Summary

A CBS Atlanta story is promoting a poorly designed cross-sectional study from Austria as though it is the best evidence we have regarding vegetarian diets.

As I hope most of my readers know, I do not quickly dismiss a study just because I don’t like the results.

In 2007, when EPIC-Oxford found a higher risk of fractures among vegans who didn’t get over 525 mg of calcium day, rather than finding some limitation of the study which is always possible to do, I started emphasizing that vegans need to get more than 525 mg of calcium per day. There are many other examples.

So when I say a study whose results I don’t like is pretty much useless, it’s because the study is pretty much useless and not because I don’t like the results.

In December 2013, a cross-sectional study from Austria aiming to compare the health status of vegetarians to other diet groups was released (1), and I wrote about it in the post Austrian Vegetarians: Good News? It was, with all due respect to the researchers, one of the most oddly designed studies I have seen to describe vegetarians.

In the study, the diet categories were:

– Vegetarian (vegans, lacto-ovo-vegetarians, and pesco-vegetarians)
– Carnivorous diet rich in fruits and vegetables
– Carnivorous diet less rich in meat
– Carnivorous diet rich in meat

The researchers didn’t define these categories for the participants when they were asking them which category they belonged to.

The researchers also created a number of health indicators that I didn’t feel confident in even though they concluded that vegetarians had the best self-rated health and the lowest incidence of chronic conditions.

Then on February 7 2014, another paper from this same study was published (2). I read the abstract and saw that their conclusions were somewhat different in the more recent paper, and less favorable to vegetarians, but given the boondoggle that I considered the study to be, I put it aside with no intention of doing another write-up.

Fast forward to April 1, when an article about the February paper was published on CBSlocal.com (Atlanta), Study: Vegetarians Less Healthy, Lower Quality Of Life Than Meat-Eaters. This CBSlocal article made the rounds quickly and so I decided it was time to comment on the study.

My criticisms of the February paper are pretty much the same as for the one from December. However, the February paper had more information.

With the diet categories so poorly designed, it’s surprising that they found a number of statistically significant differences in disease incidence between the groups. In comparing the vegetarian group to the carnivorous diet rich in meat group, the vegetarians had a higher rate of allergies, cancer, and mental illness, while the rich meat group had a higher rate of urinary incontinence. Asthma, diabetes, cataracts, hypertension, heart attack, stroke, osteoporosis, among other diseases, were not significantly different across diet groups.

As for the self-reported quality of life scores, some of the information didn’t match between the two papers; for example, the December paper lists the chronic conditions score for vegetarians as 1.45 while the February version lists it as 1.29 (lower is worse), a meaningful difference in their scheme.

Enough said about the specifics of this study, other than that the February paper is available for free at the link from the citation below, so if you want to check it out on your own you can. But quibbling over the details of either of these papers is fairly pointless. As the authors state in their relatively long section on the limitations of their study (2):

“Potential limitations of our results are due to the fact that the survey was based on cross-sectional data. Therefore, no statements can be made whether the poorer health in vegetarians in our study is caused by their dietary habit or if they consume this form of diet due to their poorer health status.”

They also bring up my earlier criticism:

“Further limitations include the measurement of dietary habits as a self-reported variable and the fact that subjects were asked how they would describe their eating behavior, without giving them a clear definition of the various dietary habit groups.”

The study from Austria, with all it’s limitations, is one thing. But the article from CBS Atlanta (link), adds insult to injury.

The CBS Atlanta article suggests that the Austrian study indicates causation (“But the vegetarian diet…carries elevated risks of cancer, allergies and mental health disorders”) and ignores a huge body of much better evidence regarding vegetarian diets, making it seem like this Austrian study is all we have to go on.

In fact, as most of my readers probably already know, much better studies following vegetarians over time have shown them to have equal or better health than regular meat-eaters in a number of diseases. You can read all about those studies in the VeganHealth.org section, Research on Vegetarians and Vegans.

For those not familiar with this research, I will point you to the fact that vegans have been found to have only a fraction of the risk of developing type 2 diabetes compared to regular meat-eaters (Type 2 Diabetes and the Vegan Diet).

Support JackNorrisRd.com

If you like my posts, please like my posts! Or share them. Thank you!

I greatly appreciate donations of any amount (click here).

Purchase anything through these links and JackNorrisRD gets a percentage:

References

1. Burkert NT, Freidl W, Großschädel F, Muckenhuber J, Stronegger WJ, Rásky E. Nutrition and health: different forms of diet and their relationship with various health parameters among Austrian adults. Wien Klin Wochenschr. 2014 Feb;126(3-4):113-118. Epub 2013 Dec 17. | link

2. Burkert NT, Muckenhuber J, Großschädl F, Rásky E, Freidl W. Nutrition and Health – The Association between Eating Behavior and Various Health Parameters: A Matched Sample Study. PLoS One. 2014 Feb 7;9(2):e88278. | link

6 Responses to “Austrian Vegetarian Study Making Waves”

  1. Dan Says:

    I agree with you. The fact that it was cross-sectional and observational makes any statistically significant results into a questionably clinically significant conclusion. Adventist Health Study-2 was at least longitudinal, meaning exposure preceded outcome. It could still be subject to major unmeasured confounding factors but here we don’t even have a longitudinal component – the patients could have selected veg. diets because they had poorer health to begin with (‘reverse causality’).

  2. Andreas Thaler Says:

    Hello Dr. Norris,

    This study has been widely discussed in German-speaking nutrition forums, so allow me to add a few points here.

    In all fairness, the study authors are unhappy (and have stated that publicly) with the simplifying (and incorrect) interpretation of their study, given that they clearly stated in their preamble that it can NOT be derived from the results of the study that living a vegetarian lifestyle would lead to lower health. This mainly stems from a German pro-meat nutritionist who cherry-picked the results of the story and widely issued a press communique stating “Vegetarians are less healthy than meat-eaters”.

    Do not blame the journalists for taking the wrong conclusions from reading the study, they did not. They simply copy-pasted that pro-meat press communique without giving it any own research or critical reasoning (THAT, on they other hand, they should certainly be blamed for).

    Some further points from the study design:

    – The base group from which the participants of the study were selected comprised a total of about 15,000 people. There were too few vegans (31) and vegetarians (125), so they added people who eat fish (187) to the “vegetarians” group to have a sufficiently large group. So about 55 % of those “vegetarians” are actually omnivores.

    – The study is a “pattern matching study”. To compare similar BMIs and similar habits (e.g. smoking, alohol consumption etc.) they matched each “vegetarian” with a meat-eater of the same BMI and habits (i.e. they matched the “vegetarians” with healthy meat-eaters). While this approach makes sense if you want to specifically find out whether a fit vegan is as healthy as a fit omnivore, it is maybe a bad foundation for a study that an overweight omnivore will then shove into your face with the words “SEE! Told you! I am healthier than you!!!”
    Or, in some cases, when authors of articles on the study insinuate that vegans and vegetarians are less healthy, DESPITE having lower BMI etc. (It is exactly the other way round)

    – There was also the comment from vegan experts that it is not a big surprise that vegETRIans who eat lots of dairy products live actually less healthy than omnivores, given that milk and cheese are not exactly the healthiest foodstuffs, despite dairy industry marketing.

  3. Heather Says:

    Thanks for the information!

  4. Steve Says:

    Hey, Jack. I just saw this paper cited in a study comparing telomere length to red meat consumption: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4944490/?report=classic It’s getting passed around as a pro-meat study, even though its only conclusion is that more research should be done.

  5. Jack Norris RD Says:

    Steve,

    I don’t think that study should be considered enough to undo this research:

    http://jacknorrisrd.com/of-meat-and-mortality-part-4-harvard-studies-conclusion/

  6. Anne van Rossum Says:

    @Andreas Thaler

    Interesting what you are saying here: they matched each “vegetarian” with a meat-eater of the same BMI and habits.

    I’m vegetarian myself now for more than 10 years. It’s very to have a normal BMI. I tend to see that it’s more difficult to have this BMI for meat-eaters if I just look at my colleagues or friends. That means that if I will have to find a matching meat-eater this person might need to have very healthy habits to compensate for his/her meat eating behavior.

    So, apart from all the mistakes in the study, that they use matching might actually influence the outcome due to the fact that meat eating is actually BAD for you. 🙂

Leave a Reply

*