Study: Vegetarianism does not endanger bone mineral density

[Tapio]

Hi! There's been quite a lot of talk regarding diet here lately, especially vegetarian vs. omnivorous. Thought this should be brought up.

From The American Journal Of Clinical Nutrition:

Effect of vegetarian diets on bone mineral density: a Bayesian meta-analysis

Conclusion: The results suggest that vegetarian diets, particularly vegan diets, are associated with lower BMD, but the magnitude of the association is clinically insignificant.

Interesting. Comments?

 

[Jorghnassen]

Hi! There's been quite a lot of talk regarding diet here lately, especially vegetarian vs. omnivorous. Thought this should be brought up.

From The American Journal Of Clinical Nutrition:

Effect of vegetarian diets on bone mineral density: a Bayesian meta-analysis



Interesting. Comments?

As with all meta-analyses, it'd be nice to know a little more about the 9 actual studies they meta-analyzed (as well as more detail on how the analysis was conducted, i.e. models, priors, etc). Also, it provides an example of the difference between statistically significant (the BMDs were lower for vegetarian diets) and clinically significant (the differences are not enough to say it affects quality of life).

 

[JJM]

As I recall, the American Dietetic Association endorses vegetarian diets as long as they are properly balanced. I don't know if they gave much consideration to BMD.

 

[fls]

As with all meta-analyses, it'd be nice to know a little more about the 9 actual studies they meta-analyzed (as well as more detail on how the analysis was conducted, i.e. models, priors, etc).

Tip: click on the "Full text" link.

http://www.ajcn.org/cgi/content/full/90/4/943

Also, it provides an example of the difference between statistically significant (the BMDs were lower for vegetarian diets) and clinically significant (the differences are not enough to say it affects quality of life).

One cannot determine this from the measure that they used (a ratio of the mean BMD between the two groups). The risk of fracture depends upon one's distance from the standard. If both groups were below the standard, an additional decrease of 4% may represent a clinically significant increase in risk. You need additional information to determine whether or not this is the case.

ETA: However, I agree that it is doubtful that this represented a clinically significant difference.

Linda

 

[Jorghnassen]

Tip: click on the "Full text" link.

http://www.ajcn.org/cgi/content/full/90/4/943



One cannot determine this from the measure that they used (a ratio of the mean BMD between the two groups). The risk of fracture depends upon one's distance from the standard. If both groups were below the standard, an additional decrease of 4% may represent a clinically significant increase in risk. You need additional information to determine whether or not this is the case.

ETA: However, I agree that it is doubtful that this represented a clinically significant difference.

Linda

I was going by the summary: clearly the BMDs were lower for vegetarians given the confidence intervals (thus a statistically significant difference), but the stated conclusion said "clinically not significant" (due to small effect size); just to highlight that one type of significance does not necessarily imply the other holds true.

 

[IMST]

I'm totally going to lord the non-clinical significance of my statistically significantly less dense bones over all y'all.

 

[fls]

I was going by the summary: clearly the BMDs were lower for vegetarians given the confidence intervals (thus a statistically significant difference), but the stated conclusion said "clinically not significant" (due to small effect size); just to highlight that one type of significance does not necessarily imply the other holds true.

I realize that you got the idea from their conclusions. It's just that they didn't actually present data to support that particular conclusion - it's a sort of a guess that they made. It seems like a reasonable guess, though. Sometimes authors will cover that issue in the background information section - a discussion of what sort of difference would be clinically significant. The problem is that the authors made-up a measure to use for their meta-analysis (so that they could combine studies that used different measures), which then meant that it couldn't be related directly to things like fracture risk.

I'm just jumping in for information purposes only.

Linda

 

[Jorghnassen]

I realize that you got the idea from their conclusions. It's just that they didn't actually present data to support that particular conclusion - it's a sort of a guess that they made. It seems like a reasonable guess, though. Sometimes authors will cover that issue in the background information section - a discussion of what sort of difference would be clinically significant. The problem is that the authors made-up a measure to use for their meta-analysis (so that they could combine studies that used different measures), which then meant that it couldn't be related directly to things like fracture risk.

I'm just jumping in for information purposes only.

Linda

You are right, checking the full article, it's really more, there's a small difference in BMD, we don't know if it's clinically significant (in terms of fractures and other things), but here's an argument that it probably is not. I'm not too convinced (or quite understand) their made-up measure to ascertain clinical significance.