says VC increases risk for stones, but b6 decreases in men
Had to review my Ridge Regression Multivariate Analysis...
First, in this study's favor is that it comes from Harvard. I mean that.
A perfect 1 to 1 linear relationship between two variables (e.g. kidney stones and vitamin C over 1 g) would be 1.0. The result was 1.41, and it isn't clear to me in what direction (from the abstract) the line varies from linear.
But lets give them the benefit of a doubt and assume that from 45,619 men, and several questionaires, they were able to input this data without significant error. Unless this were a special population, it would be reasonable to assume that less than 5%, and probably less than 1%, of the men consumed more than 1000 mg of vitamin C. (Perhaps this number can be found in the actual study paper).
1 percent of 45,619 is 456. This is probably in the neighborhood of the population in their study who consume more than 1000 mg daily. It might even be smaller. (It is interesting to note how they mention "477,700 person-years of follow-up" - I guess to make their results appear to carry more weight?) But what if the number of C takers was only 10? Found the full text and the number of subjects among the 45,619 who consumed more than 1000 mg vitamin C (mean about 1100mg) was about 190.
They documented 1473 kidney stones "events" in their population prone to stones.
There analysis tries to connect various dietary variables in the larger population to the 1473 events, 190 of which consumed about 1 g of vitamin C. If my understanding if RR M.v.a. is correct, then 1.0 is the perfect 1 to 1 correlation, and if that understanding is correct, then the lower amounts of vitamin C (e.g. < 90 mg is 1.0) have a better correlation between vitamin C intake and kidney stones.
Here is a direct quote from the full-text
After adjusting for age, there was no association between vitamin C intake and the risk of incident stone formation.
They claim that the risk appears after their adjusting the multivariats, and in their view, a higher number (e.g. 1.41 means increased risk).
The relation between vitamin C intake and stones emerged only after the inclusion of dietary potassium in the multivariate analysis; potassium intake was inversely associated with stone formation and was positively associated with dietary vitamin C
It seems to me that trying to draw a conclusion from a single dietary factor, e.g. vitamin C on such a small sample size (190) and only after admittedly fudging the analysis, is questionable at best. If their data is indeed correct, then it might be wise for people taking low vitamin C (around 1 g) to increase their potassium and magnesium.
As we have quoted Pauling elsewhere on the fact that 50% of the stones appear in alkaline urine, and in those cases, ascorbic acid should prevent those stones. In those with acidic urine, taking more of sodium ascorbate would help prevent the formation of stones.