The Autier et al review in which you note a SIGNIFICANT REDUCTION IN ALL-CAUSE MORTALITY
actually reports that in none of the studies they looked at where all-cause mortality was an endpoint was the observed reduction statistically significant
(of course I am aware that “is it statistically significant” is usually far from the most important question, but I think you shouldn’t say “significant” in this context unless someone actually found statistical significance)
observes that these observed apparent reductions in all-cause mortality seem to be larger in studies that used smaller doses of vitamin D (10-20µg/day versus higher doses)
which I think makes it difficult to maintain simultaneously (1) that these results are accurate measurements of a real improvement and (2) that a good daily dose would be something on the order of 800µg/day
observes concerning these studies that “most trials included elderly women and a sizeable proportion of individuals were living in institutions” and claims more concretely that “gains in survival are mainly in elderly women living independently or in institutional care, who are likely to initially have a very low concentration of 25(OH)D with a substantial risk of falls and fractures” (but: it is not clear to me how carefully they have verified that latter claim)
So maybe it supports your thesis, but the authors of the review itself seem unconvinced and have a specific claimed explanation for these findings which, if correct, wouldn’t indicate any benefit to younger basically-healthy individuals at negligible risk of the fall-fracture-death sequence.
(I am not passing any judgement on the rest of what you say; it does seem curious to me that the medical establishment generally seems to prefer what seem like very small doses of vitamin D, and that apparently no one has thought it interesting to try giving 1000 people 2000 IU/day or so and seeing what happens to them.)
(I am not passing any judgement on the rest of what you say; it does seem curious to me that the medical establishment generally seems to prefer what seem like very small doses of vitamin D, and that apparently no one has thought it interesting to try giving 1000 people 2000 IU/day or so and seeing what happens to them.)
There’s the VITAL study where they gave 2000 IU/day with 12,927 assigned to Vitamin D.
Good catch. They looked at cancer, cardiovascular events (heart attack, stroke, death from cardiovascular causes), and deaths-from-cancer among men aged 50+ and women aged 55+. They found some “small but nonsignificant” improvements for the ones taking vitamin D (6.13% of vitamin D takers got cancer, versus 6.36% of people taking placebo vitamin D; 3.06% of vitamin D takers had cardiovascular events, versus 3.16% on placebo). They list a number of other things for which supplemental vitamin D didn’t help significantly but for those they don’t give the actual numbers (a cynic might conjecture that those numbers show small increases rather than small reductions).
On the other hand, they found a not-so-small reduction in cancer deaths among vitamin D takers (“a suggestive 17% reduction”; I guess the language indicates that this too was statistically insignificant because the numbers of deaths in the two cases were rather small, but I don’t know; if they exclude deaths in the first two years of follow-up, which apparently is a reasonable thing to do for slowly developing diseases like cancer, though that rationale feels to me like it makes more sense for measuring getting cancer rather than dying of it?, that becomes a 25% reduction.
They also found “a suggestive 23% reduction in cancer risk” for African Americans specifically.
Which is all interesting but there’s a bit of a whiff of data-mining here that makes me reluctant to get too excited.
More details here. The cynic mentioned above would be at least partly correct. For instance, cardiovascular deaths were 11% higher among people getting real vitamin D, as were incidences of a couple of specific kinds of cancer they tracked; all-cause mortality was pretty much identical between the two groups.
The Autier et al review in which you note a SIGNIFICANT REDUCTION IN ALL-CAUSE MORTALITY
actually reports that in none of the studies they looked at where all-cause mortality was an endpoint was the observed reduction statistically significant
(of course I am aware that “is it statistically significant” is usually far from the most important question, but I think you shouldn’t say “significant” in this context unless someone actually found statistical significance)
observes that these observed apparent reductions in all-cause mortality seem to be larger in studies that used smaller doses of vitamin D (10-20µg/day versus higher doses)
which I think makes it difficult to maintain simultaneously (1) that these results are accurate measurements of a real improvement and (2) that a good daily dose would be something on the order of 800µg/day
observes concerning these studies that “most trials included elderly women and a sizeable proportion of individuals were living in institutions” and claims more concretely that “gains in survival are mainly in elderly women living independently or in institutional care, who are likely to initially have a very low concentration of 25(OH)D with a substantial risk of falls and fractures” (but: it is not clear to me how carefully they have verified that latter claim)
So maybe it supports your thesis, but the authors of the review itself seem unconvinced and have a specific claimed explanation for these findings which, if correct, wouldn’t indicate any benefit to younger basically-healthy individuals at negligible risk of the fall-fracture-death sequence.
(I am not passing any judgement on the rest of what you say; it does seem curious to me that the medical establishment generally seems to prefer what seem like very small doses of vitamin D, and that apparently no one has thought it interesting to try giving 1000 people 2000 IU/day or so and seeing what happens to them.)
There’s the VITAL study where they gave 2000 IU/day with 12,927 assigned to Vitamin D.
Good catch. They looked at cancer, cardiovascular events (heart attack, stroke, death from cardiovascular causes), and deaths-from-cancer among men aged 50+ and women aged 55+. They found some “small but nonsignificant” improvements for the ones taking vitamin D (6.13% of vitamin D takers got cancer, versus 6.36% of people taking placebo vitamin D; 3.06% of vitamin D takers had cardiovascular events, versus 3.16% on placebo). They list a number of other things for which supplemental vitamin D didn’t help significantly but for those they don’t give the actual numbers (a cynic might conjecture that those numbers show small increases rather than small reductions).
On the other hand, they found a not-so-small reduction in cancer deaths among vitamin D takers (“a suggestive 17% reduction”; I guess the language indicates that this too was statistically insignificant because the numbers of deaths in the two cases were rather small, but I don’t know; if they exclude deaths in the first two years of follow-up, which apparently is a reasonable thing to do for slowly developing diseases like cancer, though that rationale feels to me like it makes more sense for measuring getting cancer rather than dying of it?, that becomes a 25% reduction.
They also found “a suggestive 23% reduction in cancer risk” for African Americans specifically.
Which is all interesting but there’s a bit of a whiff of data-mining here that makes me reluctant to get too excited.
More details here. The cynic mentioned above would be at least partly correct. For instance, cardiovascular deaths were 11% higher among people getting real vitamin D, as were incidences of a couple of specific kinds of cancer they tracked; all-cause mortality was pretty much identical between the two groups.