That Vitamin D study.

So this vitamin D study says elderly ladies who do vitamin D get more fractures ...

Here's the study: Annual High-Dose Oral Vitamin D and Falls and Fractures in Older Women (full text, no less!)
Kerrie M. Sanders, PhD; Amanda L. Stuart, BappSc; Elizabeth J. Williamson, MA, PhD; Julie A. Simpson, PhD; Mark A. Kotowicz, MBBS, FRACP; Doris Young, MD, MBBS, FRACGP; Geoffrey C. Nicholson, PhD, FRACP.
JAMA. 2010;303(18):1815-1822.

A nice analysis of it: Falls, Fractures Linked to High-Dose Vitamin D. ("Action Points: Explain to interested patients that this study suggests that a single yearly high dose of vitamin D may not be beneficial for older women in preventing falls and fractures.")

Here's the discussion on facebook:

Karen Vaughan: Paul, what do you think of this Vitamin D study. Very high dose, but I am at a loss to explain why the data is so bad.

Paul Bergner: A few things come up. First, the editorial states that the women who took the D became more active and this may have led to more opportunities for injury. Some methodological problems, also hinted at -- there was no assessment of baseline parameters, and the two groups might not have been balanced in their tendency to injury. What if, for instance, the active group included women with a higher degree of osteopenia? Finally, methodologically, the results are just barely clinically significant, a confidence level extending down to 1.0 is not very strong, and the .047 P score is just barely under the conventional .05 cutoff. Then, the physiological norm for D from the sun is 10,000 to 50,000 per day. The dose given was ten times this highest measure from sunlight exposure. Safer probably to mimic nature.

Darcey Blue: Is it possible that it indicates one single high dose isn't effective for what vit D does? like we need it on a semi regular basis ... not once a year?

Sharon Hodges-rust I would also say that vitamin D does not stand alone- the other nutrients need to be there too- how many were on blood thinners/low vitamin K diets? what about their calcium/magnesium and vitamin A intake was. I also agree one big dose does not replace days/weeks of exposure to get the equivalent from the sun.

Jim Mcdonald: "The rationale for the single dose -- 500,000 International Units given orally once a year -- was that it would be clinically easy, would prevent decreases in vitamin D levels in the winter months, and would eliminate the problem of nonadherence, the researchers said."
they neglected to mention (or realize) that it's totally at odds with how one would naturally receive vitamin d in reality.
"studies", again and again, provide a lot of anecdotal evidence that advanced education and higher learning are not one and the same.

Karen Vaughan: It is also a way of making a rather cheap supplement expensive in the guise of making it easy to take.
The mechanism is what interested me, and aside from possibly feeling better and doing more dangerous things, there isn't much in the way of information. For instance, the sun gives us potentially valuable breakdown products in addition to the D. Or perhaps something in calcium assimilation is affected which weakens bones and causes falls when you give D without minerals enough to affect the one big shot..

Penelope Toothman: There is a sweet spot, which is age dependent in maximizing activity ( the curve is not linear but hooks down) and the blast of 500,000 iu missed the mark. More and more is not always more and more better. Had they given them 2 or 3000 units per day they probably would have seen a different response and been on a different part of the dose response curve.

Paul Bergner: The practice of giving large doses once or twice a year is common in Scandinavia, and is also used for babies with Ricketts, and has not been known to cause adverse effects in adults or children.

Penelope Toothman: I remember that the dose response curve was linear up to very high levels on 25(OH)D in people younger than 40 and was hooked down for older people. Why it is hooked down I have no idea or what other variables there were wrt diet, gender, activity level, comorbidity, whatever.

I'd love to see another study, with 5000-10000 IU / day, and the same amount of elderly people taking it, or taking placebo, for a year or three. That'd likely give a different result.