The Women’s Health Initiative Confirms That Vitamin A Intakes Are Only Associated With Osteoporosis At Low Vitamin D Intakes

Visit Us
Follow Me
The Winter 2005/Spring 2006 issue of Wise Traditions carried my article on vitamin A and osteoporosis, in which I argued that vitamin A only contributes to osteoporosis when vitamin D intakes are very low. I suggested that vitamin A intakes would not be correlated with osteoporosis risk in people consuming adequate vitamin D. This prediction has now been confirmed by a study of subjects participating in the Women's Health Initiative.

This is the third prospective study evaluating the association between vitamin A intakes and the risk of osteoporosis. Previously, the Nurses' Health Study found that high intakes of vitamin A were associated with an increased risk of osteoporosis. The Iowa Women's Health Study, by contrast, found that use of vitamin A supplements was associated with a 17 percent increase in risk, but found no association with total vitamin A intake or the amount of vitamin A in the supplements. Until now, no study had separated the subjects into high and low vitamin D intakes and looked for associations with vitamin A in the separate groups.

In the Women's Health Initiative, a modest ten percent increase in the risk of fracture was seen in women consuming more than 2,000 IU of preformed vitamin A (retinol) per day, but the risk did not continue increasing in retinol intakes higher than this. When the investigators made statistical adjustments for intakes of vitamin D and calcium, the association disappeared.

When the investigators divided the subjects into those consuming less than 440 IU of vitamin D per day and those consuming more than this amount, vitamin A intake was only associated with an increased risk of fracture in women with low intakes of vitamin D. Among women consuming less than 440 IU of vitamin D per day, the twenty percent with the highest intake of retinol, who averaged more than 8,000 IU per day, were 15 percent more likely to suffer a fracture over the course of the study than women with lower retinol intakes. Among women consuming more than 440 IU of vitamin D per day, however, the twenty percent with the highest intake of retinol were five percent less likely to suffer a fracture, although the difference was not statistically significant.

As most readers of this blog probably know, an intake of 440 IU of vitamin D is pitifully low. Ideal intakes are probably 2,000-4,000 IU in the absence of sunlight. Blood levels of 25(OH)D are a much better indicator of vitamin D status than dietary intake, since they account for both diet and sun exposure. Adequate blood levels appear to be 30-35 ng/mL. The ideal epidemiological study would separate women into those with 25(OH)D levels below this amount and those above this amount. It would be even better to add a third category of those with blood levels below 12 ng/mL, which indicates a more severe level of deficiency. We may find, perhaps, that high retinol intakes actually protect against osteoporosis when blood levels of vitamin D are ideal.

Visit Us
Follow Me

You may also like


  1. It appears that Vit D has so much more to offer than we can begin to imagine! Almost every day it seems there is more information from various studies about the benefits of good Vitamin D levels in one's system. Just goes to show that our natural instincts to be out in the sunshine, especially when requiring healing of some sort, are perhaps not to be ignored.

  2. Chris,
    What about people who are really high on Vitamin D? My level is 84.9 ng/mL. Should people like me and my boyfriend (his is 65 ng/mL) be supplementing with Vitamin D3?

  3. Health Womens Healthy Living Goals
    A thought-provoking blog with equally interesting discussion! Just like to say that the key to healthy nutrition is our body's ability to absorb the vitamins, minerals it needs. It doesn't matter how many supplements we take, if the body can't utilize/convert it properly, we will experience deficiency. Likewise, the body rely on certain vitamins/minerals etc, for metabolizing others. It cannot achieve this if it is not in a healthy state. A good starting point must be a healthy cleanse/detox to enable it to function optimally and draw all the goodness from the healthy foods we eat…and that include helping us to convert Vit D from the sun!

  4. A calcium-rich diet is only one part of an osteoporosis prevention or treatment program. Like exercise, getting enough calcium is a strategy that helps strengthen bones at any age. But these approaches may not be enough to stop bone loss caused by lifestyle, medications or menopause.


  5. Hi Chris:

    I was wondering in you have come across any information about vitamin A and Familial Hypercholesteroemia? Apparently individuals with FH have an enhanced ability to accumulate vitamin A and can experience toxicity at much lower levels than the general population. Here are a few links:

    I have one question. In the winter months, what would be a good ratio of vit D to vit A intake in IU? Thanks for this very informative blog.


  6. Stephan, these graphics about sunscreens are absurd nowadays because they dont use the best sunscreens avalaible today like Mexoryl XL at high concentration, Tinosorb M/bisoctrizole or Tinosorb S/bemotrizionol. I disagree with your last link. Everyone should avoid sun exposure because of photaging and instead of everyone must take vitamin D3 supplements. Only if you have psoriasis is really good sun exposure, always with care. As concerned by skincare I never take sun even living in a sunny country like Spain.

    Regards, excellent post.

  7. Hey Chris,
    You may wish to look up my post “Vitamin D3 or Disease, Disability and Death” in my Golden Trails blog ( While there, under the Side Bar Health Trail, I have also included some well researched links on this subject which will be helpful to you and your readers.

  8. Chris,

    If you go to the documentation page:

    You can download a PDF file by clicking the link “Dose-Response of Vitamin D and a Mechanism for Prevention of Cancer”

    The presentation “Skin Cancer/Sunscreen — the Dilemma” is also well worth reading. It includes absorption spectra for common sunscreen UV blockers. The amount of data indicting sunscreen is shocking.

  9. Ted, the first study concluded that the reaction was second-order and the second study concluded that the reaction is biphasic proceeding from first-order to zero-order. The first study concluded that the enzyme seems to reach saturation when 25(OH)D levels reach 40 ng/mL, the second study concluded this occurs at 35.2 ng/mL. According to Bruce Hollis (personal communication), the first study was preliminary and the second more accurate.

    I don’t know of any evidence that humans require vitamin D3 reserves in adipose tissue for any physiological function. The health endpoints are tied to nutritional status, which is measured by 25(OH)D levels.

    Of course, if one is living in a situation where vitamin D input would drop during a portion of the year, it would be good to have adpiose reserves if one is not supplementing during that time.


  10. Chris,

    There is pretty consistent observational evidence that cancer drops off a cliff above 50 ng/mL 25(OH)D3. There’s a good summary of the data at

    Whether this represents a true cause-effect relationship is open to debate, but vitamin D does have antiproliferative effects in cell culture and some animal cancer models.

  11. With respect the data presented in Figures 4 and 5 of 25-Hydroxylation of vitamin D3: relation to circulating vitamin D3 under various input conditions appear to me to confirm rather than contradict Hollis’s earlier work.
    Below 100nmol/l the body has little reserves of Vitamin D3 but above 125nmol/l 50ng we see larger stores.
    Neither of these papers appear to me to support your suggestion that 87.5 nmol/l 35ng provides adequate reserves of stored D3 for the majority of the population.

  12. Thanks for the good words Tom, and thanks for posting the paper Ted. However, that paper is old and preliminary and consequently innaccurate. See this paper instead: Tom, looking at the full text of this paper should answer your questions.

    Homertobias, see my article “From Seafood to Sunshine” on the WAPF site. There is one study showing three times the risk of heart disease with 89 ng/mL 25(OH)D or higher. Granted, it is a case-control study, but since there is no evidence whatsoever that it is beneficial to have 25(OH)D anywhere near this high, it is not worth the risk.

    25(OH)D should be at least 30-35 ng/mL. This provides substantial benefit over lower levels. There is evidence of very modest benefit to going up to 46 ng/mL for blood sugar, but it is tiny. There is no evidence for benefit at levels of 60 ng/mL that I know of.

  13. Tom: D3 has a half life of about 8 hours, D2 has a half life of about 8 days. Your skin stops converting uvb to vit d when your level is about 60. Paleolithic man had a surplus of unpolluted sunshine and had no need to store. You can supplement your level higher but it is not physiologic. A level of 90 does not immediately harm you but may not be good for you. When your level is > 60 d is stored in….fat.

  14. Slowly the light illuminates the dark, and Chris was there with a candle before most everyone else.

    Really enjoyed the article in the current Wise Traditions! It made me realize I need to be more careful about A/D balance.

    Do you know any research that determined how transient 25(OH)D levels are in the body? Although it’s fat soluble, my understanding is that it is not stored in the body to the extent that vitamin A is (cf. the A versus D content of land animal livers), thus constant input may be more important.

Leave a Reply

Your email address will not be published.