Evidence links low vitamin D levels with increased risk of death

Vitamin D, nutrient that we can get for free when the sun’s rays hit our skin, has been associated with a reduction in risk of a wide range of health issues including heart disease, cancer, diabetes and multiple sclerosis. In theory at least, having high levels of vitamin D in the body should protect against these conditions, and as a result may also reduce risk of death. In a recent study published in the journal Clinical Endocrinology, researchers in Austria assessed the relationship between vitamin D levels and overall risk of death in a group of 614 men and women studied for an average of 6.2 years each [1].

In this study, individuals were split into four groups according to vitamin D levels. The individuals in the group corresponding to the lowest levels of vitamin D, compared to those with higher levels, were found to be at more than twice the risk of death. However, in an ‘epidemiological’ study of this nature, it is impossible to tell if lower levels of vitamin D actually increase the risk of death, or is just associated with it.

To help here, it is possible to ‘control for’ what are known as ‘confounding factors’. If the association between two factors (in this case, vitamin D levels and risk of death) remain after confounding factors are accounted for, then this makes a ‘causal’ relationship between two factors more likely.

In this particular study, the researchers went on to control for various factors including age, sex, presence of diabetes and high blood pressure, smoking habits and waist-to-hip ratio. Even after accounting for these factors, there was as still a statistically significant relationship between vitamin D levels and risk of death. And again, risk of death was roughly doubled in individuals with the lowest levels of vitamin D in their bodies.

This evidence supports the idea that higher vitamin D levels in the body can reduce the risk of death, but does not prove it. For that, we require intervention studies. Not so long ago, a study was published which grouped together 18 studies in which individuals had been treated with vitamin D or placebo over a period of time [2]. Despite relatively low levels of vitamin D supplementation, those taking this nutrient saw a small but statistically significant reduction in overall risk of death.

The current state of the evidence suggests that vitamin D has genuine life-saving properties. For many of us, the weather at this time of year can cause vitamin D levels to plummet. For information on how to maintain vitamin D levels through supplementation see here.


1. Pilz S, et al. Vitamin D and mortality in older men and women. Clin Endocrinol (Oxf). 2009 Feb 18. [Epub ahead of print

2. Autier P, et al. Vitamin D supplementation and total mortality: a meta-analysis of randomized controlled trials. Arch Intern Med. 2007 Sep 10;167(16):1730-7.

15 Responses to Evidence links low vitamin D levels with increased risk of death

  1. Dougal McGuire 26 February 2009 at 2:56 am #

    So, if I consume lots of foods that contain Vit D my risk of dying is lower? I estimate my risk of dying (at current Vit D intakes) to be roughly 100% – how much should I consume to have a better chance of evading death?

  2. Ted Hutchinson 26 February 2009 at 8:44 pm #

    It isn’t however just sun exposure, upper atmospheric pollution and the wearing of sunscreens that is responsible for lower vitamin d status.
    Association of Leptin, 25-Hydroxyvitamin D,
    and Parathyroid Hormone in Women
    explains how leptin levels can reduce our ability to hydroxylate vitamin d3 into the active form and
    Reduced plasma half-life of radio-labelled 25-hydroxyvitamin D3 in subjects receiving a high-fibre diet.
    explains how consuming a high-fibre diet may also reduce the availability of Vitamin D3.

    It worries me that given we know that “Fructose-induced leptin resistance exacerbates weight gain in response to subsequent high-fat feeding”
    the amount of High Fructose Corn Syrup being consumed is not only contributing to obesity by increasing calorie consumption but also it may also, by increasing leptin resistance, be reducing 25(OH)D status and our ability to hydroxylate calcidiol into calcitriol the active metabolite.

  3. Paul Anderson 27 February 2009 at 5:07 am #

    The evidence seems to be growing daily that vitamin D supplementation is worthwhile, particularly for those of us living in Northern latitudes.

    Can anyone recommend a good source of affordable vitamin D in the UK or is it necessary to import from the US.

    On a related note, just how easy is it to get your vitamin D level tested in the UK. When I last mentioned it to my GP he showed no interest whatsover.

    Increasingly I am viewing the NHS as a very poorly run, second rate service. Its next to impossible to get testing done for things like insulin, C peptide, vitamin status, etc. My mother has been on blood pressure medication for something 20 plus years, and yet her doctor generally doesn’t even bother to check her blood pressure levels, let alone check for things like potassium levels and kidney damage. Of course, he is concerned about her cholesterol levels. This of course would have nothing to do with the incentives paid to GP’s for prescribing statins to “at risk” patients.

    Sorry for wandering off topic.

    PS – the next available appointment for my GP is 24 March. Although if I could make an evening appointment they can bring it forward to 23 March. Good news indeed. Fortunately its not an emergancy.

    Paul Anderson

  4. Dr John Briffa 27 February 2009 at 2:03 pm #


    The overall risk of death referred to in the blog post above relates to risk of death over a defined period of time (i.e. the length of time that individuals in the studies were monitored for).

    “I estimate my risk of dying (at current Vit D intakes) to be roughly 100%” If you remove any time constraint, then I would contend that your risk of dying is not “roughly 100%”, but “100%”.

  5. Liz Smith 27 February 2009 at 4:19 pm #

    When I went to my doctor complaining of feeling like death warmed up he offered to take a blood test – “if it made me feel any better!” The test came back negative, regardless of me having every symptom in the book which suggested I had a problem. He told me because of the tests there was nothing wrong with me and no further action would be taken. It took two tests privately (£250) to prove how very below par I was and two years slowly to get back to feeling normal using an alternative practitioner. In USA people are told to go to their doctor and get a blood test for Vit D, of course they have to pay for all tests, so what’s the chance I’ll get one on NHS as I am very interested in this theory of Vit D. I always thought that Vit D and Vit A should be taken together to work efficiently is that still true?

  6. Ted Hutchinson 27 February 2009 at 5:40 pm #

    GRASSROOTSHEALTH.ORG D Action currently offer $30 postal 25(OH)D tests. Simple spring loaded lancet pricks your finger and you drop 2 drops onto the sample tissue, Post back and they email you a result in 2~3weeks Results in ng/mL but X 2.5 to convert to nmol/l

    There is also a chart online
    “Disease Incidence Prevention by Serum 25(OH)D Level”
    so you can see how much extra D3 you need to reduce you risk of these conditions. Takes roughly 1000iu/d extra to raise 10ng or 25nmol/l.

    This chart
    shows UK average monthly status.
    To get from current to 60~70ng/mL 150~175nmol/l requires roughly 5000iu/d extra female and 6000iu/d extra men.

    For children and overweight work on 100iu for 3lbs and 1000iu/d for 30lbs.
    Anything under 10,000iu/d is safe.
    I use Now foods 5000iu/d from IHERB code WAB666 saves you $5. that also allows you to order 3 packs 360capsules, and using the code sneak under the £18 threshold for customs duty the p&p charge $6 doesn’t count for tax purposes. The problem is not so much the customs tax but the extortinate £8 handling charge out PO levy to collect it.

    I really came here to point you to 2 Grassrootshealth.org videos that are now on YOUTUBE
    They are
    Vitamin D and Diabetes-Can We Prevent it?
    Vitamin D and Cardiovascular Disease Prevention

    If you search youtube with the words
    uctelevision vitamin d
    You will find the full series of videos that talk about Vitamin D deficiency, sunscreen and skin cancers, and both Cancer incidence and the role of D3 in cancer treatment.
    It’s all cutting edge science so you may want to download rather than watch these online. The information load is fairly heavy and if you are not familiar with this stuff it takes time to grasp the importance of what they are saying.

  7. Ted Hutchinson 27 February 2009 at 5:55 pm #

    Sorry I forgot Paul’s point
    “Can anyone recommend a good source of affordable vitamin D in the UK or is it necessary to import from the US.”
    I prefer to import, bear in mind the points in my previous post about the importance of using a currency converter to ensure that after discounts have been applied your order value, before p&p is less than £18.

    Those IHERB work out at 5p daily if you do as I say. They are oilbased gel caps that Dr Davis Heartscanblog regards as the most effective form.

    However from the UK TheVitaminService offer
    Bio D Mulsion Forte™ (Vitamin D3) – 1fl.oz (30ml) – BioTics Research these are 2000iu drops that you can add to food/drink as you like.
    using 5 drops 10,000 over 2 days averages 5000iu/d

    With vitamin d supplementation you need to be aware that D3 is fat soluble and has a half life of 21days about. So providing you average over that period a daily intake of 5000~6000iu/d it doesn’t matter on which particular day you take that total amount.

    Only pregnant and nursing mothers NEED to take D3 DAILY to ensure the fetus/baby gets a regular even supply.

    So for everyone else if you forget you daily D3 then double up the next day. If you are useless at remembering to take them then dedicate on day each as D DAY and take all 7 days worth that particular day each week.

    Remember that D3 requires your body to perform two hydroxylations before it becomes an active substance. On it’s own D3 is inert.

  8. Dougal McGuire 28 February 2009 at 1:37 am #


    “I estimate my risk of dying (at current Vit D intakes) to be roughly 100%” If you remove any time constraint, then I would contend that your risk of dying is not “roughly 100%”, but “100%”.

    Maybe…but I remain optimistic…I apologize, I was being a little facetious. To atone for my bad behavior I will (attempt to) engage with you more seriously.

    I must confess I came across this blog by accident. What caught my eye about this article was the headline which (I feel) extrapolates form the data too far. the headline makes a general claim that I don’t think is supported by the data. The study was performed in a group of elderly people…I am in my thirties, does the data from this study support advice for me to increase my vitamin D intake or the population in general? I would contend not.

    I also feel this particular study is methodologically weak with a small sample size, too few mortality cases (which has resulted in wide confidence intervals) and woefully inadequate assessment of vitamin D status. It was conducted exclusively in an Austrian town which is a major limitiation (the data is not really generalizable). While the authors controlled for known confounders, studies like this are still likely to be heavily confounded. What was the study groups previous vitamin D intakes/exposure to sun? Is Vitamin D a predictor of mortality or is mortality a predictor of poor Vit D status? That is, does underlying disease have an effect on vit D metabolism, food choices or exposure to sun that reduces vitamin D concentration? (reverse causality).

    The meta-analysis only suggests a minor effect of Vitamin D on mortality risk (statistically significant…a different thing to practical significance as you are aware). Whether this reduction is ‘significant’ is something that could be argued long and hard about (personally I think that seemingly small reductions are significant in nutritional epidemiology). However, I am not sure there is sufficent evidence at this stage to make Vit D recommendations with sufficient confidence.

    Conducting research on Vitamin D is fraught with difficulty…as is any diet related research. My humble opinion, for what it is worth, is that due to the problems inherent in nutrition research it is wise to be very conservative when providing dietary advice. However, it is fair to point out that I am a skeptic when it comes to diet and health (not a denier mind).

  9. J R HODGSON JONES 28 February 2009 at 5:58 am #

    Thank you for an informative aricle. I am wondering what D3 has to be taken with to make it available? Not being a scientist, I do not understand ‘hydroxlations’ means.

  10. Dr John Briffa 28 February 2009 at 4:19 pm #


    “headline which (I feel) extrapolates form the data too far.”

    I didn’t extrapolate, I just summarised the findings of the study. Also, I think it’s worthwhile taking this individual study (whatever its deficiencies) in the context of the other available evidence in the area, including the meta-analysis of controlled intervention studies I specifically cited.

  11. Paul Anderson 28 February 2009 at 8:57 pm #

    I think the evidence supporting increased intake of vitamin D as well as the logic: less exposure to sunlight, less dietary intake, increased obesity reducing release of this fat soluable hormone//vitamin and the fact that there is no evidence whatsover of adverse side effects at the dosage’s suggested is pretty strong.

    ….. somewhat stronger than the case made for taking statins, or following the almost universally advocated low fat high carb diet.

    Humans require vitamin D and cholesterol in order to live. You can functioin perfectly well without taking a statin or consuming grains, high fructose corn syrup, or sugar.

    By all means challenge the validity of any study. To me it makes more sense to take vitamin D3 than it does to take the patented version, D2. Better still, get some sunlight and spend some time outside in the sun.

    Keep up the good work, Dr B. You continue to be one of the best bloggers on health/diet related matetrs on this side of the Atlantic.

    Paul Anderson.

  12. Hilda Glickman 1 March 2009 at 1:22 am #

    I understand that fair skinned people make vitamin D more easily from the sun than dark skinned people, even dark Caucasians. That would make sense as genetically fairer people are from cooler climates with less sun, so this would need to happen. Conversely if people were genetically adapted to hot weather they may make too much vitamin D and vitamin D can be toxic. So can we say that fairer people NEED less sun than darker. Also people who migtrate from hotter climates where they have been for generations, should try to get as much sun as possible here (a tough job).

    Hilda Glickman

  13. Paul Anderson 1 March 2009 at 2:30 am #

    PS – Ted, thanks for providing the advice on where to obtain vitamin D. Paul.

  14. Dougal McGuire 1 March 2009 at 3:29 am #


    In the interests of debate…a fair summary would then be that vitamin D status is related to mortality in older people. Without that qualification the headline gives the impression that this study provides data indicating low vitamin d status is linked to mortality in the general population…this would be an extrapolation from the dataset. Maybe I am being too pedantic but I don’t think pedantary is unwarranted.

    I don’t want to give the impression that I am dismissing the Pilz et al. study entirely. While limited it certainly adds to the literature. The problem is, as I see it, that it is very difficult to do randomized controlled studies relating to nutrients/foods and health. Therefore, there is going to be a reliance on epidemiological/mechanistic and quasi-RCT data to determine the links between diet and health. This reduces the confidence we have in any nutrition advice/policy and this should be recognized. We should also be careful about not over-hyping results from single studies or limited datasets (this is not an accusation mind, just we should be very precise in describing what a study actually supports).

    Now, I think there is likely to be a link between Vit D and disease risk – I have colleagues working on this issue. However, I don’t think the link is beyond reasonable doubt. Plenty of people dismiss nutrition advice/research either because of ideological reasons or because advice appears to change so regularly that they cease to believe any of it. My belief is that it is in everyones best interest to err on the side of caution and be pretty conservative in promoting the link between diet and health or disease.

    I am not looking for an argument for the sake of it…more aiming for honest debate I guess.

  15. bad NHS ON THIS SUBJECT 21 April 2009 at 3:12 pm #

    I too have been misdiagnoised by the NHS. I mentioned about vitamin D and got told no one is low in it. My muscles are weak and ive lost my job over it. I even got a lawyer but they are very busy dealing with complaints about the NHS. A year ago i started to take vitamin D and couldnt get any doctor to test me i searched and found at home blood spot testing for vitamin D but its from America. I contacted them and yes they send them to the UK. I got it through the vitamin D council ( do a search you will find them) My first test came back low and my second test has just been done it came back lower so im uppng the vitamin D. It takes 3 weeks from me sending my blood spot back to them for me to get the results here in the uk. Why is there nothing like this in the UK? So much for a good Health Service in the UK. They help you if you smoke, take drugs or alchole but if your low in vitamin D they dont test for it.

Leave a Reply to bad NHS ON THIS SUBJECT Click here to cancel reply.