Vitamin pill anyone?

According to a recently published study (download full paper here) on the prestigious Archives of Internal Medicine (a scientific peer reviewed journal), “in older women, several commonly used dietary vitamin and mineral supplements may be associated with increased total mortality risk”.

So, what exactly did the authors find? They found that in older women (mean age 61.6 years at baseline in 1986) participating in the Iowa Women’s Health Study the absolute risk of dying was increased over the 22 year study by 2.4% for those taking multivitamins, 4.1% for vitamin B6, 5.9% for folic acid, 18% for copper, 3.9% for iron, 3.6% for magnesium and 3.0% for zinc all when taken as a supplement. There was, however, a 3.8% reduction in the absolute risk of mortality when taking calcium supplements. Also, of interest, the risk of dying increased with increased supplemental iron, so the more you took, the higher the risk of dying. With calcium, its suggested protective effect was lost at higher doses (the authors found a reduction of the risk of dying with supplemental doses of up to 1300 mg/day with the protective effect disappearing with daily doses higher than 1300 mg/day). This was also reported on the news here and here and here but also on science related news sites such as here here and here.

This study has in its favour the size of the sample studied (38 772 women), the longitudinal repeated measures design with use of dietary supplements queried 3 times – at baseline in 1986, 1997, and 2004 (this reduced the risk of incorrectly describing use of supplements) and an average follow-up time of 19 years. In addition, the authors also adjusted their analysis for age; educational level; place of residence; diabetes mellitus; high blood pressure; body mass index; waist to hip ratio; hormone replacement therapy; physical activity; smoking status; and intake of energy, alcohol, saturated fatty acids, whole grain products, fruits, and vegetables. I will explain what this means later on.

The authors also highlighted, quite rightly, some limitations:

1) Being diagnosed with e.g. cancer or cardiovascular disease during follow-up may increase use of supplements (these people then die earlier because of the condition and not because of the use of supplements). What did the authors say about this? They did not find evidence, on this cohort, that having CVD, diabetes of cancer prior to the baseline measure increased use of supplements later on. In addition, for those diagnosed with cancer during the study, no evidence of increased use of supplements was found. As mentioned above, the authors adjusted their analysis for diabetes mellitus; high blood pressure; body mass index; waist to hip ratio; physical activity; and smoking status (amongst other variables).

Why? To exclude the impact of these factors on the estimate of risk of mortality. In other words, after some complicated statistical analysis (I will not bore you with that…), they wanted to be able to say that the differences found, if any, for the mortality risk between the group taking the supplements versus the group not taking the supplements were not due to one of the groups having more cases of diabetes, hypertension, obesity, or having more people physically inactive or smoking.

You want to limit the impact of a number of different variables that may explain the differences in mortality rate across the two groups so that any differences found may possibly be explained by the variable you are actually interested in, i.e., use of supplements vs. non-use of supplements. So, all these extra factors act as potential confounders. By adjusting for these potential confounders that are known to affect survival, one may actually increase the precision with which one can estimate the effect of giving supplements.

So, going back to the results highlighted earlier in this piece, the authors found that the absolute risk of dying was increased by 2.4% for use of multivitamins after adjustment for the other explanatory variables in the model. So, by removing the impact of the other variables known to impact on risk of dying for which data were available, one was still left with a statistically significant increase risk of dying of 2.4%. May seem very small but it’s there.

2) Those taking supplements may be the least likely to need them. Dietary supplement users tend to have underlying diets (and lifestyles) that are healthier. Those taking supplements may be the least likely to need them, the so-called “worried well”. This healthier lifestyle should, in theory, result in a reduction of risk of chronic disease and premature death. So, in the same sense that you have to adjust for diabetes or hypertension, you also have to adjust for these healthier lifestyle variables.

3) The nature of the statistical analysis performed requires that the p value be dropped for every variable you adjust for. As there were 15 variables, the p value dropped to 0.05 / 15 = 0.003. What does this mean? This means that the authors could not just be satisfied with being 95% sure that the difference found in regards to risk of mortality was a true difference and not just something that happened by chance. They had to be 99.7% sure that the difference did not occur by chance. However, they did not achieve this for all the variables tested – this level of certainty was only achieved for multivitamins, copper, and calcium.

The authors concluded by saying that “in this large prospective cohort of older women, we found that most dietary supplements were unrelated to total mortality rate. However, several commonly used dietary vitamin and mineral supplements were associated with increased total mortality rate, most strongly supplemental iron; calcium showed some evidence of lower risk.” A very important message from this paper, also highlighted by the authors, was that association does not equal causation. The authors never said that supplement use was the cause of increased mortality. What they said was that use of some supplements was associated with increased mortality, meaning both variables were present at the same time but you cannot establish whether one led to the other. This is an obvious conclusion as this study is an observational study.

This study raised concerns about the long-term safety of various vitamin and mineral supplements. Whilst there is undoubtedly a lot of supporting evidence for the use of supplements to address nutritional deficiencies, the case for using them with a goal of achieving health benefits beyond those achieved by a healthy diet is not established (what some people refer to as optimum nutrition – even if the concept of optimum nutrition cannot be defined biologically – how do you know you have achieved it if there is no definition, criteria or test for optimum nutrition?). We do not know what the implications are of the long term intake of supplements that exceeds the upper recommended levels of intake. Do they bring any benefit? Most importantly, do they do no harm?

A significant number of people take supplements in supra-physiological doses, far in excess of the levels found on the Dietary Reference Values book or the guidance set by the Expert Group on Vitamins and Minerals (Food Standards Agency, 2003) in its publication entitled “Safe Upper Levels for Vitamins and Minerals”. In these instances they act as drugs. For a drug to come on the market, they need to go through a rigorous and very long process of safety testing before it is allowed to be put on the market. However, “Dietary supplements, unlike drugs, do not require rigorous RCT testing, and observational studies are often the best-available method for assessing the safety of long-term use”. So why do we take and accept them so light heartily?

Take away messages summarised from here
1) “More is not always better”
2) “The truth that we need to change our eating and exercise habits is harder to swallow than a supplement pill”
3) “Supplements can help some people, harm others, and have no effect on most”
4) New nutrients aimed at your money, not your health à there are no magical nutrients

PS. In 2009, sales of dietary supplements in the UK totalled £670 million pounds. At the top of the list you could then find multivitamins and fish oils (close to £140 million each).

PS2. You are all doing a Bachelor of Science. This means that you will become scientists by the end of your degree. Remember, science should always be objective and unbiased. Remember that when you read comments posted on newspapers websites or if you come across some potentially “biased” websites. Not many topics generate such fervid debate as the use of supplements and what these can theoretically do. Dismissing a scientific paper simply because you do not agree with it or because you don’t like its findings (supplements may do harm) and then accepting another paper because it supports your views (supplements may do good), not reading a paper and then making wide assumptions about it, reading a paper with the intent of finding something to dismiss it….none of this is science. So, don’t do it!

This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

To prove you're a person (not a spam script), type the security word shown in the picture. Click on the picture to hear an audio file of the word.
Click to hear an audio file of the anti-spam word