Two vitamin studies were reported this week, one from Iowa examining death rates in women over the age of 55 and vitamin use; the other was a randomized controlled trial assessing the effect of extra vitamin E and/or selenium on the development of prostate cancer. The headlines highlighted the hazards of some of these supplements, but here's my take on these studies:
1. A multivitamin is an insurance policy against an inadequate diet in people over 55 (those in this study; multivitamin and DHA use is much more important in women of childbearing age). These Iowa women consumed an average of 6.3 servings of fruits and veggies a day – more than twice the national average in the US and Canada, so maybe they didn't need the insurance policy.
2. We know that extra iron for 50-year-old (or older) women and men who do not have iron deficiency is harmful. Even so, the incremental hazard ratio was in the under-10% range – we typically look for hazard ratios that decrease risk by 40% or increase to 240% of baseline in epidemiologic (association) studies like the Iowa study before we put a lot of faith in them or act on them prior to randomized studies.
3. We know vitamin D is beneficial as most do not get enough from their diet.
4. We know that extra vitamin E (large amounts seen in this Iowa study and in the prostate study) are harmful from randomized studies – yet a benefit was shown in the Iowa study. Additionally, extra magnesium in the elderly has been shown to be beneficial, yet is indicated harmful in this Iowa study. That is the problem with epidemiologic studies – you use them to suggest questions that need answering (that is, the results suggest areas for further study), but they often do not answer a lot by themselves.
5. Quality of Life measures were not examined, such as decreases in memory degradation you'd expect with DHA (the most important omega-3 fat for the brain and eyes) or decreases in eyesight loss (macular degeneration) from DHA plus lutein, plus some parts of a multi, or even bone fractures.
My overall impression is the Iowa study changes nothing, and reinforces iron (and maybe copper) hazard. Avoid these two in multivitamins if not of childbearing age. What this leaves us with is more puzzles and questions for future studies to solve.
I'm still taking (because data supports such):
1. A multivitamin without iron and with only a low dose of vitamin A (less than 3500 IU a day) and low amounts of vitamin E (as mixed tocopherols, about 30 IU). I split my multivitamin in half, and take one half in the morning and one half at night.
2. 1200 IU of vitamin D3 a day
3. DHA (900 mg) and lutein (20 mg)
4. A probiotic
5. A little extra calcium (600 mg) (with magnesium 300 mg)
6. Two baby aspirins with plenty of water.