Taking Vitamins May Lead To Earlier Death In Older Women

The study:

Dietary Supplements and Mortality Rate in Older Women, Archives of Internal Medicine, October 2011

They looked at intake of 15 supplements:

• Multivitamins (Absolute Risk Increase, ARI, 2.4%)
  
• Vitamin A
  
• Beta-carotene
  
• Vitamin B6 (ARI 4.1%)
  
• Folic acid (ARI 5.9%)
  
• B complex
  
• Vitamin C
  
• Vitamin D
  
• Vitamin E
  
• Iron (ARI 3.9%)
  
• Calcium (Absolute Risk Reduction 3.8%)
  
• Copper (ARI 18%)
  
• Magnesium (ARI 3.6%)
  
• Selenium
  
• Zinc (ARI 3.0%)

The 8 with numbers after them were associated with either an increased risk of mortality (all but calcium) or a decreased risk (only calcium). The others ... this study found no significant impact.

I think the risk increases are small. What do you think? Just 2.4% for taking a multivitamin? Copper stands out though, doesn't it.

There were 5 dose options on the questionnaire for selected vitamins. The benefit for calcium was lost at its highest dose. Also for calcium, "no clear dose-response relationship was observed." "For vitamins A, C, D, and E, as well as minerals selenium and zinc, no dose-response association was found." This was surprising to me. It looks like taking the vitamin at all, regardless of dose, was risky.

This was an epidemiological study, an analysis of data from participants in the large Iowa Women's Health Study. It's not appropriate to draw advice from this type of study, especially one with such weak associations. It's not a randomized controlled trial where they can specify doses, combinations, and types. You can however formulate or refine hypotheses. And the hypothesis that supplements fail to prevent chronic disease or even do harm is gaining momentum.

I would like to draw advice from it because, as you know, I'm biased. I think popping vitamins willy nilly is not a good idea. They often contain a pharmaceutical dose of an isolated chemical, along with fillers, binders, and other residuals like plastics and heavy metals that the food which contained the nutrient may not. For people who buy organic and local, who support Country of Origin labeling, why not apply this standard to supplements? I read that nearly all the vitamin C in the world, as well as glycerin used as filler, comes from China. Not that there's anything wrong with that.

Assuming there really is an increased risk of death from consuming these vitamins, that risk, it must be said, doesn't apply to anyone but older (mean age: 61.1), white (99.2%) women who were postmenopausal (98.6%) who lived in the state of Iowa. I used to think, "Oh, sure it does." But you just can't say that since iron and folic acid supplements are harmful for a postmenopausal woman, they're equally so for a woman of childbearing age. And while multivitamins might not be good for older, white women in Iowa, they might be good for poor, urban teens who exist on cakes, chips, and soda, or for anyone with deficiency. People have different needs and tolerances.

The study's strengths:

• It was big (over 38,000 women).
  
• It collected data on supplement use three times over the time-course, not once at baseline which is what I normally see. As they said, "The use of repeated measures enabled evaluation of the consistency of the findings and decreased the risk that the exposure was misclassified."
  
• It adjusted for a lot of possible confounding factors, including:
  
  Age
  Energy intake
  Education level
  Place of residence
  Diabetes
  High blood pressure
  Body mass index
  Waist-to-hip ratio
  Hormone replacement therapy
  Physical activity
  Smoking
  Dietary factors: saturated fat, fruits, vegetables, whole grains, and alcohol (there was a food questionnaire twice during the time-course)

In fact, the assumptions people make about people who take supplements ... that they're usually more educated and take better care of themselves, were applicable in this case. Supplement users in this study had a lower prevalence of diabetes, high blood pressure, and smoking; had a lower BMI and waist-to-hip ratio; had a higher education level; were more active; ate fewer calories; ate more fruits and vegetables, etc. The analysis prior to adjustment for these factors showed taking supplements resulted in a lower risk of death compared to not taking them. Only after adjustment did the increased risk become apparent.

I think vitamins are helpful sometimes. For the well-fed, they can be a dross on the system.

Related: Comparison of this study's design to the Is Elevated Cholesterol Beneficial? study's design. This one is so much better.
________