People who take vitamin and mineral supplements are risking their lives. This was the message carried by hundreds of articles posted on websites and in newspapers and magazines on 10 and 11 October 2011. The writers of these alarming articles were faithfully summarizing the press release issued by researchers at the University of Eastern Finland and the University of Minnesota, but apparently none of them had time to read the actual study. So we've done that for them.
The researchers published an epidemiological study in the Archives of Internal Medicine based on data gathered for the Iowa Women's Health Study. This involved almost forty thousand women, who were monitored between 1986 and 2008. At the start of the study the women's average age was 62.
The researchers' publication is about the women's supplement use. When the study started 66 percent of the women took supplements. By 2004 the amount had risen to 85 percent.
When the researchers compared the mortality rate of the supplements takers with that of those who did not take supplements, they observed that supplements use had no noticeable effect. The mortality rate was slightly higher among the supplements takers, but the relationship was not significant. Epidemiology is not a hard science. Only when epidemiologists notice at least a doubling of chances, do they consider it a 'fact'. And that was certainly not the case in this study.
The figures with a d next to them are the ones that the researchers say are statistically significant. Calcium supplements supposedly reduce the likelihood of dying; multi-vitamin and mineral supplements apparently raise the likelihood of dying by a mere six percent, and copper supplements [does anyone still take them?] would be responsible for a 45 percent increase in mortality. But a doubling in the mortality rate? That's nowhere to be found.
Well it is actually. When the researchers broke down their data, they noticed that the mortality rate in some people who took iron supplements doubled: among users of unusually high amounts of iron.
A multi-vitamin tablet contains 10-15 mg iron. These doses don't affect mortality rates, but women who take hundreds of milligrams of iron every day are likely to experience a rise in their mortality rate. However, these are not doses found in normal supplements. These are doses found in medical preparations that doctors prescribe for serious anaemia, which is sometimes a complication after heavy operations.
So the study provided few shocking results. But the researchers believe that taking extra vitamins and minerals is more dangerous than their data shows, and that's why they sounded the alarm.
"Based on existing evidence, we see little justification for the general and widespread use of dietary supplements", the researchers conclude. "We recommend that they be used with strong medically based cause, such as symptomatic nutrient deficiency disease." The irony is that the only 'hard' relationship that they found between mortality and supplements use was a 'medically based' one.
The researchers acted pretty strangely. They discovered that extra minerals and vitamins don't lead to a rise in mortality – and not to a decline either – but don't draw this conclusion. They undermine their own research and say: we've really found nothing, but nevertheless we think taking supplements is dangerous. Indeed, not very scientific. But they get away with it.
And the media? They all just copied the contents of the press release without even checking it.
Arch Intern Med. 2011;171(18):1625-1633.