Vitamin pills and risk of death studied

“Women have been told there is ‘little justification’ for taking vitamin supplements in an academic study that found the tablets may be actually linked to a higher risk of death,” The Daily Telegraph reported.

This study looked at deaths and causes of death in women with an average age of 61 years, and their use of vitamin supplements. It found that women who used certain supplements had a greater risk of dying earlier than women who did not use those supplements.

The study has some strengths, including its large size and its repeated assessments during the follow-up period. However, it also has some limitations. For instance, the results were analysed using several different models, each designed to take into account the various confounding factors (such as age and smoking status) that could have affected the results. These analyses had conflicting results, with the most complex finding that the only supplements significantly associated with dying earlier were multivitamins and copper.

Overall, the study cannot show that vitamin supplements cause early death. It is possible that women were taking supplements in response to an illness that could have caused their earlier death. In particular, many women take iron for anaemia, which is associated with chronic disease, injury and major surgery – factors that could have affected the results.

Most people can get all the vitamins and minerals they need by eating a balanced, varied diet. High doses of vitamins and minerals may have side effects, and can also interact with certain medications. Some groups who are at risk of deficiencies are advised to take supplements, details of which can be found within the Common Health Questions section. Those who have been advised by their doctor to take supplements should continue to do so.

Where did the story come from?

The study was carried out by researchers from the University of Eastern Finland; the University of Minnesota, US; Yeungnam University, Republic of Korea; and the University of Oslo, Norway. It was funded by various sources including the National Cancer Institute and the Academy of Finland.

The study was published in the peer-reviewed journal Archives of Internal Medicine.

The study was generally reported fairly by the papers, with most reports including comments by independent experts on its limitations.

What kind of research was this?

This cohort study examined whether there was any association between the use of vitamin and mineral supplements and the risk of death in nearly 39,000 women, with an average age of 61 years. The researchers say that although dietary supplements are now commonly taken to prevent chronic disease and maintain good health, the long-term health consequences of many of their ingredients are unknown.

Cohort studies, which enable researchers to follow large groups of people over many years, are often used to look at possible links between lifestyle and health outcomes. However, they cannot prove that one factor causes another. A randomised controlled trial would be needed to do this.

What did the research involve?

For this study, the researchers used data from a large US women’s health study aimed at examining associations between lifestyle and dietary factors and the incidence of cancer in postmenopausal women. At the study’s start in 1986, a total of 41,836 women aged 55 to 69 years completed a validated, 16-page questionnaire on their dietary and lifestyle habits, including supplement use. They were also asked about other factors such as their age, height, education, physical activity, diet and certain medical conditions, such as high blood pressure.

This current study included 38,772 of these women, excluding from the analysis any women who had not adequately completed the initial questionnaire on food and supplement use. The women also filled in further questionnaires in 1997 and 2004.

Each year, the women who had died were identified using state and national death registries, until December 2008. The underlying cause of death was also ascertained, and women were excluded who had died due to injury, accident or suicide (since it is unlikely that supplement use would be related to these outcomes).

The researchers analysed the data for any association between supplement use and mortality, using validated statistical methods. They performed additional analyses for shorter follow-up intervals, from 1986-1996, from 1997-2003, and from 2004-2008. They adjusted the results for other possible confounders such as age, lifestyle, smoking, alcohol and certain medical factors.

What were the basic results?

Among the 38,772 women that were followed up, there were 15,594 deaths over the average follow-up time of 19 years. Several commonly used dietary supplements were associated with a higher risk of dying earlier (mortality) compared with non use (of those particular supplements). The main adjusted results are summarised below.

Multivitamins were associated with a 2.4% higher risk of earlier death (hazard ratio (HR), 1.06; 95% confidence interval (CI) 1.02-1.10)
Vitamin B6 had a 4.1% higher risk (HR 1.10; 95% CI 1.01-1.21)
Folic acid had a 5.9% higher risk (HR 1.15; CI 1.00-1.32)
Iron had a 3.9% higher risk (HR 1.10; CI 1.03-1.17)
Magnesium had a 3.6% higher risk (HR 1.08; CI 1.01-1.15)
Zinc had a 3% higher risk (HR 1.08; 1.01-1.15)
Copper had an 18% higher risk (HR 1.45; 1.20-1.75)
Calcium had a 3.8% lower risk (HR 0.91; CI 0.88-0.94)

The findings for iron and calcium were replicated in separate, shorter-term analyses at 10, six and four years of follow-up.

How did the researchers interpret the results?

The researchers say their study raises concerns regarding the long-term safety of certain dietary supplements. They are particularly concerned about iron supplementation, for which they found a ‘dose response’ relationship, in that the higher the dose taken, the higher the risk of mortality. This association was also consistent across shorter intervals of time. They conclude that dietary supplements should only be used where medically necessary and are not justified for general use.

Conclusion

The study has some strengths in that it had a large number of participants who were followed over a long period of time. Also, their supplement use was assessed at the start of the study, and also in 1997 and 2004, and these repeated assessments help to strengthen the argument for the reliability of the study.

However, the study also has some limitations and it is not possible to give a definitive answer as to how these findings should be interpreted. For instance, the results were analysed using several different models, each designed to take into account the various confounding factors that could have affected the results. As an accompanying commentary points out, the most complex of these models found that the only supplements significantly associated with dying earlier were multivitamins and copper.

Importantly, the results cannot be interpreted to mean that supplements can cause early death. As the authors concede, it is possible that women were taking supplements in response to illness, which may have been the reason for their earlier deaths. In particular, many women take iron for anaemia, which is associated with chronic disease, injury and major surgery. All these factors could have affected the results.

The Department of Health advises that most people can get all the vitamins and minerals they need by eating a balanced, varied diet. Some groups who are at risk of deficiencies are advised to take supplements, and details of these groups can be found within the Common Health Questions section on NHS Choices.

In conclusion, few randomised controlled trials have been undertaken on the safety of dietary supplements, and further research is required in this important area. Dr Glenys Jones, a nutritionist at the Medical Research Council's Human Nutrition Research unit says: “Research in this area to date has had inconsistent findings, and what is needed to determine whether supplement use actually causes an alteration of mortality rate is a number of well-controlled intervention studies that can then be brought together and reviewed.”