"Multivitamin" Benefits Roughly Equal to Harms

During an average 11 years of follow-up, 28,851 deaths were identified. In Cox proportional hazards models controlling for tobacco use and other potential confounders, no associations were found between multivitamin use and mortality from all causes
(for users vs. nonusers:
hazard ratio = 1.07, 95% confidence interval: 0.96, 1.19 for men;
hazard ratio = 0.96, 95% confidence interval: 0.85, 1.09 for women), cardiovascular diseases, or cancer. The findings did not vary across subgroups by ethnicity, age, body mass index, preexisting illness, single vitamin/mineral supplement use, hormone replacement therapy use, and smoking status.


The 1.07 hazard ratio (7% higher chance of dropping dead at any moment) for men is indeed not significant; a simpler model that doesn’t control for as many things (including use of some single supplement e.g. selenium) gives a HR=.99, for example.

It’s unlikely that vitamins do nothing. Yet they appear to, in this fairly large study of old people. I’d say this is reason to look at fine-tuning vitamin/mineral intake. Those men who took 1-6/week instead of 1x or 2x/day seemed to fare slightly better than 0/week (TOTALLY not significant though).

Too much of some things in diet+multivitamin is likely quite harmful to health. If you ignore statistical significance, the weak evidence is that those old people who take multivitamins for a 5 year stretch are slightly more likely to end up dead 10-15 years later.

Or, supposing that multivitamins aren’t harmful (they obviously vary tremendously in composition), it’s really true that it’s unlikely that your normal diet is lacking anything you need, and the non-significant effect is mostly due to confounds like taking vitamins because you feel your health start to fail, or taking vitamins because you live a healthy lifestyle already, etc.

Unfortunately, that multivitamins do nothing, or are both helpful and harmful, with the harms slightly outweighing the benefits, is the simplest explanation of many such null results in the literature. But I don’t want to accept it, because I know of some specific causal evidence suggesting that individual vitamins, when supplemented, yield health benefits in a majority of people. So I’ll hold the slightly more complicated: “most multivitamins are too-highly dosing at least a few things, in a way that harms health as much as the rest of it benefits health”, tentatively.

Another nearly siginificant pattern: people who’ve used vitamins for at least 5 years die less than those who recently started or never used (but again, not really significant). Motivated speculation (not wanting to believe multi-vitamins are actually net-harmful or -useless): old people who suddenly start taking multivitamins did so due to correctly feeling their declining health and imminent death. So their higher rate of death shouldn’t be blamed on having recently started multivitamins.