by Mario Rizzo
A couple of weeks ago I received a letter from the Harvard Medical School inviting me, among many others, to participate in a placebo-controlled trial of the efficacy of omega-3 (fish oil) supplements and high dose Vitamin D in preventing disease. The trial would last for five years.
I thought about it for a few minutes. But it seemed very risky. Why?
Based on my reading of the evidence (as presented in the popular press but also in a few journal articles that I have read) I am convinced that these supplements are highly likely to be efficacious. In fact, I current supplement with 1200 IUs of Vitamin D and 3 grams of omega-3 per day.
If I were to enroll in the trial I would have to give that up and take the chance that I would be taking a placebo. Why should I do that?
The only reason I can think of is to be socially beneficent. I could help establish more definitively whether, in fact, this supplement regimen is beneficial. On the other hand, I am just one person – and one person or less won’t make much of a difference.
But the expected cost to me personally is significant. So “beneficence” would really be an altruism of sorts. This is a significant difference.
More importantly, those people most likely to participate in the study are those who are less informed, I should think, about health matters. They are probably less likely to follow healthful practices in the first place. I do not know how this will affect the results. It could be that scientists will find a greater or lesser effect in this group.
The point is that this problem affects participation in those cases, especially, where people are generally healthy now and simply want to stay that way. If, on the other hand, people have a hard-to-treat disease and the trial gives them a possible access to a drug that they cannot obtain in any other way, participation would be much more likely.
Is it time for people in clinical trials, such as the one to which I was invited, to be paid or given some other benefit? I realize that this may compromise certain criteria of randomness. But no system is perfect.