Omega-3's Didn't Work

Marangell LB, Martinez JM, Zboyan HA, Kertz B, Kim HF, Puryear LJ.  A double-blind, placebo-controlled study of the omega-3 fatty acid docosahexaenoic acid in the treatment of major depression.  Am J Psychiatry. 2003 May;160(5):996-8

Why? Are there any clues in the study?  Is it just that DHA doesn't work as well as an EPA/DHA combination?

Here are some details from Drs. Marangell and colleagues report.

A.  Small sample size:  can make a study "negative" despite some trend toward benefit, because results don't reach formal "statistical significance".  There were 18 patient in the DHA group, and 17 in the placebo group.  Their sample size would allow them to detect a "real" difference (5 points on the HAM-D, for you detail types) 80% of the time.  They found only a 3.7 point difference in response, placebo versus DHA.

B. Bad luck (whoa, really bad luck) during the randomization: their placebo group, apparently just by chance, was significantly more depressed than the DHA group.  So when they improved, as placebo groups generally do, they had a bigger distance toward "normal" that they could jump.  Some people did get that much better, perhaps from some of the good care that often goes along with a study like this.  But that made the improvement that omega-3s would need to cause, to be better than placebo, pretty big. 

Remember, it's not that the patients on DHA didn't improve, it's just that they didn't improve more than placebo, by as big a margin as was expected, and as much as would be necessary to be sure the difference wasn't just chance.  Whatever the reason for this "negative result", we'd still have to wonder whether using DHA alone (as opposed to EPA and DHA together, as in most fish oil products), might explain the lack of improvement compared to placebo. 

For the moment, we should probably conclude:  DHA alone is not a good idea at this point, based on this study compared to the results of EPA/DHA combination.  Similarly, the EPA-alone approach tested at high dose by Keck and colleagues, which also did not show benefit, suggests that for now, the best omega-3's to try are fish oils containing a combination of EPA and DHA.  Check the labels if you're going to try this approach. 



Visits since counter set October 15th, 2005: