Here's the original report: http://jama.jamanetwork.com/article.aspx?articleid=1357266

I'll caveat that I'm not an expert but my gut reactions are:

1 - The study is a meta analysis (a study of different studies) with no standard controls in place. We don't know if the subjects had pre-existing conditions, eating habits, how much Omega-3, etc. Were these people eating McDonald's every day or some other traditional inflammatory diet (wheat, trans fats, etc)?

2 - I need to validate this but Omega-3 alone doesn't dictate potential benefits to cardiovascular health. I think the Omega-3/Omega-6 ratio is more important (http://www.ncbi.nlm.nih.gov/pubmed/12442909). Not sure if this info was available from the different studies being reviewed

3 - Along the lines of Omega-3 amounts, I can't tell from the article how the subjects were receiving Omega-3 supplementation. Some methods are more bio-available than others. Was it actually supplements or did they look at people eating fish and count that as Omega-3?

No. Without seeing the study, I would very seriously doubt that there was anything other than Omega 3 as the change to the patients. As paleo people, I think it's the reduction of Omega 6 that is as important if not more important than just cramming a bunch of Omega 6 down our throats. Raising the good doesn't mean getting rid of the bad. Working out regularly is great, but if you're still blowing lines of coke every night, it's probably not helping you much.

So, if they were still eating whole wheat bagels for breakfast and pizza for dinner, thrown in with some fried anything for lunch. You're still taking in all that stuff. The added Omega 3 benefits would be hard to prove if at all effective in that scenario. At least that would be my assumption without further data.

This study will have a big impact on clinical support for omega-3 PUFA dietary supplementation for improving cardiovascular health outcomes. It's an important study and it would be remiss to ignore it.

It's worthwhile pointing out the authors acknowledge there is evidence of the triglyceride lowering effect of omega-3 PUFA's - but given there is only a small association between triglyceride levels and coronary heart disease as per the two studies they cite (1) the triglyceride lowering effect alone may not significantly contribute to improving health outcomes.

The biggest source for bias was that almost half the numbers in this meta-analysis were derived from open-label studies.

Other issues were a lack of data on the EPA/DHA ratio and source of omega-3.

The biggest issue I have with this study, and, as is becoming apparent as time goes on with other nutrient-health interaction studies, is the nutrigenetic confounder (2). In other words, the nutrient-health outcome is only revealed when adjustment is made for genetic responders vs non-responders.

Personally, I would not be trashing the omega-3 supplementation just yet, but I would be following the research closely to determine if supplementation is of benefit for your type of nutrigenetic profile.

(1) 1992a 1992b

(2) 2006

I don't think you should get as much 3 as possible anymore, I think AA has some benefit and DHA too but that's about it. Keeping them low in the diet is more important than the ratio, they oxidize and that's not really a good thing. Just my opinion though.

http://raypeat.com/articles/articles/fishoil.shtml

http://www.cholesterol-and-health.com/PUFA-Special-Report.html

Yes, indeed, maybe some of the participants thought those fish shaped crackers that contain zero fish counted as fish. Maybe as Quoc Ngo mentioned, fish sandwiches at McD's. Maybe it was low quality fish oil that was cut with soy oil. Maybe they ate GMO farmed fish that's been fed soy/corn and had no exposure to their normal diet and had the same n3:n6 profile as CAFO chicken?

Maybe the participants lied. Maybe the sponsors of the study were subscribers to the "Heart Healthy" logo for their grain crap-in-a-bag products.

As Harry mentions, the study will have some impact, the question is, does it match reality, or will it serve to sway opinion away from the truth?

Maybe n3's don't have an effect on heart disease and stroke, but they're likely beneficial in other ways. Not everything is a cure to a disease. Sometimes removing the cause of a problem (cough, cough, SAD eating, especially hydrogenated industrial seed oils) is more important than supplementing with something that's beneficial.

Simply adding more Omega 3 onto a standard diet is unlikely to improve health.

Robb Wolf moderated his Omega 3 supplementation recommendations and explained why here: http://robbwolf.com/2011/10/05/fish-oil/

Here is a relevant quote (note, linoleic acid is an omega 6; EPA/DHA is omega 3):

"What I had not initially considered is signaling, be it anti or pro-inflammatory is a cell by cell affair, and this is based largely on the make-up of fatty acids in our cell membranes. In an inflamed, sick, standard American diet model, individuals have a significant overabundance of linoleic acid in their cell membranes. The idea of front loading more EPA/DHA to change the fatty acid profile of our cells is great until you run into the brick-wall of our metabolic machinery. Cell fatty acid turnover cannot be “goosed” from behind like shoving a bungee jumper off a bridge. Fatty acid turnover has a rate limiting step that is not “substrate limited.” Or, in non-geek-speak more fish oil will not make the process go faster. Instead we need to limit the intake of linoleic acid, keep a decent intake of EPA/DHA, but we need not, and in fact should not hammer that dosage, as we will see when we look at oxidative stress and free-radical chemistry."

Chris Kresser also wrote a great article on why reducing Omega 6 is more important than simply dosing up Omega 3's. http://chriskresser.com/how-much-omega-3-is-enough-that-depends-on-omega-6

Here is a great summary from that article:

To summarize, for someone who eats approximately 2,000 calories a day, the proper n-6 to n-3 ratio could be achieved by:

1) Making no changes to n-6 intake and increasing intake of EPA & DHA to 3.67g/d (11-oz. of oily fish every day!)

2) Reducing n-6 intake to approximately 3% of calories, and following the current recommendation of consuming 0.65g/d (three 4-oz. portions of oily fish per week) of EPA & DHA.

3) Limiting n-6 intake to less than 2% of calories, and consuming approximately 0.35g/d of EPA & DHA (two 4-oz. portions of oily fish per week).

Although option #1 yields 60% tissue concentration of EPA & DHA, I don’t recommend it as a strategy. All polyunsaturated fat, whether n-6 or n-3, is susceptible to oxidative damage. Oxidative damage is a risk factor for several modern diseases, including heart disease. Increasing n-3 intake while making no reduction in n-6 intake raises the total amount of polyunsaturated fat in the diet, thus increasing the risk of oxidative damage.

This is why the best approach is to limit n-6 intake as much as possible, ideally to less than 2% of calories, and moderately increase n-3 intake. 0.35g/d of DHA and EPA can easily be obtained by eating a 4 oz. portion of salmon twice a week.

This study does make me wonder about the efficacy of fish oil supplementation. There was a similar study done in April that questioned the benefits of fish oil supplements and reached similar conclusions. Still, the data does not discredit the benefits of eating wild fish, only the benefits of consuming those fatty acids in supplement form. It may be the case that though a diet high in fish promotes wellness, consumption of the same Omega 3s in isolation does not. Sometimes taking a supplement just isn't as effectual as eating real food. I'm no scientist, and, at least for now, I'll continue taking my supplements; but the prudent move is likely to employ a preference for eating actual fish. That's what I'm going to try to do because, as I said, this study does indeed have me wondering.