Essential Fatty Acids: Ratios versus Amounts


Continuing the discussion from DIY - Optimally efficient bio-available micronutrient sources: research / articles / spreadsheets?:

There are a lot of threads here discussing ratios of Omega 6 to Omega 3. As I’ve spent time looking into it, I found that older research tends to wrap their findings in terms of the ratio of Omega 6 to Omega 3, but the newer research tends to just look at the level of Omega 3. Theory has evolved, and these days the medical and scientific community has moved away from a focus on ratios. It seems the important part is simply the level of Omega 3’s.

A lot of research was done based on ratios, but when you re-cast their results based on Omega 3’s, the results are comparable or even clearer. And as anyone who has looked at a DIY formula quickly realizes, in practice, it’s very hard to bring down the ratio of 6:3 by reducing Omega 6, and still get any decent level of fat in the diet - it’s far easier to bring up the level of Omega 3.

The idea of most “ideal” ratios seems to be a wrong turn for the nutrition of fats, but rather than dying off, as it has in the scientific community, it’s taking off in popular nutrition… I think this especially because of the Paleo trend. Those trying to mimic various “traditional diets” look at the ratios in those foods and try to emulate them. Within Paleo, I even see people trying to come with ideal ratios of Omega 9 : Omega 6 : Omega 3. I don’t think those attempts are on sound scientific footing.

Advice on DIY soylent

In all your research have you come across anything that looks like a general consensus of an acceptable omega 3 intake range?


I have the same concern, and assume that the current ω3 to ω6 ratio is limited by cost factors (not taste or nutritional guidelines). As far as I can tell, the myco-source ω3’s are only available from DSM at this point, and you better believe that a single-source ingredient is expensive (even if the process is more sustainable and scalable than fish-based ω3s).

(I’m not going to discuss ω9’s because there was never a major dietary source for early hominids, which is why the body can synthesize what it needs from other fatty acids endogenously.)

It makes sense that research would focus on amounts of fatty acid ingested rather than 3/6 ratios because each type of fatty acid produces its own unique types of eicosanoids – even if in general ω6 eicosinoids are pro-inflammatory while ω3 eicosinoids are less so, the actual impact of each type of eicosinoid on health and disease is less well defined. One thing to consider though, is that those studies don’t usually have the ability to control fatty acid intake through diet because it is simply 100x cheaper and easier to give a subject a supplement on top of their normal diet. With Soylent, you have control of all fatty acid consumption, and could potentially look at ratios for all the fatty acid (not just 3/6) and subsequent impacts on inflammatory conditions.

I wish the label indicated the amount of ALA in addition to the stated amounts of EPA and DHA… For example, research related to the affective benefits of EPA generally find clinically significant doses with 1000mg or more per day – a far cry from the 270mg/day in Soylent 1.2. Since the eicosanoids produced by EPA are different than those produced by DHA and ALA, those ω3’s do not have the same positive impact on major depressive disorder. DHA has been shown to have positive impact on cognitive decline at doses of 900mg/day (closer to, but still higher than in the 1.2 formula). While all three ω3’s have a positive impact on cardiovascular health and rheumatoid arthritis, but by influencing different mechanisms, so you can’t replace one with the other.

TLDR: I would like to see better EFA labeling, higher amounts of EPA & DHA without reducing the current amount of ALA (matching levels of clinical significance for inflammatory conditions rather than just meeting minimum dietary recommendations). However, the the current levels are probably restricted by cost.


It’s still very broad.

EPA and DHA still aren’t “essential,” because the body can produce them from ALA… but the conversion rate is very poor for women, and worse for men. There’s a an AI level (Adequate Intake, a sort of minimum) of 1.6 g ALA for men, and 1.1 for women… and no requirement for EPA or DHA. Coversion to EPA/DHA is roughly 10% for men, and 30% for women, so you’re getting a small fraction of 1 g of EPA/DHA from ALA.

On the other hand, there are so many benefits to getting closer to 1 g of EPA/DHA, that some groups are advocating it directly (usually by advocating fish a couple times a week).

Internationally, the recommendations are based on percentage of total calories - most saying to get between .5% and 2% of energy from omega-3’s. In a 2000 Kcal diet, that’s between 1.1 and 4.4 g of omega-3.

Things are muddied a bit because a lot of the research on omega-3’s is based around looking for a therapeutic benefit in specific disease conditions… but at those higher levels, ALA thins the blood, and EPA/DHA can lead to real bleeding problems. Individual response varies… the AHA recommends anyone with CHD get 2-4g of EPA/DHA per day, but under a physician’s care, and your pharmacist needs to know, too, because of interactions with the blood-thinning effects of other medications.

If you’re eating a typical American diet, which pre-disposes you to CHD and atherosclerosis, getting 500 mg or maybe 1g per day is probably a good idea, even for perfectly healthy people. On the other hand, if you’re strictly consuming Soylent, it may be completely unnecessary, unless you already have atherosclerosis from your prior lifestyle.

It’s a very hard call to make, and it’s very personalized. Me? In addition to sometimes eating fish, and to the ALA in my canola oil when I’m on DIY soylent, I take 1800 mg of EPA/DHA per day as fish oil caps, which is a bit high. On the other hand, I’m a large man (6’5", 225 lb.), and I’m also in a family with a history of coronary artery disease, and my personal genetic testing results show me having 3.4 times higher risk than average for venous thromboembolism - so getting a blood-thinning effect is probably a good thing.


Fungus sourced?

Yes, but at that level and above, you’re talking about therapeutic doses. You’re effectively medicating. That’s counter to the stated goal of Soylent: everything you need, and nothing else. Anyone who needs/wants to can add a fish oil cap to Soylent. Forcing people to discard the oil is another matter entirely.

Personally, I disagree, and want more than 100% of the DV for many of my micronutrients… but I accept that it’s a personal decision, and that it’s not what Soylent is trying to do.


Whoops, I meant algal-sourced.

I get your point and definitely agree that individual diet and genetics influence your need for dietary supplementation, and that its not a big deal to take an extra gel tab with your Soylent, but I’m not convinced that the doses are actually “therapeutic” are not simply protective. The AHA and FDA recommends up to 3g/day ω3, because there is not a lot of research on the impact of doses higher than that (and those standards were set in the early 2000s). However, there are a few recent studies have looked into the risks of high ω3 consumption and not found significant anticoagulation risk with up to 4g/day (check out page S211 of this metanallysis Considering all the protective mechanisms discovered by high EPA and DHA consumption, I’d like to see 1g of each.


By the way, looks like DSM doesn’t have a monopoly, they just dominate the market:


It’s interesting that you say the trend is leaning towards a focus of Omega 3 alone. I won’t dispute that by any means, although I must confess I had actually gotten a sense of just the opposite. Granted I have no specific source to back that up, and that “sense” may be influenced more by diet fads more so than medical research, as you say.

But let’s say that holds true… what’s the level of DHA/EPA to shoot for? This study suggests that up to 5g/d of DHA/EPA combined a day in healthy adults presents no adverse affects. But that also begs the question: is 5g/d necessary? I do not have any known bleeding/cholesterol concerns, but is there a real benefit to exceeding more than, say, 3g/d? Or even less?

Assuming there is some validity to maintaining a proper ratio - you mentioned the difficulty of keeping Omega 6 low while also raising Omega 3, anyway - I wonder if settling in the 2-3g/d for EPA/DHA and then bringing O6 down closer to 10g/d? But then you’re below the minimum recommendation… Lol.

I feel like this is something that I’m likely worrying entirely too much about, but hey, that’s the point of this forum!


I see the good Dr already beat me to saying just about the same thing about Omega 3s… But the ratio would need to be addressed still. Unless a 14-17:2-3 (6:3) is perfectly reasonable.

/edit 2

In regards to medical vs therapeutic, I think that’s a very valid concern. But Soylent has to operate at a macro level to protect itself; DIY we can get much more individually detailed and push the limit. That said, I’m not particularly interested in paying for excess nutrients that aren’t giving me any discernible benefit.


Check the ratio section here, it echos the OP:

As long as there is enough of each EFA (ω6=AA, LA, GLA, DGLA, ω3=ALA, EPA, DHA) to get its beneficial effects, the ratio probably not so important. But there is still more research that needs to be done.

The Greenland Inuits derive 50% of their calories from fat with a ratio of 1:7 ω6:ω3 and they still have lower rates of cardiovascular disease than Americans.


And a higher rate of stroke due to excessive omega3 consumption causing bleeding issues.


Is it true? i genuinely want to know.


I did read that early studies associated high rates of nosebleeds with the high n3 diets in Inuits, but I can’t find them. I looked at a few, this is the most comprehensive which looks at impact of the diet rich in n-3’s on a number of morbidities.

I also found this study which says n3’s are associated with lower rates of stroke by thrombosis (blood clot) but “slightly higher rates” of hemmoragic stroke (from bleeding) but in the end concludes that the safety risk of n3 therapy is not conclusive.

That is also echoed by this paper on n3 therapy safety:


Yes, with a caveat.

“Although excessively long bleeding times and increased incidence of hemorrhagic stroke have been observed in Greenland Eskimos with very high intakes of EPA + DHA (6.5 g/day), it is not known whether high intakes of EPA and DHA are the only factor responsible for these observations.” That’s from LPI citing Food and Nutrition Board, Institute of Medicine. Dietary Fats: Total Fat and Fatty Acids. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol, Protein, and Amino Acids. Washington, D.C.: National Academies Press; 2002:422-541.

So it’s not directly proven, but it’s a strong theory. EPA/DHA have blood-thinning effects and are known to increase clotting time. And the Inuits there, who eat more EPA/DHA than pretty much any other population, have problems with excessive bleeding and strokes.

All the omega-3’s, but especially EPA/DHA, are associated with increased bleeding times and blood-thinning effects. Prevalent theories say that this is exactly why they are beneficial against cardiovascular diseases. They also have anti-inflammatory effects, and there has been a lot off research around other conditions, but the effect and evidence are not as strong as for cardiovascular problems.


I thought I’d clarify something on strokes, because the discussion above may seem contradictory.

A stroke is when brain tissue is damaged because it stops getting blood. This happens two different ways:

  • one, “thrombosis,” involve a clog in the plumbing, so blood flow is reduced or stopped
  • one, “hemorrhage,” involves a leak in the plumbing, so blood spills out instead of
    getting delivered

Either way, the blood doesn’t reach the part of the brain that needs it. You can imagine that thickness level of your blood can influence both of these.

Thinner blood can more easily get through narrowed arteries and around clogs, so EPA/DHA helps here, and can even help prevent clogs, which are made of blood clots, because it has anti-coagulant properties - anti-coagulants inhibit clotting.

On the other hand, blood that is too thin will more easily leak - either through large ruptures or small pores. And if a hole does open, the body relies on the clotting action of the blood to seal it up quickly. So excess EPA/DHA, which thins the blood and inhibits the clotting action works against you.

So, two different types of stroke. Clogs get worse if blood is too thick, so adding EPA/DHA helps. But if the blood gets too thin, then leaks get worse, so taking too much EPA/DHA makes it worse.


It’s very easy to get a particular impression from the lay press. The lay press is writing about journal articles, but writing for the general audience. The general audience usually doesn’t have access to the underlying journal articles, and has to go by the press coverage.

Often, in the lay press, you’ll see things like this: they cover a new study that found benefits for EPA/DHA. They talk about the benefits associated with increasing EPA/DHA in the amounts studied, and the measurements of the benefits in the study. And then they close with, “this is just more evidence that you need to keep your omega-3:omega-6 ratio above 1:4,” or something like that. That part is their own opinion, and you can’t even say that’s it’s inherently wrong, because working to keep your ratio at 1:4 will most likely mean increasing your EPA/DHA, as in the study.

If the author of the lay press article is in the Paleo camp, an intake ratio-type-conclusion seems always seems to get injected, but if you read the article carefully, the underlying study is generally not talking about ratios, at all.

In fact, in the few studies that actually talk about ratios, they’re usually talking about plasma phospholipids. That is, they’re talking about the ratios of different oils actually in the blood, not the ratios being eaten. Plasma (blood) fat ratios do not necessarily correspond to intake (food) ratios at all, except inasmuch as eating more of a particular fat usually leads to a rise of that fat in the blood (or a rise in something related to that fat.)

There are some sources which are specifically set up as medical professional interpreting the research for lay people - for example, the Harvard School of Public Health. Places like this don’t pepper their advice with technical terms and journal citations, but they also don’t have an agenda associated with a particular dietary belief (like Paleo) or selling a product (like a supplement store.) Here’s their take on Oemga-3 fat ratios:

From Ask the Expert: Omega-3 Fatty Acids,

What are omega-6 fatty acids? Should I be concerned about the ratio of omega-6 fatty acids to omega-3 fatty acids in my diet?

Omega-6 fatty acids (also known as n-6 fatty acids) are also polyunsaturated fatty acids that are essential nutrients, meaning that our bodies cannot make them and we must obtain them from food. They are abundant in the Western diet; common sources include safflower, corn, cottonseed, and soybean oils.

Omega-6 fatty acids lower LDL cholesterol (the “bad” cholesterol) and reduce inflammation, and they are protective against heart disease. So both omega-6 and omega-3 fatty acids are healthy. While there is a theory that omega-3 fatty acids are better for our health than omega-6 fatty acids, this is not supported by the latest evidence. Thus the omega-3 to omega-6 ratio is basically the “good divided by the good,” so it is of no value in evaluating diet quality or predicting disease.


Based on everything in this thread and elsewhere, can anyone find a certifiable fault with the following conclusion:

Omega 3: 1.6g ALA (per AI) + 500mg - 1g EPA + 500mg - 1g DHA = 2.6 - 3.6g/d
Omega 6: 14-17g/d (per AI)
Ratio: Prioritization debated, however due to the competition between ALA and LA, it’s reasonable to suggest that a lower ratio would be preferred. The priority is an adequate Omega 3 intake.

(… I am still looking for solid evidence to confirm the Omega 6 requirement is actually much less than the DRI recommends, which I have seen suggested in multiple places, but cannot confirm.)

As it stands, I will build my recipe focused on getting 3g/d~ of ALA/DHA/EPA, and just keep an eye on the Omega 6 levels. Which, coincidentally enough, is more or less what most of you have been saying! :wink:

Making sense of USDA database Essential Fatty Acids - "18:3 undifferentiated," etc

I see no major fault in the general ranges you’re looking at for Omega 3’s or Omega 6’s.

I assume you mean DHA, not DPA. (DPA exists, but it’s a relatively uncommon Omega-3 which is an intermediary between EPA and DHA.)

One thing I’d add: unless you want to drive yourself crazy, don’t try to figure out what ratio of EPA:DHA you should be taking… old recommendations seemed to be to get more EPA than DHA, but DHA is enjoying a resurgence. Most fish oil caps have more EPA than DHA, but some don’t, depending on the fish source, and algal oil usually has more DHA than EPA.

On the plus side… anything you don’t need probably just gets used like regular fat - burned for energy or stored, unless you’re taking so much that it causes problems from blood thinning… so I just try to get enough and try not to worry about it.


Oops, yes you’re correct. I fixed it to DHA.

Do you have any insight or opinion on minimum requirements for Omega 6 (assuming proper Omega 3 intake)? Along with everything else, there is conflicting literature out and about. I’m not particularly concerned at this point, but I am curious.


I haven’t read much on Omega 6; I just accepted the US Adquate Intake level of 17 g (for men) per 2000 Kcal and spent my “oil time” figuring out what I could about Omega 3’s. International minimums for Omega 6 go as low about 9 g, and the range goes as high as 25 g.


That sounds like a very healthy approach.

I think we can all agree that there are optimal levels of each essential fatty acid for healthy bodily function and disease prevention – I am simply biased in my interpretation of the research toward heavier n3 intake due to working on clinical trials for depression therapy for a few years with n3’s and a pilot associating rheumatism with mood and n3 intake in elders. I’m in 95% agreement with MentalNomad, am just am a little heavier on the n3 intake and am not convinced that there is a linear relationship between n3 intake and blood coagulation beyond a few grams in either direction. Regardless, scientific consensus is still a decade or two away.