Saturated fat: The advice and the science

Saturated fat

Is the advice to eat less saturated fat based on any good science? Well, not necessarily according to this fantastic new review:

Nutrition: Saturated fat and cardiovascular disease: The discrepancy between the scientific literature and dietary advice

The article analyzes the references for three sources of dietary guidelines: USDA, IOM and EFSA. They all recommend avoiding saturated fat. But they need all their imagination to find the science to back it up: 

Lots of embarrassment

I recommend the article above, it’s an amusing read. I particularly liked how the USDA refers to Jacobsen at al, a pooled analysis of 11 cohort trials, and its finding that eating more carbohydrates (like bread) or monounsaturated fat (like olive oil) instead of saturated fat (like butter) actually seem to increase the risk of heart disease.

To put this embarrassing finding “in perspective” they referred to another observational study that found the opposite correlation. The only problem? That study was actually an outdated version of one of the 11 studies pooled in Jacobsen et al. In other words, they chose to believe in one old study, instead of 11 studies combined – including a newer version of that same study!

More embarrassment in the Nutrition article.


  1. Violeta A.
    The article's summary states that "a meta-analysis of 61 prospective studies found the ratio of total to HDL cholesterol to be the strongest predictor of ischemic heart disease mortality". What is considered to be a good ratio in this respect? Thanks.
  2. This is an interesting paper. A brilliant overview of available scientific data. I find the conclusion in line with a recent article I wrote on my blog; "The Case Against Saturated Fat".

  3. This is old now, to be honest. Also find Uffe Ravnskov's new commentary on the paper (

    If you read the discussion between Hoenselaar (the guy who wrote the review your praising) and Pederson (some other guy) he actually adduces original ecological data based on European statistics that shows a statistically significant inverse (protective) association between saturated fat and CHD. The graph is particularly compelling.

    I blogged about this here ->

  4. Dr. Andreas Eenfeldt, MD Team Diet Doctor
    Very interesting. I just sent you a mail regarding this.
  5. For those who haven't caught up on this arguement where online blogger takes on and wins against the researchers you can find it detailed here.
    Heart disease and stroke are highest in countries with low saturated fat consumption. I'm sure that isdatzo is Hoenselaar R

    Hoenselaar R has an English Website here Diet and disease Evidence based information for healthy and sick people

  6. Dr. Andreas Eenfeldt, MD Team Diet Doctor
    I'll do a post on this spectacular inverse association in Europe. It's simply mindblowing.

    The French paradox may have turned into the Pan-European paradox.

  7. RE Uffe Ravnskov's new commentary on the paper (
    More discrepancies around saturated fat and cardiovascular diseases Uffe Ravnskov Free full text online here.
  8. Moreporkplease
    Hi Andreas,

    What French paradox? In the last 10 years obesity in France has doubled; diabetes is rapidly rising. The French now eat nearly as much sugar as Americans. They have switched to buying processed food at supermarkets. People now drink soda at home. American-style diet plans are making inroads:

    We can't stop the crisis until we admit it's truly global and all about sugar, soda, processed food.

  9. Dr. Andreas Eenfeldt, MD Team Diet Doctor
    Yes, that's just sad!
  10. Ralph, Cleethorpes, UK
  11. Arshad
    Just found the source of the above news:

    Associations among 25-year trends in diet, cholesterol and BMI from 140,000 observations in men and women in Northern Sweden

    Ingegerd Johansson, Lena Nilsson, Birgitta Stegmayr, Kurt Boman, Göran Hallmans and Anna Winkvist

    Before commenting on the study, we need to read full article....

  12. @ Arshad
    But if we look at the graphs here
    Is the use of cholesterol in mortality risk algorithms in clinical guidelines valid?
    Surely the optimum Cholesterol level for men is 5.5mmol/l 21mg/dl and for women it may be higher?
    Look at the WHO data on cholesterol and CVD or All Cause Mortality
    Zoe Harmcombe has plotted it out here Scroll down the page.
    Health h Correlator together with O Primitivo were checking this data in 2009.
    There is an optimum level, sweet spot, in cholesterol levels below which All Cause Mortality rises, as indeed so does CVD.
    Eating more saturated fat raises cholesterol levels to that sweet spot and therefore we find countries with higher saturated fat intakes have lower CVD incidence.
  13. It may also be helpful for people to refresh their understanding of the role of cholesterol in the body.
    I'm afraid this does require a certain amount of work but Peter Attia has done a good job of explaining cholesterol in the series of posts here The Straight dope on Cholesterol I found the links to his sources and the discussion in his comments arising also helped.

    I think the misleading message that cholesterol lowering is always beneficial needs to be challenged.

    For people my age each 1-mmol/L increase in total cholesterol is associated with an approximately 12% lower risk of noncardiovascular mortality So is it sensible to suggest I should be trying to get my cholesterol below 5.5mmol/l ~ 212ng/dl?

    By eating more saturated fat my TC level has risen to that associated with the lowest rate of all cause mortality, explain to me why that isn't a good thing for me?

  14. Arshad
    Thanks a lot, Ted Hutchinson. Your comments are always great, & very informative.
  15. Had my blood checked a week ago and the results of eating a high fat diet seems to have been good to my numbers.
    Before hflc: Total 178, hdl 58, ldl 94, vldl 26, triglycerides 130.
    After a few months of hflc: Total 141, hdl 49, ldl 84, vldl 8, triglycerides 40.
  16. Jennifer Morris
    I'm wondering if the low fat, high carb eating is the reason we're seeing more Alzheimers in the Baby Boomers? As well as other neurological diseases, like Parkinsons. I believe fats are essential to neuro health. If this is so, the damage done by high carbs is terrifying!
  17. Jennifer Morris
    I've got fifty pound to lose, stopped statins, and I'm not losing weight. I'm eating on average daily fats 55- 65%, carbs 6-10%, protein 25-35%. Calorie count is over my recommended amount by 150 cal or more. I read that it can take 4-6 weeks to become ketoadapted, but not losing anything!? Am I doing this wrong I wonder? Eating too many carbs?
  18. Zepp
    One have to be Ketoadapted first.. to know how your body do react to this way of eating!

    There is no one size fits all.. and one have to learn how ones body do react on a high fat diet!

    The thing about keto diet is that ones body do uppregulate the capability to use fats as predominant fuel!

    And after som beginners problem and struggle to eat a ketogenic diet.. one must learn about your own special body and mind how to do it best.. for that goal one have.. and to that scrutinise if the goal is realistic!

    Its a great beginners diet plan so far.. one can always eat more fat.. but thats your own desicion.. if one have to be hardcore or not!

    What one should take notice about is if ones apetit get less.. its often that how is a sign that your body do take some of the energy from your fat depots.

Leave a reply

Reply to comment #0 by

Older posts