Sunday, April 8, 2012

Fast Food Only Makes Us Happy via the Power of Clowns

Just in time for my Easter dinner of Slim Jims and Mickey Ds (kidding!  I ate grassfed steak and potatoes with spices and kerrygold in honor of Walter Willett) comes this downer of a news article:  Link Between Fast Food and Depression Confirmed.

That science daily piece is based upon a study newly published in Public Health Nutrition, Fast-food and commercial baked goods consumption and the risk of depression.  Now Public Health Nutrition sounds like one of those buttoned-up nutrition journals filled with good intentions and soy, so let's see what the quality of the work is, shall we?

All right, so this is a revisit of the SUN study.  I'm pretty sure I have some SUN study stuff around here somewhere, oh, here.   Basically it is an epidemiological visit to sunny Spain, where nearly 9000 university graduates were followed for about 6 years, answering questions about diet and subsequent health.

And guess what?  If you fill your plate with empty industrial fast food calories (weirdly, they call fast food "sausages, pizza, and hamburgers" in this study, which certainly can be fast food, but also not so fast…) and "commercial baked goods" including "muffins, doughnuts, croissants, and other commercial baked goods."  Okay.  Though I've probably eaten more homemade muffins than commercial ones, can't ever remembering making my own doughnuts or croissants, so they've got me there.

So all the folks did not have diagnosed depression at baseline, and over the years, several hundred began taking antidepressants or were diagnosed with clinical depression by their doctors.  And, three guesses as to how diet correlated with the diagnoses????

That's right, a linear, dose-dependent correlation between the amount of "fast food" at baseline and depression.    The "fast food and commercial baked goods" eaters were more likely to be young, and more likely to not eat other, more delectable foods, such as olive oil, fish, nuts, and vegetables (*cough* Mediterranean diet *cough*).  The fast food shovelers were also more likely to smoke and to work more than 45 hours a week (not recommended!).

If you recall, all those studies listing relative risk of whatever (such as every statin study ever written) are all hiding something.  In general you end up with headlines such as "A 49% increase of DEATH DEATH DEATH" but in actuality the risk of death goes from 1% to 1.49% which is hardly exciting.  I had to edit this next part after Jonathon's comment because I carelessly reported the numbers incorrectly the first time around.  I have several excuses having to do with poor time management and sleep deprivation, but no good excuses because I am not typically so careless… anyway, the percentage of change of the magnitude of the association varied as we went farther and farther from the original dietary measurement, discounting all the folks who had been diagnosed in the months prior to the last measurement of depression… these variations were from 1% or so to 14% or so for commercial baked goods.  The absolute numbers were 493/8964 people diagnosed with depression, or about 5% (this number is low compared to the general population, but many folks were excluded from the sample due to already having a diagnosis of depression prior to the study, which skewed this sample toward healthier folks with later onset of depression symptoms).  The hazard ratios aren't that impressive, but the trend is linear and significant for the fast food consumption.

The discussion is also interesting. The authors suggest that those who eat more fast food and baked goods have a higher trans fat consumption (likely true), and therefore poorer glucose control and poor glucose control (staggered they mention trans fats and not saturated fats here!!  Do these folks actually read the nutrition literature?) Then, even more surprisingly, the authors have a cogent discussion of how  high glycemic carbohydrates (found in abundance in commercial baked goods) will increase plasma tryptophan and increase serotonin uptake into the brain and thus decrease depressive symptoms (blah blah blah)… but these effects seem to have been studied only short term, and that long term, inflammation and pro-inflammatory cytokines are more associated with fast food-style diets and these are more likely to be important to longer term depression.

Phew, I was beginning to think these researchers were way too sensible when they went into a discussion of how high-fat, ketogenic diets can disrupt cognition (pray tell me how likely it is a fast food diet is highly ketogenic?)

And, let's not pull punches on the major weakness of this study.  Diets were only assessed at baseline, and follow up occurred up to 8 years later.  That's a lot of unknown eating to account for.  I also thought the previous study of trans fats wasn't that helpful because it seemed that college graduates in Spain didn't have that much variation in how they ate.  Meaning hardly anyone ate that much vegetable oil, and trans fats were mostly from dairy, as opposed to an American sample, where trans fats can make up a scary percentage of calories.

Still, correlations being what they are, if you are the sort of person to be quaffing from the drive-thru, you are also apparently far more likely to be the sort of person to be diagnosed with a clinical depression.  Confounders galore, but there you have it.  Now, back to my Slim Jims.


  1. Did they have depression scores recorded at baseline and then compared with their scores at the end or was it just a yes/no depressed/not proclamation? If I was a no-good rascal I would suggest that people with subclinical depression, even if they don't have clinical depression yet, eat more highly-stimulating food to try to boost their spirits.

    But I'm not a no-good rascal. Right?

    Nah, I'm pretty down with the inflammatory/nutritional/metabolic energeic influence on depression. No need to convince me! More data isn't ever a bad thing as long as we use it right. And we do!

  2. Actually Stabby they were very careful to avoid that confounder and didn't even include people diagnosed with depression in the first 0-12 months (in some scores, 0-6 months in others, which is troubling, but c'est la vie… my main issue with this cohort is that well-educated Spanish folk all tend to eat fairly well)

  3. Aha! Nicely done. Since I already agree that it is probably causation I guess I'm just happy to see researchers being meticulous. Yay science!

  4. Someone seriously suggested once that places like McDonalds could offer statins next to the ketchup packets. Why not offer an assortment of anti-depressents too (heehee)? I just wish more common knowledge would be, "poor quality nutrition (meaning low in micronutrients, not calories) causes ______" (depression, adhd, obsity, dibetes, fill in the blank). Then someone with ______ could see if diet really did make their symptoms worse.

  5. Hahaha! That's the best blog post title ever.

  6. Serotonin resistance. This happens if you take SSRIs for too long, so why not on high-carb diets?
    It's how you make use of what you get (most tryptophan on a protein-adequate diet is degraded via kyneurate anyway, proportionately little is used for proteins or serotonin/melatonin).
    Given that serotonin has effects on blood statis and inflammation, you'd want to keep a tight rein on it.

  7. @George plz provide scientific evidence for serotonin resistance. Altie quack-dar is on the blink right now.

    Depression is about a whole lot more than serotonin (#1) and crappy diets cause depression for tons of reasons other than serotonin (#2) and eating a crappy diet may be a marker for depression rather than causative (#3). I don't know about you but if I am sitting staring into the abyss wondering how I am going to keep living , feeling as if life is a constant rewardless struggle and all that lies in my future is pain and everything is hopeless, I"m not going to be grilling up some salmon and making fresh olive oil salads. I'll probably be eating canned food, not bathing daily, and not leaving house as often.

  8. How did you get the 0.6-1.2% vs 7.9-13.6% figures?

    As I read it, that was the change in the association between fast food and depression once they eliminated those in the study who were diagnosed with depression 6,12,18,24 months after baseline.

  9. Hello doc,
    I have noticed that if I reduce or try to quit prozac,the symptoms of ocd, depression bounce back with more power. What do you think , what is the reason for that?

  10. Emily,

    do you know anything about not eating 'enough' healthy stuff?

    Give care,


  11. Another thing Emily,

    I like the title of your blog, Evolutionary Psychiatry, and I recognise your profile picture is a realistic manner. But why don't you, or we together, think how psychiatries can have good influence in their patients their lifes..

    Like medical treethments against smoking, like drinking enough water in the way I figured in my own life water with a few drips of concentrated orange juice is much nicer than fast sugar drinks and light drinks, that exercise is a must and must be made tempting and pleasant and more regularly offered, like vegetarian meals to be low fat as they tend to be very more fatty than meat diets, like rules on listening music..for instance to always use headphones and to offer them for rent or so..and to use classical music to enter a stage of relaxation or sensible core, like to make rules that one patient has nothing to do with the other, etc.. :)


Tired of receiving spam comments! Sorry, no new comments on the blog

Note: Only a member of this blog may post a comment.