The American Heart Association: Do These People Read Their Own Data?

After writing my last post, I got to thinking a bit more about the UCLA study that set me off – the one in which the authors noted that 72.1% of people hospitalized for a heart attack had LDL levels below 130 mg/dl (the recommended level) and yet, instead of concluding that high LDL doesn’t cause heart disease, speculated that LDL targets should be set even lower.

Specifically, I wondered what percent of the population is below that level. After all, if 95% of all adults have LDL below 130 but only account for 72.1% of the heart attacks, then you could reasonably say LDL may have something to do with heart attacks. Or looking at it from the other direction, if 5% of all adults have LDL over 130 but account for 27.9% of all heart attacks, then yes, high LDL may be a problem.

It took some digging, but I eventually founds the statistics I wanted on the American Heart Association’s web site, both in web pages and PDFs. Here’s the magic number:

Percent of American adults (over age 20) with LDL Cholesterol at or above 130 mg/dl: 32.6%.

Just roll that one over in your brain for a moment. Now combine it with the UCLA study data, and here’s what you get:  people with “high” LDL make up 32.6% of the population, but account for just 27.9% of the heart attacks.

Boy, doesn’t that just make you want to toss back a Crestor cocktail? And keep in mind, that group would include people whose LDL levels are 130, 150, 175, etc.  We’ve been told for decades that the higher your LDL, the more likely you are to clutch your chest in the middle of the night. But if the “high” LDL group experiences slightly less than their share of heart attacks, how can that possibly be true?!

The lead investigator for the UCLA study suggested that the “optimal” LDL level hasn’t been set low enough; perhaps it should be 40-60 mg/dl.  Excuse me? Are we now supposed to believe that LDL is dangerous, but all levels above 60 are equally dangerous? Given the statistics I just cited, that’s the only possible explanation, except for this one:  LDL doesn’t freakin’ matter. It’s a meaningless number.

The AHA site also offered a slew of statistics broken down by race. Don’t bother trying to lock these in your brain because I’ll revisit them in a moment:

Among non-Hispanic whites age 20 and older, the age-adjusted prevalence of total blood cholesterol levels over 200 mg/dL is 45.0 percent of men and 48.7 percent of women; 15.3 percent of men and 18.1 percent of women have blood cholesterol levels of 240 mg/dL or higher.

Among non-Hispanic blacks age 20 and older, the age-adjusted prevalence of total blood cholesterol levels over 200 mg/dL is 40.2 percent of men and 41.8 percent of women; 10.9 percent of men and 13.1 percent of women have blood cholesterol levels of 240 mg/dL or higher.

Among Mexican Americans age 20 and older, the age-adjusted prevalence of total blood cholesterol levels over 200 mg/dL is 51.1 percent of men and 49.0 percent of women; 16.8 percent of men and 14.3 percent of women have blood cholesterol levels of 240 mg/dL or higher.

The age-adjusted prevalence of U.S. adults age 20 and older with LDL cholesterol levels of 130 mg/dL or higher, which is associated with a higher risk of coronary heart disease, is…

For non-Hispanic whites, 31.5 percent of men and 33.8 percent of women.
For non-Hispanic blacks, 34.4 percent of men and 28.6 percent of women.
For Mexican Americans, 42.7 percent of men and 30.4 percent of women.

“Age-adjusted” means they’ve balanced the data to compare people of similar ages.  If the average Mexican-American is 27 and the average white American is 39, it’s a bit pointless to compare the two groups directly.  With that mind, let’s organize these numbers so they make sense.  (I can’t stand clunky terms like “Non-Hispanic Whites,” so at the risk of offending the tender-hearted, I’m just going to refer to all three groups of over-20 Americans as White, Black, or Mexican.) 

Men   White Black Mexican
Cholesterol > 200 45.0% 40.2% 51.1%
Cholesterol > 240 15.3% 10.9% 16.8%
LDL > 130 31.5% 34.4% 42.7%

Wow, looks those Mexican-American men are in deep trouble.  More than half have total cholesterol over 200, and nearly half have LDL levels above 130.  Fewer blacks than whites have high total cholesterol, but more blacks are in that “high LDL” category, so let’s see … going by the prevailing theories, I’m going to say Mexican-Americans have the highest of heart disease, then blacks, then whites.  So let’s check that analysis against the actual rates of coronary heart disease, also available on the American Heart Association site, and add that as our final table row:

Men White Black Mexican
Cholesterol > 200 45.0% 40.2% 51.1%
Cholesterol > 240 15.3% 10.9% 16.8%
LDL > 130 31.5% 34.4% 42.7%
% with CHD 9.4% 7.8% 5.3%

Whoops! Just exactly the opposite of what the lipophobes have been telling us all these years. Mexican-American men, the “worst” in every cholesterol category, have the lowest rate of heart disease. Well, perhaps the experts were talking about women all along. Let’s check the data:

Women White Black Mexican
Cholesterol > 200 48.7% 41.8% 49.0%
Cholesterol > 240 18.1% 13.1% 14.3%
LDL > 130 33.8% 28.6% 30.4%

Looks like the black women caught a break here. They’ve got the best cholesterol levels in every category and therefore surely have the lowest rate of heart disease. Let’s add the actual rates of coronary heart disease in our final row:

Women White Black Mexican
Cholesterol > 200 48.7% 41.8% 49.0%
Cholesterol > 240 18.1% 13.1% 14.3%
LDL > 130 33.8% 28.6% 30.4%
% with CHD 6.9% 8.8% 6.6%

Uh, gee, that’s kind of embarrassing, isn’t it? Mexican-American men, the group with the worst LDL figures, have the lowest rate of heart disease. Black women, the group with the best LDL figures, have the second-highest rate of heart disease. I thought perhaps the higher rate of heart disease among black women could be explained by a higher rate of smoking, but it turns out the AHA had those figures as well: 20.6% of white women smoke, but only 17.8% of black women.

Still think LDL levels have anything to do with heart disease? Please, print this out and hand it to your doctor if he ever tries to put you on a statin because your LDL is “too high.”

And here’s what really fries my bacon about these numbers: I found them in documents published by the American Heart Association — one of the many organizations still trying to scare us about high LDL. Their own statistics say they’re wrong.

In comparing the cholesterol statistics with the heart-attack data, I did notice a figure that seems to actually matter, either as a cause or a marker: low HDL. Here’s a quote from the press release about the UCLA study:

The study also showed that HDL cholesterol, or “good cholesterol,” levels have dropped in patients hospitalized for heart attack over the past few years, possibly due to increasing rates of obesity, insulin resistance and diabetes. Researchers found that 54.6 percent of patients had HDL levels below 40 mg/dL.

And here’s the magic number from the American Heart Association: American adults with HDL levels below 40 mg/dl: 16.2%.

Now we’re looking at a small fraction of the population that accounts for more than half of the heart attacks.  It should be clear to anyone with a functioning brain that raising HDL is a heck of a lot more important than lowering LDL.

If you’ve seen Fat Head or read any of the books or blogs I recommend, you already know what raises your HDL: eating more natural fat … exactly what the American Heart Association tells us to avoid.

I guess they’re too busy putting their seal of approval on boxes of Cocoa Puffs to actually check the relevant data — including their own.

p.s. — I’m leaving tomorrow for the low-carb cruise.  I’ll check comments when I can, but won’t be writing another post until I get back.


23 thoughts on “The American Heart Association: Do These People Read Their Own Data?

  1. Katy

    Wow! I do wish somebody would inform the famous cardiologist Dr. Mehmet Oz so he’d stop advising people to avoid foods “…full of unhealthy saturated fat that can raise levels of ‘bad’ LDL cholesterol, a risk factor for heart disease and strokes.”

    The Great Oz has heard Gary Taubes speak, so he’s aware of the alternate theories. I’m afraid he’s one of those who built his career on the low-fat diet theory and can’t possibly change course now.

  2. djinn

    Tom, you’re not going to believe this, but I have convinced my doctor about chloresterol madness. And she has stopped prescribing statins. I’ll show this to her and we’ll both have a good chuckle.
    (I said you wouldn’t believe me……. )

    Your doctor is a gem. Most would never be persuaded by a mere mortal … I mean, patient.

  3. Ellen

    Hey, that AHA sells that heart healthy logo for big bucks – For each of the approximately 630 “heart healthy” logos on cereal boxes or other food products, the AHA gets $7500 a year.

    Well actually, the AHA charges companies on a per product basis: $7,500 for 1-9 products, $6,750 for 10-24 products and $5,940 for 25-99 products in their first year. So if you are a Kraft marketing exec, you can really get a good deal on some AHA logos.

    The AHA ain’t gonna jeopardize that kind of cash flow – the truth and the health of American citizens is of no consequence when you’re talking billions in assets at risk.

    Yeah, it’s a shame. And they actually disputed the Lipid Hypothesis until Ancel Keys got himself on the board.

  4. Sizzlechest

    LDL values as given on a standard cholesterol test are terrible indicators for a number of reasons. First, there’s the inaccuracy of the measurement itself. The number is estimated based on the other values and never measured directly. Second, it provides no insight to the “quality” of the LDL particles. The small-dense LDLs are the ones most likely to be oxidized, and a cholesterol test doesn’t distinguish the small from the large. I have seen literature of some new LDL tests, I think Jimmy Moore blogged about them, that give a true picture of your LDLs. I think those LDL tests are a better predictor than even HDL, but I haven’t looked into them since my insurance doesn’t cover those tests.

    That’s correct; LDL is usually calculated, and if you have low triglcyerides, the calculation is often way off. If you’ve got high HDL and low triglycerides, the odds are very good that you’re mostly producing large, fluffy LDL.

  5. Dan

    Why let a little inconvenient data interfere with all those profits from drugs and “cholesterol lowering processed foods”???

    I tried to share the data with my doctor, but he didn’t want to argue. At least he’s not pressuring me to take statins if I don’t want to.

    I guess that’s a start, at least. My dad’s doctor was furious when he stopped taking Lipitor.

  6. Ed Terry

    A year and a half ago, my cardiologist was thrilled that my total cholesterol was 97 (LDL 44)! I accomplished that by only eating fats that came from huge industrial facilities (e.g. soybean, etc). None months later when I went back for a checkup and he wanted to put me on a statin because my LDL had shot up to a staggering 80! When I pointed out that my HDL was now 66, he didn’t even acknowledge that as a potential improvement.

    I pointed out to him that there’s never been a study demonstrating that lowering LDL reduced mortality. He grudgingly admitted that perhaps the slight benefits offered by statins were due to their “pleiotropic” effects, and he was adamant that I go on a statin. By the way, most of the fat I eat was at one time walking around and making noise.

    Later that year, I was impressed by my primary care doctor. She asked me how I raised my HDL to 70 and I told her she wasn’t going to like the answer. After I briefly explained my low-carb approach, she didn’t even blink.

    I don’t think I’d want my total cholesterol so low. Glad your HDL is nice and high now.

  7. Dave, RN

    “LDL values as given on a standard cholesterol test are terrible indicators for a number of reasons”.

    Another reason for that is there are two ways of calculating that number. If your trigs are above 100 you use what most all labs use in the US (Friedewald). I’m guessing that US labs use Friedewald because most americans trigs are above 100. . If it’s below, then you’re supposed to use the “Iranian” method to calculate. But, they are still both just calculations and not actual measurements.

    And yea, a total of 97 is… frightening.

    And neither method tells us diddly about particle size.

  8. Ailu

    Tom, do you offhand know what the test is called that measures the particle size? (I want to be completely prepared & armed when I get my annual next month. lol) Thanks.

    I believe it’s called a VAP test. You might also want to check out the calcium test Dr. Davis recommends:

  9. Josh Goguen

    Since reading this, the numbers have been banging around in my head and I’m just becoming more bewildered. Why this cholesterol hysteria keeps going seems maddening.

    LDL isn’t an indicator of heart attack but should be watched because it causes heart attacks. Cats and dogs living together. MASS HYSTERIA!

    Thank you, Tom. You’ve discovered a way to slowly drive me crazy.

    I’m afraid it keeps going because it’s a multi-billion dollar industry now. I’m not sure what will finally kill it.

  10. Todd

    In Anthony Bourdain’s latest No Reservations episode, titled Obsessed, he closes the show on the topic of how he’s been on Lipitor for a while, and first his doc had him cut out butter, then cheese, and finally pork. Here’s a guy who needs to find out that it’s his carnivore tendencies that have kept him in reasonably good health thus far, and his docs advice is actually going to hurt him more than help. Somebody needs to bring Fat Head to his attention!

    Here’s the clip where he talks about his doc (he starts right at the end of part 4, and continues here in part 5):

    Yee-ikes! Somebody introduce this man to Eric Westman or Mary Vernon! I just watched a lecture this week by Dr. Vernon in which she outlined some key metabolic pathways. It’s sugar and starch that lead to bad cholesterol profiles, not fats.

  11. Ted Hutchinson
    Vitamin D production after UVB exposure depends on baseline vitamin D and total cholesterol but not on skin pigmentation.

    So as you lower your cholesterol level you also lower the capacity of skin to generate vitamin D from cholesterol when exposed to UVB from sunshine.

    Low solar UV-B exposure and/or low 25(OH)D concentrations are associated with an increased risk of CVD. Large nonrandomized studies indicate that CVD mortality is more than twice as high in older individuals with deficient 25(OH)D concentrations compared with those individuals who have adequate 25(OH)D concentrations.

    Good info. I wonder if the drive to lower cholesterol also increases the risk of skin cancer.

  12. donny

    Peter makes the idea of ripping cholesterol out of artery walls and transporting it back to the liver look pretty stupid here. Um, maybe your body makes HDL when you’re healthy– less oxidized ldl in the artery wall because ldl are part and parcel of the freaking repair process. Hdl’s job might be to assist in ldl turnover– so that in case the need for repair does actually come up, the ldl is up to the job.

    And now for a little crazy;

    Yup; I believe high HDL may be more of a marker for good cardiovascular health than an actual cause. So eating lots of good quality fat may be improving our health and merely raising HDL as a by-product, in which case the HDL-raising drugs coming down the pike won’t do much. I’d rather eat the butter and skip the pills.

  13. Goat

    “I’m afraid it keeps going because it’s a multi-billion dollar industry now. I’m not sure what will finally kill it.”

    What will kill it is enough people refusing to care what their LDL levels are, refusing to take the statins, refusing to eat “healthy-whole-grains” as prescribed.

    That is, of course, unless we get socialized medicine, at which point the conventional wisdom will become law. Then it will be “go on a statin or lose your medical care”.

    The current health-care bill will, from what I’ve heard, put the government in charge of prescribing specific treatments. That’s going to be a disaster if it happens.

  14. Wes Hopper

    My wife and I went on low carb about 14 years ago, and I suffered the agonizing fate of losing 55 lbs of ugly fat around my middle. We avoid anything marked “low fat” for the poison it is. I had a complete checkup that included artery scans and ultrasounds last year and my arteries showed no signs of clogging anywhere.
    Kind of surprised my cardiologist because most guys “my age” are clogged somewhere. Thank you, Dr Atkins, RIP.

    Well said. I don’t touch anything marked “low fat” either.

  15. Chris

    About the low fat products. I was a little surprised to see that the low-fat yoghurt that is sold at the supermarket contains as much (and sometimes more) sugar than is found in coca-cola (per 100mL, that is). Both being about 10-11% sugar.

    Yup; my wife likes full-fat yogurt, but it’s hard to find sometimes.

  16. D. Stanek

    When I was in engineering school, we used a term for the eggheads who were wedded to their nutty highbrow ideas. In this case the term applies to the AHA. They are a bunch of “highly educated nitwits.” I’ll take the advice of a well read person with common sense (like you) over a PHD or doctor any day. Is there any hope these people’s paradigms will ever change. Unfortunately physicist Ernest Rutherford I think said it best, “New ideas are accepted only when the older generation of experts dies out.”
    Keep up the good work.

    The paradigm will shift when the people whose careers depend on it are gone.

  17. TallDog

    Does anyone know what percentage of the heart attack victims in the UCLA study had “dangerously high” cholesterol (over 240)?

    I can’t seem to find this anywhere. I’m starting to think they buried the number.

    I didn’t see that figure in the study.

  18. Jean Bush

    A couple of weeks ago, my doctor said my “cholesterol” was slightly elevated, would I like to try a low dose statin? No thanks. He didn’t even tell which cholesterol it was. Idiot.


Leave a Reply

Your email address will not be published.