Not necessarily, of course.
Like the rest of the human race, I am always in search of better health. I am an intrepid explorer of the findings of scientists, digging into the abstracts. What were the actual findings? Who was the study based on? How did they know? Was it correlation or just causation? It’s disappointing how often it turns out to be hogwash.
Today, it was the push-ups article in the NY Times. Last week, the cheese. I can’t tell, once I read enough to discover, Aha! I knew it! whether I should feel smug or irritated. Should I blame the scientists or the journalists? Or myself, for continuing to search for the easy fix and the fountain of youth? Could I solve it by combining them, say, to get push-up cheese?
The pattern of scientific study recaps is fairly standard. Headline: Do This! Because a recent study says so. The photo is vaguely related, usually exaggerated. In a reputable paper, the digest of findings is somewhat specific, although it may blur some rather key details. If it’s not a reputable paper, the digest is
plagiarized summarized from somebody else’s write-up, with most of the key details omitted or exaggerated. Sometimes, scientists are quoted trying to explain causality, though that’s really guesswork given the nature of studies which can’t control for variables enough to make that connection. Never mind! At the end, there’s a snappy quip, often a nonsequitur. If you read the online comments (but don’t!), people responding seem to completely miss the point. Perhaps it’s just as well.
If, like me, you are a searcher of fact facts, you might click through to the actual findings wherein you read all the further limitations, spot the exaggerations, and sometimes even learn that isn’t what the study said at all. Often, the abstract is all you can see because the actual study requires you to pay, so you’re left wondering if you should do anything as a result of what you read. Oh, and did I mention that in the middle of trying to read something which might save your life, you get continuously interrupted by other nonsense, “headlines” for absurd fake stories or music videos for products linked to whatever “product” the “how to live longer” article might ultimate sell?
MOM GETS NEW TONGUE MADE FROM ARM AFTER CANCER DIAGNOSIS…WOMAN WHO DIDN’T KNOW SHE WAS PREGNANT GIVES BIRTH WHILE IN COMA, DISCOVERS SHE HAS 2 UTERUSES…
–from the February 2019 FoxNews push-up story
Mo’ Push-ups, Mo’ Firefighters
The gist of the recently reported push-up study is that being able to do a lot of push-ups is a better indicator of future cardiovascular health than being able to do very few. A lot is over 40. Few was under 10. So far, this is not surprising. Really, without digging much into the “science,” I can probably guess that a person who is capable of doing a lot of exercise will be healthier than a person not capable of doing much.
But here’s where, if you dig, you get both interested and ticked off. Because you know where the Action Plan is headed, right? Everybody should do a lot of push-ups, definitely at least 41. The ad for personal trainers will start popping up in the middle of the articles. The messaging in the photo is not so subtle. Where’s this young healthy fella? In the snow! On gravel!
Before you head outside to the running path in February, you should be aware that this study published in the Journal of the American Medical Association followed firefighters. Male firefighters between 30 and 50 (roughly). In Indiana. This is the part where I always have to compare–is a middle-aged Indiana firefighter enough like me for me to change my behavior?
The study absolutely positively (in the Times and FoxNews versions at least) concludes that if you can’t do at least 10 push-ups, you will have a heart attack sooner than if you can do over 40. Here’s the data from the study which concludes this:
I noticed a few things in the characteristics of this study group. The people who could do the 40+ push-ups had an average age of 35 (avg BMI 27), while the people who could do less than 10 averaged 48 (avg BMI 33). The statistical significance in the results above was calculated based on 37 incidents out of these thousands of firefighters. Now, I have a statistics background, and I am a believer in the magic of statistical significance.
Still, my stomach still churns when I read that conclusions are drawn, in part, by the difference between 1 heart incident for the 40+ push-up group vs. 8 for the under 10 group, given that they started out at different ages and BMIs. The study included age, but it didn’t “control” for age; you can’t sub-stratify when there’s only 37 cases. The older, fatter guys–that’s not pejorative, I’m older and fatter myself–were more likely to have heart attacks than their younger, skinner counterparts.
It gets even murkier. Look at the sub-group who did 31-40 push-ups. They actually had an uptick in heart disease incidences over those who did only 21-30 or 11-20 push-ups. Maybe you need to do just 21 push-ups, not over 40. The actual goal of the researchers was to find out how predictive treadmill tests were of heart disease. What the scientists concluded was not that a lot of push-ups is better than just a few, but that a simple push-up test might be an equally good indicator of future heart disease compared with an expensive complicated treadmill test. Don’t start doing a lot more push-ups just yet.
Eat More Fondue!
There’s also been a recent spate of reports out on how we should all be eating more cheese. For example, here’s the first part of from a Newsweek review of a long-term international study, which clearly “contradict[s] dietary recommendations.” Take that, you dumb heart disease specialists who have been telling us to reduce fat intake! You’re wrong!
After analyzing the diets of more than 130,000 people in almost two dozen countries, scientists found that eating the equivalent of one serving (244 grams, or 8.6 ounces) of full-fat milk or yogurt, a 15 gram (0.6 ounce) slice of cheese or a teaspoon of butter could benefit health.–Newsweek, 9/11/18
What’s the photo that accompanies the article? What would you estimate is the amount of butter and cheese included in this particular sandwich? Looks great, doesn’t it, perfect for a cold winter’s day! Let’s add a cup of tomato bisque to that! (If you’re going to have a mound of melted fat, you might as well add a whole bunch of sodium with cream in it. )
I’m such a troublemaker–would you like to know what the Lancet article actually says?
Higher intake (>1 serving vs no intake) of milk (HR 0·90, 95% CI 0·82–0·99; p trend=0·0529) and yogurt (0·86, 0·75–0·99; p trend=0·0051) was associated with lower risk of the composite outcome, whereas cheese intake was not significantly associated with the composite outcome (0·88, 0·76–1·02; p trend=0·1399). Butter intake was low and was not significantly associated with clinical outcomes (HR 1·09, 95% CI 0·90–1·33; p trend=0·4113).–Lancet, 11/24/18
I’ll translate. Consuming at least 2 servings of milk or yogurt daily with some fat in it led to less heart disease than consuming no dairy. Eating cheese and butter wasn’t associated with anything, although cheese and butter are classified as dairy products. That part of the Newsweek article, and the accompanying picture, weren’t something you could conclude from this study.
Meanwhile, another article in the Times pointed out that South Asians (India, Pakistan, Sri Lanka, etc) have a four times higher risk of heart disease. This is true even when controlling for body composition; even skinny South Asians are more prone to cardiovascular problems. Did the Lancet study control for genetic predisposition? Not that I could tell. Many of the low dairy group could have been South Asians, so that their heart disease could have been driven by entirely different factors.
Need I go on? One study about eating more cheese looked only at Finns. How much is my diet like that of the average Finn? Another study was based on a population 65 years or older. This one was conducted in Poland. And I’m not even able to dig into which ones are funded by the dairy industry. But more than one of the magazine write-ups have ended the deification of the cheese industry by pointing out that, after Brexit, fondue sales are way up. Say what?
Steve Martin and Michael Pollan Were Probably Right
I do believe in science, and I think we should fund scientific studies. Maybe I just shouldn’t be so thin-skinned. After all, if I didn’t want the magic formula myself–please tell me exactly how much and what exercise to do and exactly what to eat–I wouldn’t get disappointed. If you want to oversimplify the results of a study, you’re going to lead to conclusions the data didn’t suggest. If you dig into the details, you’ll find they aren’t the magic formula. You end up where Steve Martin did:
It’s so hard to believe in anything anymore. I mean, it’s like, religion, you really can’t take it seriously, because it seems so mythological, it seems so arbitrary…but, on the other hand, science is just pure empiricism, and by virtue of its method, it excludes metaphysics. I guess I wouldn’t believe in anything anymore if it weren’t for my lucky astrology mood watch.–Steve Martin
As for diet and exercise, if you look across all studies, all data, all recommendations, they do aggregate to much the same. It’s what Michael Pollan wrote before he kept writing the same book over and over again (he’s now apparently into psychedelics). At least, I have found it true for me, which is where scientific studies really ought to be conducted. My body has found that, for me, this was probably right:
Eat Food. Not Too Much. Mostly Plants.
I’ll add the corollary:
Exercise. Every Day if You Can. Not Too Much.
Really, you don’t need the 130,000 people summarized by the Lancet. Use yourself as the scientific experiment. You control for every factor that is irrelevant to you, and you can do longitudinal research on yourself. I know for a fact, for example, that eating large quantities of pizza makes me fatter. Excess of fat? Sodium? Carbohydrates? Take your pick. Going to Zumba class every Friday makes me feel better. Because it’s aerobic >20 minutes? Increased blood vessels? Endorphins? Science can’t tell exactly why it helps me, in particular.
Forget the cheese and the push-ups. Just dance.