Our favorite* NYT reporter, Gina Kolata, is back at it, this time with a write up presenting findings of a meta-analysis entitled “Adverse Metabolic Response to Regular Exercise: Is It a Rare or Common Occurrence?” The study’s authors looked at insulin, blood pressure, triglyceride, and HDL levels to define an ‘adverse metabolic response.’ They found that 8.4% of ‘exercisers’** had their insulin numbers get worse; 12.2% their blood pressure; 10.4% their trigs; and 13.3% their HDL. And about 7% experienced worsening of two of those four.
One way to look at the study is that 93% of the time, three of the four categories improved or stayed the same. From this perspective, I would answer the researchers’ question by noting that adverse metabolic response to regular exercise is, indeed, rare. I would also title a write-up of the study something other than “For Some, Exercise May Increase Heart Risk.” And I would certainly refrain from opening that write-up with this:
Could exercise actually be bad for some healthy people? A well-known group of researchers, including one who helped write the scientific paper justifying national guidelines that promote exercise for all, say the answer may be a qualified yes.
Reasonable minds can disagree on what ‘some’ means; that’s all fluff anyway. There are still two key questions about the study that Ms. Kolata leaves unanswered and unexplored: (1) What was the exercise? and (2) Do the health markers studied actually predict heart disease? After all, the study doesn’t matter without answers to those questions – talking about ‘exercise’ without defining it is as vacuous as talking about a ‘healthy diet’ without defining health objectives, or using the phrase ‘hook up’ to describe a romantic encounter without detailing the particulars…
That is unless you blindly accept the government’s definition of exercise, or whatever type of exercise is routinely used in studies like those analyzed. Is exercise a brisk walk? Is it Cross-Fit? Is it a hike? Is it a leisurely bike ride? Is it 10 sets of 10 on the bench press Monday, Wednesday, and Friday [I just described the routine of 78% of 15-25 year-old American males or, for those living in Pacific Beach or Orlando, about 89% of men between 15 and 38]? The meta-analysis itself doesn’t answer the question either, instead just referring the reader to the six underlying studies. But it does say that all basically involved ‘endurance’ exercise.
I read somewhere (whoever you are, thank you for the insight, assuming it’s accurate, and if it’s not, oh well) that researchers always study the effects of exercise through endurance training because it’s easy to measure. Something like VO2 max can be calculated and compared, as can be the minutes that a subject pedals on a bike. It’s much more difficult to measure the activity of someone who is walking outside, or lifting weights, or playing soccer, or doing any of the thousands of other activities that most of us would consider ‘exercise.’
There are two points to take away from this. First, whatever the findings of the meta-analysis are, they only apply to the type of activity looked at in the underlying studies. The findings do not apply to any other activity.
In other words, Ms. Kolata presents a study about a narrow range of activity and writes it up for the New York Times as if it applies to all exercise. That’s bad journalism. It’s not just misleading; it’s false. Resistance training is exercise, but its effects weren’t studied. Swimming 25-yard sprints is exercise, but it also wasn’t studied. Another rigorous, typically late-night, activity was also left out—and I’m certain that its non-STD-related health effects are 100% positive.
The other point is that the study and write-up disguise the interesting finding that endurance exercise could have bad effects. The paleo community — led by Mark Sissons’ admonition against chronic cardio — has long been questioning the benefits of steady-state ‘endurance’ exercise. The U.S. first went through an aerobics epidemic; now it’s going through a marathon one (thanks Biggest Loser!). Although someone like Dr. Doug McGuff might disagree, I’m guessing that jogging is one activity that most Americans consider ‘exercise.’ But jogging feels nothing like sprinting, or lifting weights, or a Cross-Fit met-con session, nor is there any reason to assume the body responds to these activities the same way. Maybe some of these activities do not trigger the same response meta-analysis revealed about ‘endurance’ exercise. But the way the researchers wrote the article, and they way Ms. Kolata reported it, you’d never know.***
The other elephant is whether the biomarkers studied (insulin, blood pressure, trigs, & HDL) even matter for heart disease, or for other diseases. Ms. Kolata briefly touched on the issue:
Some critics have noted that there is no indication that those who had what Dr. Bouchard is calling an adverse response to exercise actually had more heart attacks or other bad health outcomes. But Dr. Bouchard said if people wanted to use changes in risk factors to infer that those who exercise are healthier, they could not then turn around and say there is no evidence of harm when the risk factor changes go in the wrong direction.
Right. So the question remains: Are what we now consider risk factors really risk factors, and if so, what are their relative weights? That’s for another post, which will be consist of little but links to books and blogs.
*I use ‘favorite’ here not to mean “I love her work; everyone should read her” (the way I would call AJ Liebling, DFW, & Joseph Epstein some of my favorite essayists), but in the sense of “I always read her articles because they make me furious” — a sort of Google algorithm of favorite.
**More on what exercise meant in the study later on. Those quotes aren’t there for nothing.
***The ‘what the f*** is exercise’ issue reminds me of the ‘what the f*** is fat’ issue that we encountered in HBO’s “Weight of the Nation.” That, too, will be another post.