Case study of a study: Getting fat but staying fit

So, you may have noticed that from time to time, you will see headlines to the effect of, “Can you be fit but fat?!” [“Yes!!!” or “No!!!” depending on which study or whatever prompted the headline.] Gretchen Reynolds has noticed it, too. And so have some researchers who decided to conduct a study to try to shed some more light on the issue. (“OBJECTIVES: This study sought [to] examine the independent and combined associations of changes in fitness and fatness with the subsequent incidence of the cardiovascular disease (CVD) risk factors of hypertension, metabolic syndrome, and hypercholesterolemia.”)

It’s easy, with a study like this, to think the study is saying more than it is. I think part of this is just human nature, to want to connect things to the big picture. I think some of it could be that narratives around obesity are so entrenched, we’re like, “I know how this song goes!” and our minds sort of take off on their own. Let’s look at some of the things it’s not saying.

You might think, with the objectives listed and Reynolds’ framing, that it will tell us about how being fit interacts with being fat (having a high BMI) in affecting people’s health. (I did at first.) Not quite. It looks at how weight/fat gain (vs. weight/fat maintenance) interacts with changes in fitness level in affecting three risk factors for cardiovascular disease (CVD). So it actually tells us little about how <i>starting out</i> at particular BMIs or fitness levels affects people’s health, just how changes in BMI and body fat percentage or fitness levels affect people’s risk factors for CVD, and whether one changes how the other affects risk factors for CVD.

It’s likely that they restricted the BMIs of the subjects to a certain range so that they could just zoom in on how changes in weight affected them, without any noise from possible differences in BMI. (I can’t tell for sure based on just the summary.) For example, it could be that weight gain in underweight people doesn’t increase the risk of developing CVD risk factors, but in overweight people, it does. Or it may not be the case, and maybe gaining weight after you’re 21 or so is equally bad for you no matter what your BMI. We don’t know, at least not based on this study, because that’s not what the study was about.

Here’s another thing it doesn’t tell us: how likely these people are to die. Notice I just said, “the risk of developing CVD risk factors”? Not that that makes it a bad study. Linda Bacon’s original HAES study also looked at risk factors, not mortality. If you want to find out, “How does being fit and/or being fat affect mortality?” a study that measured actual mortality would be the gold standard. It would also take longer and be more expensive. If you want to look at health, not just mortality, then actual diseases would be the gold standard. Hypertension, metabolic syndrome, and hypercholesterolemia are risk factors for CVD, but they don’t cause death or poor quality of life by themselves*. (Metabolic syndrome is a group of risk factors.) So the study is looking for risk factors for one particular disease (CVD) that can cause death, not actual cases of that disease, or deaths from that disease, or deaths from other diseases, or total number of deaths.

So if you started out wondering how this fits in with, say, Flegal’s study of BMI and mortality, keep in mind that A) it is looking at changes in BMI, not absolute BMI, so you can’t really compare them, and B) even if it was looking at absolute BMI, it’s looking at risk factors for CVD disease, not death from all causes, so people with BMIs >25 could be less likely to develop CVD when risk factors are present than people with BMIs <25 (I think I’ve seen some evidence of this); BMI >25 could increase of survival of CVD (ditto evidence), and fatness could increase the risk of some diseases but reduces the risk of others (ditto evidence).

Diabetes also may make people more likely to gain weight, and diabetes is a risk factor for hypertension. The summary of the study says that it studied healthy adults, so presumably none of them had diabetes. I’m not sure if that would also exclude people with prediabetes, and I’m not sure if prediabetes is also a risk factor for hypertension, but if so that would call into question the assumed causality of the weight gain causing the hypertension. There may be other factors like this that would also follow a “C causes A + B” causality, that we don’t know about.

Let’s also look at who wrote the article. Gretchen Reynolds often writes for the Well blog. Based on her articles that I’ve read, she’s neither an anti-obesity cheerleader nor an outspoken skeptic like Gina Kolata. If the article were written by a “cheerleader”, I’d be skeptical…okay, more skeptical, and if it were written by Gina Kolata (a lone wolf, in DebraSY’s terminology), other people would be skeptical, and I couldn’t exactly blame them, because she has basically made her reputation on being a lone wolf. She has books to sell, just like Gary Taubes, even if I do find her take more convincing than Taubes. But being right doesn’t make you immune to assumptions.

Let’s also do a quick google of Duck-Chul Lee, the leader of the study. Here’s a paper by Lee advocating that doctors look more at cardiorespiratory fitness as a risk factor, and here’s a similar one. Oh, and here’s a study Lee did, very similar to the study we’re looking at, that looked at actual mortality instead of risk factors and found that fitness was an important predictor, but change in BMI was not. Overall, it looks like his primary interest is in finding the benefits of exercise.

Most importantly, don’t take away from this study, “Well, there have been a bunch of other studies on this, BUT THIS ONE SETTLES IT.” This is yet another study. It basically adds another data point; it’s not necessarily more definitive than the other ones. It is probably more useful in telling you what to do than studies that look at absolute BMI, because it looks at how changes from a baseline affect risk factors for CVD; although stuff like “whether you gain weight in middle age” and “how well you respond to exercise” can have a basis in genetics, they’re both more easily influenced by your actions than your absolute BMI or fitness level.

So, after all that explanation of what the study didn’t say, what did it say? According to the summary of the study, “Maintaining or improving fitness was associated with lower risk of developing each outcome, whereas increasing fatness was associated with higher risk of developing each outcome, after adjusting for possible confounders and fatness or fitness for each other… In the joint analyses, the increased risks associated with fat gain appeared to be attenuated, although not completely eliminated, when fitness was maintained or improved. In addition, the increased risks associated with fitness loss were also somewhat attenuated when fatness was reduced.”

Based on the wording here, it sounds like staying fit decreased the risks of gaining fat, and reducing fat decreased the risks of losing fitness, but the first effect was stronger than the second. Reynolds’ article also notes that most people did gain fat, and few lost fat.

A quote from the end of Reynolds’ article:

“The message is simple,” Dr. Lee concludes. “So much attention gets focused on weight reduction, but reducing body fat is very difficult for most people. Our study suggests that,” in terms of heart health, “maintaining your fitness over your lifetime is just as important, and for most people is probably more achievable.”

OK, good that they’re recognizing that, but given their field, I would think they’d know that maintenance, not reduction, is what’s really hard. Not that losing weight doesn’t take effort, but most people who attempt it will succeed. The same is not true for weight maintenance.

On further reflection, it’s quite possible that this is just an example of the same kind of slip that I keep making and correcting throughout this post. I keep writing as though they studied absolute BMI and not changes in BMI, even though I know that’s not what they studied. I’ve noticed another frequent mistake that people make is to describe weight maintenance as weight loss–I remember that that was all over the comments in a NYT article about people who ate a weight-increasing diet, and the ones told to exercise before eating anything gained less weight than the ones told to exercise at a different time, but people kept saying they’d lost more weight.

What’s the takeaway from this study? To me, it’s that fitness is more important than weight gain. If keeping active and following HAES keeps my weight stable even when I hit menopause, great; if not, I can reduce the risks from it by keeping fit. And the risks are hardly a death sentence, so don’t worry about it too much. Since dieting is likely to leave me at a slightly higher weight, that’s probably not a good idea. And even if I magically became thin tomorrow, it would not be prudent to quit exercising.

*I consider “by themselves” to also mean that I’m excluding things like “eating less salt” or “how people treat me because I have a big belly”–things not directly caused by high blood pressure or metabolic syndrome.

This entry was posted in Uncategorized. Bookmark the permalink.

One Response to Case study of a study: Getting fat but staying fit

  1. One other thing I meant to mention: notice how the picture that accompanies Reynolds’ article is of a smiling, exercising inbetweenie/small fat (who has a body much like mine) instead of a headless slumping/eating/holding their fat supersize person? I wonder how much of that is due to the media starting to listen to criticism of the “headless fatty” and how much of it is because of the subject matter, and because the title contains the phrase “staying fit”.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s