We at the Fit is a Feminist Issue news desk are committed to bringing you up-to-the-minute news, commentary and perspective on a variety of topics– no matter the day, no matter the hour.
So when co-editor-in-chief Samantha FB messaged me to get on this story, I got straight to it (after eating my dinner, that is; you can’t do good investigative journalism on an empty stomach).
The story is– what are we to make of the mainstream news stories claiming that “Fat but fit is a big fat myth”? This was the headline of a BBC article, out today. What’s the deal?
Here’s the upshot: At this year’s European conference on obesity, researchers from the University of Birmingham gave a paper suggesting that the notion that people could be obese, metabolically healthy, and therefore not at increased risk of heart disease and diabetes is false.
This reporter will dig into more detail for Sunday’s blog post, but for now, I’ll tell you what the news article says, and then what questions I have (as an academic who researches and writes about this medical and scientific literature).
First, here’s what an article by the Guardian has to say about the new study (which is not even in article form, much less submitted, much less reviewed, much less accepted for publication, much less published):
Several studies in the past have suggested that the idea of “metabolically healthy” obese individuals is an illusion, but they have been smaller than this one. The new study, from the University of Birmingham, involved 3.5 million people, approximately 61,000 of whom developed coronary heart disease…
The scientists examined electronic health records from 1995 to 2015 in the Health Improvement Network – a large UK general practice database. They found records for 3.5 million people who were free of coronary heart disease at the starting point of the study and divided them into groups according to their BMI and whether they had diabetes, high blood pressure [hypertension], and abnormal blood fats [hyperlipidemia], which are all classed as metabolic abnormalities. Anyone who had none of those was classed as “metabolically healthy obese”.
The study found that those obese individuals who appeared healthy in fact had a 50% higher risk of coronary heart disease than people who were of normal weight. They had a 7% increased risk of cerebrovascular disease – problems affecting the blood supply to the brain – which can cause a stroke, and double the risk of heart failure.
The article then goes on to foment panic among health care providers and consumers. Well, I think the tone is a little frantic. See what you think:
Susannah Brown, senior scientist at World Cancer Research Fund, said the study’s finding, “emphasise the urgent need to take the obesity epidemic seriously.
“As well as increasing your risk of cardiovascular disease, being overweight or obese can increase your risk of 11 common cancers, including prostate and liver. If everyone were a healthy weight, around 25,000 cases of cancer could be prevented in the UK each year.”
Right. Now, let’s all take a deep breath. As some of you know, I’ve posted often about how real science is complicated, so we should not take sensationalist news headlines as telling the truth about new, or controversial, or counter-intuitive, or nuanced research results. So here are some questions I have:
When the researchers talk about increased risk that people with BMIs over 30 have for various illnesses, are they showing statistically significant increases or clinically significant increases? What I mean here is this: a researcher can find a shift in risk that ends up being irrelevant to the real-life clinical likelihood of developing particular conditions (this is complicated but important). We don’t know, as there is no paper yet.
What sorts of risk profiles did the researchers find for people with BMIs under 25? under 20? When one peers at the fine print in the data tables in medical and epidemiology papers (as I am wont to do), one finds interesting and potentially reportable risks for folks who are co-called underweight– BMI under 18.5. Are there increased risks in those groups? Are they comparable to the risks in the BMI> 25 and >30 groups? We don’t know.
What about the same BMI in different age groups? How do those risks vary over the life trajectory? It turns out that at various life stages, different BMI groups have very different risk profiles for medical conditions and death due to medical conditions. For instance, men in their 50s with BMIs <18.5 have a pretty high all-cause mortality risk (generally from cancer). We don’t know anything about this yet.
Then there’s how they defined their terms. Sounds pedantic (and yeah, it is, but this is my job), but it’s important to know very precisely how the researchers defined metabolically healthy in terms of blood pressure, cholesterol, blood sugar (e.g. Hemoglobin A1c), and then what clinical end points (that is, records of diseases people in the study ended up having) or surrogate end points (e.g. blood pressure, cholesterol, blood sugar) they used.
And last (until Sunday; consider yourself warned): it seems to me (and lots of scientists agree) that the jury is out, sooooo out, on a clear understanding of the ways in which both genetic and environmental determinants of human metabolic processes contribute to body weight change and maintenance. Topics currently being investigated include:
- the role of body fat
- the role of body shape
- the role of abdominal fat
- the role of visceral vs. subcutaneous fat
- how visceral fat and trigliceride levels interact
- the role of body weight variation over the life trajectory
- what science and medicine can do about the body weight of the population (since so far medicine has come up with nothing effective other than gastric bypass, which has its own problem)
- And much much more.
One last comment: the standard view in medicine and medical research is that having a BMI over 25 is bad for people, and a BMI over 30 is much worse (and don’t even get me started on over 35, over 40, etc.) Studies and articles that have come out challenging that assumption have been pilloried by a lot of public health and medical experts. But, like any scientific paradigm, there’s research on the edges, and sometimes that research gives rise to a new paradigm. I don’t know if we’re on the way to a new paradigm, but I know that the current paradigm has left a lot of important and foundational questions unanswered. Of course, that doesn’t make for good headlines.
A 50% increase in the probability of a common disease (coronary heart disease) would certainly seem clinically significant, although I’d like to see the confidence intervals around that number and the details of how the statistics were done.
Yes, I agree with you on all counts. a 50% increase in CVD risk is very troubling. But it gets distributed among the population differentially, so I want to know which subgroups are at most risk. And yes, knowing the confidence intervals is very important. For me, the big question is WHY. We really don’t have a clear view of the mechanisms, which helps explain why we have so little control over trying to change body weight and then maintain it. Thanks for reading and commenting.