Overweight people have lower death risk…#yesterdaysnewstoday

I must be getting old. Things are beginning to repeat themselves. It feels like déjà vu all over again. In fact, even that play on the expression ‘déjà vu’ seems recycled. I know that it actually is, by the way. Things are beginning to repeat themselves, you see. I must be getting old.

It seems like just yesterday (last November maybe. Or last July. Or back in 2009. Or even 2005. Or was it 1995? Or how about waaaaaay back in 1990?) when I was reading about how being a bit overweight might be associated with having a longer life expectancy. Not obese, mind. Just a bit overweight. The medical literature often talks about “not being lean” or somesuch euphemism.

Anyway, the medical consensus for decades seems to have been that being underweight and being significantly obese are both definitely bad for your health. Unsurprisingly, the empirical data have shown for years that both are associated with increased mortality. What happens in between these categories is disputed. But lots of studies have suggested that carrying a small bit of weight can be important in helping you withstand illnesses (especially those that result in weight loss) and so — statistically — can give you a slightly lower death risk compared to skinny people.

Whatevs. My main point is that it’s an old idea. Well old.

So imagine my surprise to see that tweet — the one at the top of this post — circulated last night by the UK’s posh broadsheet, The Independent, trailing their next-day (i.e., today’s) front page story. “Recipe for a long life“, ran the headline, “Overweight people have lower death risk“.

In response to that headline, and without seeing the actual story, I myself tweeted the following observation:

And then, being a bit moody about it all (it was after midnight), I single-tweeted three bold predictions:

When I got to see the actual story-behind-the-headline this morning, I saw that it is a well prepared and balanced account of a new research review paper published in JAMA, the journal of the American Medical Association. Clearly, the Indy have put a lot of work into this, and they explicitly flag the debates around the issue and the fact that similar findings have appeared frequently before. Nonetheless, I found that each of my Twitter predictions were pretty much correct.

Firstly, the story well and truly was about people who carry a little bit of extra weight and not to people who are seriously obese:

…carrying a few extra pounds may actually reduce the risk of premature death…the new study shows that people who are modestly overweight have a 6 per cent lower rate of premature death…only the severely obese, with a body mass index above 35, have a significantly increased mortality…

Secondly, the study does indeed relate to mortality (i.e., death) and not to morbidity (i.e., poor health). Being fat is still bad for your health, and still causes diseases that damage your quality of life while costing the economy millions in healthcare and lost productivity. It just won’t actually kill you any time soon:

…the research should not be taken to mean that there were no negative health implications associated with being overweight or obese… Mortality [the death rate] is one thing but morbidity [the disease rate] is another…

Even my prediction about statistical vagueness (which, if I’m honest, I thought was a bit risky) was borne out to some extent. One of the key findings behind the whole headline was…wait for it…not statistically significant:

Mild obesity (those with a BMI between 30 and 34.9) brings a 5 per cent lower premature death rate, according to the study. Although this was not statistically significant, it suggests there is no increased risk of premature death attached to that weight range…

Incidentally, not only was the finding not statistically significant, but the very fact that this is so does NOT “suggest” that conclusion offered by the Indy. After all, absence of evidence is not evidence of absence and so on…

By way of an overall observation, I also tweeted this:

You see, lots of news stories are prioritised on the basis of fitting in with the prevailing mood or interests of the readership. Like when stories about global warming appear more frequently in the media when the temperature outside is hot. In this case it’s a stale-tasting headline about weight issues just when folks are grappling with the post-festive binge and their society’s proclivity for establishing New Year’s resolutions. In fairness to the Indy, they more or less admit this themselves:

It sounds too good to be true, coming at the end of the season of excess…The news will seem heaven sent to those contemplating a new year diet…

I don’t think I’m clairvoyant. I’ve just become a bit accustomed to the way scientific news (especially health research) is reported in mainstream media. It all seems to be the same-old same-old, with minor updates and lots of margin-for-error stuff, pitched during suitable gaps in the actual world-news schedule. And if readers can end up basically predicting the news, what’s the point of news-reporting?

Finally, with glorious profundity, I also tweeted this:

Sorry about that. But then that’s Twitter for you. I guess my point was that this isn’t actually news (in that it’s not truly ‘new’), while there probably was lots of other stuff going on in the world that was.

As I say. It feels like déjà vu all over again.



Categories: Academic publishing, Authority appeals, Dieting, Food, Health, Mathematics, Statistics, The Independent, Twitter

3 replies

  1. I haven’t read the study but from what you’ve said here, it seems the Independent headline is accurate. ‘Overweight’ has a medical definition of a BMI between 25 and 30, and this study seems to confirm various other studies which have found a reduction in mortality at this level. It’s interesting that there was no increase in mortality even at the 30-35 level and unsurprising that there is an increase at the ‘severely obese’ level.

    I wouldn’t say this necessarily warrants front page headlines, but since countless news stories throughout the year lead us to think that any weight above the arbitrary 25 BMI threshold is hazardous to health, I have no complaints. The fact that there have been numerous similar studies in the past just means it’s more likely to be true.

    • Many thanks for your comment. I can’t say I disagree with your assessment. The headline itself is indeed technically accurate — although it does rest heavily on the reader’s ability to appreciate the technical meaning of the term ‘overweight’ (I’m guessing that some readers will, at first glance, take it to include *all* above-average BMI ranges).

      My main reservation was with this being plastered all over the front page in something of a ‘War Begins!’ fashion. The research is empirically valid and represents an advance, but it’s an incremental advance on an old story. In my estimation, it is not really ‘front-page’ news in the classic sense. Therefore, carrying it on the front page produces the risk of reader confusion: readers may form the (erroneous) impression that something dramatic and unexpected has been revealed in this research.

      But it is certainly important for readers to understand the nuanced relationship between BMI and mortality (and morbidity), given the equally simplistic counter-message that is often presented. Perhaps this headline evens things out a bit.

  2. Survival models looking at mortality rates of relatively healthy humans aren’t really that interesting. So it’s common for rehashed material to come out. Given that at this time of the year the “fitness” and “health” resolutions come out and gym memberships are sky high, it’s sensible timing more than anything!

»So what do you make of this?

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

WordPress.com Logo

You are commenting using your WordPress.com 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

Follow

Get every new post delivered to your Inbox.

Join 1,053 other followers

%d bloggers like this: