22 September 2009

Research: how to read obesity reports

I've been reading quite a bit of obesity research recently and I want to share some of my thoughts about how fat people might read such research with a critical eye.

I know the idea of reading material that is intensely fatphobic is not everyone's idea of fun, but I think it is important that we dip in to this stuff from time to time so that we can: keep up with what they are saying about us; develop better research models for fat; develop a critical eye in order to distinguish between research that provides useful information, and research that makes things a lot worse for fat people.

You don't have to read heavy research reports to get a flavour of current obesity research. This is the stuff that also crops up in news report after news report. You know the type of thing, it starts with a sensationalist headline making some kind of preposterous claim about fatness, there's invariably a picture of a headless fatty, some quote from an obesity expert, and the reiteration that being fat is a very bad thing. What I'm going to say below applies to this kind of report as much as it does to the more formal scholarly publication.

Think of what follows as a mental check-list to help you read material that claims to be obesity science, it's like reading a food label to check for dodgy ingredients. Maybe approach this kind of material in the same way that you might do if you were lifting a rock to have a look at the worms and insects wriggling away underneath, all that stuff is interesting to look at but you're really glad that you don't have to live down there.

1. Check the date, is it silly season? This is the time of year when people are likely to be away on holiday and the media is increasingly desperate to find material to fill its dead air. More stories, especially salacious fat panic reports, get through that would otherwise flounder under quality control guidance.

2. Google any experts that are quoted. Find out their interests, especially how they make a living. It is common for such experts to be paid employees or directors of weight loss companies, or organisations directly sponsored by the weight loss industry, such as the International Obesity Task Force, the Association for the Study of Obesity, the National Obesity Forum, and others. Decide for yourself how neutral or trustworthy an expert you think they are. Also, anyone who refers to themselves as an obesity expert is likely to be a bit of a dick, especially if they are not at all fat.

3. Think about how the news story came to be made. Journalists and editors may twist research findings for the sake of an exciting story (I have done this!). Think of the media as a distorting mirror for research, bear in mind that it has its own vernacular and pressures, that it is likely to simplify, reduce and mis-quote complex research findings, or that stories are often cobbled together quickly from a press release without much quality control.

4. Think about why the research is being done. What kind of starting out assumptions does it make about fat people? Does it begin with a paragraph or two about the perils of the obesity epidemic? Does it appear to question such an epidemic? What is it supporting? Do the researchers use Body Mass Index as a measure of health without any critical understanding of it? Do you think BMI is an accurate representation of heath? What does this tell you about the values implicit in the research? Do the research findings support these values?

5. Where are they coming from? Try and imagine how the researchers might answer if you asked them: do you think being fat is a problem? This can help you work out what kind of perspective they are bringing to their research, which is important but not always stated clearly. You could also ask: do you think fatphobia is a problem?

6. Think about what claims are being made by the research in terms of its scientific purity. Is it claiming to present truth or facts? If so, go back and reconsider the perspectives being put forward by the authors. Remember that 'truth' and 'facts' depend on what people think and believe; 'facts' made by the weight loss industry about fatness vary a great deal from 'facts' that I know about my own fat body, for example. Looking at the research findings, what other versions of the truth could be made?

7. Try and find out who is funding the research. Don Kulick writes in Fat Studies in the UK that all research about pet obesity is produced by pet food companies, for example. I know pets are different to humans, but it illustrates how funding can affect the scope of the research and its findings, which then get reported as facts. Sometimes you may have to dig a little for this information.

8. Think about the process by which the researchers got their hands on the funding. Try to imagine what they might have had to say in order to get the money. Might they have had to downplay any interest in fat politics, for example, or play up their support for the treatment and prevention of obesity? You can't know the answer to this for sure, but who gets the funding and why they get it, and what gets left out, is part of the context for obesity research. Also, what happens to researchers who have no funding?

9. How big is the research sample? By sample I mean the people who are being studied. One of the National Health Service Care Pathways for dieticians in the UK is based on research on a group of nine people. Do you think a study of nine people can make conclusive claims about all fat people? No! So size makes a difference in the outcome of the study.

10. What does the sample look like? If it's a sample of fat people, are they suffering from any prior ailments? This affects research claims made about fat people and health. Is there any acknowledgement or accommodation in the research of social influences on health, for example discrimination? How might discrimination or stigma impact on the sample or affect the findings? How representative is the sample of all the rad fatties you know?

11. How are variables defined and interpreted? Variables are the things that the research is studying, for example weight loss, ethnicity, activity. The way the research is set up means that although variables appear to be neutral, the way they are defined and interpreted is not neutral at all. Here's an example: Jane Ogden, a well-respected obesity expert, presented a paper about weight loss surgery at the Size Matters? conference earlier this year. She defined 'success' as someone who had lost weight after surgery. This means that cases could be defined as 'successful' where the person who had had surgery was suffering terrible surgery-induced health problems, as long as they had lost weight. That doesn't sound like a 'successful' surgery to me, quite the opposite.

12. Have a look at the source material cited in meta-studies about obesity. Such big studies are basically studies of studies, and they sometimes make pompous claims about being very reliable. But if they are based on source material that is not particularly reliable, for any of the reasons I've mentioned here, then their reliability too is questionable. It's also a good idea to see what meta-studies include and exclude, for example do they include material that is critical of taken-for-granted claims about fat? If they don't then they're missing out a lot of important stuff.

13. Ask to see the original data and report, if you can.

14. Think about where the research has been published. Peer-reviewed publications are seen as the gold standard for reliable research, but there have been reports recently about fake journals, people being paid to put their names to dodgy research, and in-house publishers owned by the businesses benefiting from the research. Do some homework and decide on the reliability for yourself.

15. Become a fan of Bad Science and make sure you read this post.

16. Make time for self care after immersing yourself in the strange world of obesity research. Blog or share your findings, do something fun to get any residual fatphobia out of your system. Keep breathing.

Edited to add: I forgot to mention a few more things...

Health. Most obesity research is about fat and health because this is the agenda that most interests upholders of fat panic. Much of my comments here refer to health research. The fact that, aside from researching weight loss, other kinds of obesity research are sidelined also says a lot about what gets funded and what does not, and what is deemed important. If I was the boss of all research funds I would fund a far broader range of stuff, it would be interesting and useful, for example, to know more about the effects of fatphobia on people of all sizes.

Sampling strategies. How researchers find samples also affects the research outcomes. There are books about this, go and have a look at one if you can tolerate this level of geekiness. What I will also say, however, is that the sample is really important, so check for possible bias in it. For example, a study about people's attitudes to fatness based on a sample of fat women who go to Weight Watchers is going to have a different outcome to a study of fat women who go to NOLOSE.

Stats. There's some stuff I could say about statistical maths too, which I won't because I barely understand it myself. Suffice to say that there are different ways of manipulating statistical/quantitative data to provide different research outcomes.

One final thing, a really important thing. Studies may find a correlation, or a relationship, between a number of variables. So a study could find that there's a relationship between fatness and unhappiness, for example. But this doesn't mean that being fat necessarily makes you unhappy. A statistical relationship is just that, not a cause or an explanation.


wonderful woman said...

"if you were lifting a rock to have a look at the worms and insects wriggling away underneath, all that stuff is interesting to look at but you're really glad that you don't have to live down there."

I love this. And this post is *so* incredibly useful. Thank you! I have avoided this kind of crud as much as I can but agree - we have to know what is being said about us to fight it. And we have to know why.

This should be in FSUK Vol 2... if anyone ever produces that!


Charlotte Cooper said...

Thanks WW. I added a few more things at the bottom.

Don't be afraid to look at this stuff, it's interesting in a gruesome way.

Jennifer said...

Just a quick comment to say I was liking Bad Science until I ran across some totally unnecessary fat bashing in this post:


Charlotte Cooper said...

Jennifer, oh that's a shame. He's not particularly fat-friendly, I know, but his column is great fro debunking bad science and clearly there's an overlap with fat stuff, which is full of bad science. Oh dear.

Jerome said...

This is a FANTASTIC post. Thanks for posting, I think people need to be reminded constantly to be careful about taking reports regarding research studies at face value, since a lot of them are straight-up bullshit and/or funded by [insert industry here]. That being said, I generally assume any research with a fat-phobic conclusion to be bullshit automatically...