Thursday 28 July 2011

Statistical dehumanisation

[trigger warning]

Dorset, Leeds, and now Nottinghamshire. Yet again, applying a century-old statistical population measure to individuals gives an utterly meaningless result.

I've already said what I want to say about the science behind this, so I'll just link to that and summarise below. I still have more to say about the attitudes, though.

[Director of Public Health Chris Kenny] said: "Obesity is on the rise in this country and what this weight management programme is designed to do is to raise the issue with parents.

"I wouldn't want this one particular case to undermine the whole programme."

Nor, I suppose, would he want the fact that the "science" he's using is not actually supported by the evidence to undermine his programme.

Or the fact that it's counter-productive and bullying of children who really should be considered too young1 to be given My First Body Issues sets.

The problem isn't really the horribly flawed science. That's bad - and there's a serious group-think problem within the "obesity research" field that stops this getting much time.

The problem is the attitude to people that current "public health" schemes seem to regularly get sucked into.

"Public health" is not an easy job. Done well, it can save lives, improve health, and free up valuable funds that are no longer needed to deal with preventable diseases. However, it's something that only really works at a population level.

The infamous bacon cancer study found that the risk of a particular group of cancers increased from 5% to 6% through increased consumption of processed meats. Now, as was pointed out at the time, this means that for 99% of people, whether or not they eat bacon will make basically no difference to their risk of cancer. For many of those people, bacon is a tasty food, and giving it up or cutting back will make them unhappy.

From an individual perspective it's highly unlikely to be a good deal - permanently give up bacon now, to marginally decrease your chances of getting cancer thirty years later? Most of the people who would take that deal probably didn't like bacon that much in the first place, and so probably didn't eat enough for cutting it out to make a difference.

From a public health perspective the calcuation is completely different. If everyone stopped eating bacon, rates of this group of cancers would drop about 20%, saving millions of pounds a year, and saving thousands of lives. Who wouldn't want that?2

The conflict comes between the individual view and the population-statistics view of public health policy. It's very easy to forget that those statistics are made of individuals, all of which - even if they're identical on the measured statistics - have their own very different lives.

Population statistics is the "easy" way to do "public health". It's also - because it reduces people to a few broad numbers - inherently dehumanising. With that, it becomes easy to see why they won't be dissuaded from this course: they're saving [abstract] lives, so if a few [real] lives are harmed as a result, it's still worth it.

Because the underlying attitude is itself dehumanising, some bad science that reduces health to a couple of convenient numbers is very attractive. Rather than looking at individual health, fixing the numbers becomes the goal - and if your numbers are "wrong", that makes you a problem. For people who are already dehumanised regularly on an individual level, the effects combine very strongly

The "hard" way to do public health does also happen, and there are professionals working very hard towards it - providing green spaces, looking at food availability, considering safety issues, and so on - rather than blaming individuals for making the "wrong" choices. An approach that humanises rather than dehumanises is not only more compassionate, it's also more effective.

Not surprisingly, those public health professionals don't make the news for things like this.

Also not surprisingly, the government is quite happy to, for instance, cut funding to local councils and so force them to close leisure centres (something that is certainly bad for public health), but continues paying for schemes like this to shame people with non-approved bodies.

Footnote

1 No, there isn't an appropriate age. The product failed safety tests.

2 Given the current high unemployment, anyone whose job depends on people eating bacon, for a start.