How unstable are our politicians?

I am 100% ready to believe that MPs are stressed as heck. All jokes aside, a large percentage of them are decent people trying to do good by their constituents, and in return they get death threats. One of them was murdered a few years ago. They are ranked second-last in the “most trusted professions” index, above only advertising executives. And being involved in trying to get Brexit through – with its utterly irreconcilable, contradictory demands – must be an unbearable nightmare.

But this study tells us nothing.

For one thing – even if it did indicate precisely what percentage of MPs suffered from various mental health conditions, it can’t  say anything about what causes them. It’s a snapshot, a so-called “cross-sectional” study, which looks at a population at a given time. It doesn’t tell you how they change over time, or what causes the things you see.

And while stress could be causing the mental health problems, there’s an obvious alternative hypothesis. We’re dealing with a bunch of people who are similar in one important way: they all chose to go into politics and become MPs. That makes them, immediately, an unusual group.

Not to put too fine a point on it, you have to be pretty weird to want to go into politics. The sort of person who becomes treasurer of the union at university, or joins the Durham Young Conservatives, is not representative of the average student; most of us were too busy drinking, sleeping late and trying to convince our friends that we were having sex. So might it be that the sort of person who wants to become a politician is at greater risk of mental illness?

I can’t find any studies to confirm that theory. Interestingly, there are three – looking at Italian politicians (2010), German politicans (2011), and American politicians (2016)– that find that politicians do have detectably different personality types from the population at large. But the findings are weirdly contradictory: two of them find that politicians are more “open”, one finds they’re less; some find they’re more “agreeable”, some find less.

They do, as you’d expect, tend to be more extraverted, and two studies find they’re less neurotic, i.e. more emotionally stable, which I think would suggest they’re less prone to mental illness. But since they’re an unusual group, it’s not inherently surprising to find that they’re different from the population at large.

But, more importantly, this study is just badly designed. Here’s the problem. There are 650 MPs in total, if you include Sinn Fein and the speakers. The researchers sent the study questionnaire to all 650 of them. Of those, 146 responded.

Now, imagine that there are two MPs, both of whom receive a questionnaire in their pigeonhole with a title like “survey into MPs’ mental health”. One of them is entirely happy and well-adjusted and hasn’t spared a thought for her mental health in years because she is lucky enough not to struggle with it. The other is struggling with depression and anxiety. Which of the two is more likely to fill out the form?

So the group of respondents are even more unusual: they’re not just MPs, they’re MPs who chose to fill out an entirely optional questionnaire about their mental health. That may mean that they’re people for whom mental health issues are particularly salient. It’s also worth noting that the ones who replied are disproportionately young, and younger people are more aware of – and therefore more willing to discuss – mental health issues.

As I said: I’m happy to believe that MPs’ jobs are stressful and even that those jobs cause them to have worse mental health. But this study does very little to confirm that. (I asked a psychologist friend about it and he replied “God, this study looks awful.”)

I think it’s important to push back against it because it’s part of a wider conversation in society, about how we are apparently in the grip of a mental illness epidemic. But I am far from convinced that’s true.

As the UCL neuropsychologist Vaughan Bell pointed out recently, the World Health Organisation and national governments track the impact of mental health disorders on populations. The WHO has found that in western Europe in general and Britain in particular, harms caused by mental health have not risen noticeably in the last 30 years. Another study shows significantly decreasing harms in western Europe since 2000. An NHS/NatCen report found a slight increase in the UK since 1993, but attributed it at least partly to people being more likely to acknowledge it. Those claims are based on large-scale survey data, so they’re as solid as you’re likely to get in this area.

And yet, we are constantly worrying about it, in much the way that we are constantly worrying about air pollution, loneliness and suicide. They’re real problems, absolutely, but they don’t seem to be getting worse. And while on the one hand calling attention to mental health issues might reduce stigma – the NatCen report said that greater awareness might be behind the increase in reporting – it is not an unalloyed good: if we overestimate the prevalence and importance of these conditions, we will dedicate too many resources to combating them. Each pound spent on NHS psychotherapists is a pound not spent on dialysis machines, which is a point I make fairly regularly.

None of this is to say that mental health problems aren’t very real, or – to return to my original point – that MPs’ lives and jobs aren’t stressful, of course. I wouldn’t do it for twice the money they get: it sounds horrible. You’d have to be weird to choose that life. But, then again, that’s the point.

---

Tom Chivers is a science writer. His second book, How to Read Numbers, is out now.

TomChivers