Wealth and health

One of the tropes of the current crisis has been that to re-open the economy is to sacrifice lives on the altar of profits. It's also been widely pointed out that this is a false dichotomy. In fact, wealth and health tend go together. Economic downturns lead to deaths as predictably as viruses. In this post I just want to re-state this view one more time, since I think it's a crucial one to grasp if we want to react sensibly to this (or any other crisis).

As the above graph suggests, there's a positive association between longevity and GDP. GDP and child mortality seem to be inversely related. Deaths from the five most lethal infectious diseases have declined as the global economy has grown.

Correlations like this aren't a slam-dunk case that wealth causes health. Studies that have looked at the association in detail have found it to be a slightly complicated one. But even if the exact causal mechanisms at various stages of growth can be difficult to disentangle, the basic picture seems clear: wealth and health tend to go hand in hand. That's the case not only if you look at individual countries through time, but also if you look the set of countries at a particular point in time and compare the well-being of people in richer and poorer parts of the world (even controlling for other factors).

Moreover, unlike in the case of spurious associations (scores for M. Night Shyamalan's films going down on Rotten Tomatoes in line with newspaper sales, for example), it's not hard to think of reasons why these two variables might be linked, and why the wealth of a country might help its people be healthier. Richer countries can give more funding to health services. They can invest in better-quality housing, safer infrastructure, and a more comprehensive social safety net. Its citizens are wealthier, and they can spend more money on their well-being.

So far we've been looking at the positive side of the story, with better wealth being associated with health. But there's also a dark side to the association, with poverty being associated with disease and shorter life-spans.  You can see this effect with economic downturns even in the rich world: opioid deaths rose by 85% in parts of the US where car factories had closed down (and here too it's easy to think of how this might have happened, with unemployment leading to despair and addiction). A 1% rise in the unemployment rate makes working-age men 6% more likely to die of any cause.

And that's in the rich world. Economic growth is even more vital to the developing world, since increases in wealth just make the citizens of rich countries even healthier, whereas people in poor countries live much closer to death and disaster. They're highly dependant on trade and exchange with the rich world. It's no surprise that UNICEF is now predicting a 45% rise in child mortality because of lockdown-related disruptions.

The relationship between economic downturns and health does have its complications. Some studies suggest fewer people die in the actual course of recessions than normal (although more people die of particular causes, like suicide), but there are lasting health costs over the longer term. So even if the coronavirus lockdowns are followed by a V-shaped recovery, with locked-up demand immediately bursting out again, then we might expect our health (on average) to be slightly worse than it otherwise would have been over the next few decades.

There probably are a few people out there who would put profit for themselves ahead of other people's lives. But most of the people raising warnings about the lockdowns are probably just trying to draw attention to the harm we can do to ourselves if do too much damage to the economy. Whether Covid-19 is dangerous enough to justify the public-health costs associated with the economic downturns that are now starting to bite - that's a different question, and one that's best left for another day (and maybe even another website). But it's not a simple question.

'Your money or your life?' isn't a question that, in ordinary circumstances, anyone would want to be asked. But it's actually a much easier question to answer than the one many countries are faced with now.

What are my chances like?

One evening in that bygone age known as a few weeks ago I was at a dinner party, and the conversation turned to this new virus everyone was making such a fuss about, the coronavirus (or the novel coronavirus known as Covid-19, to be completely accurate). Like a lot of other people, I'd looked at this breakdown by age groups of people who were known to have died from the virus in China. So I told the other people at the dinner party, 'In our age group, we have a 0.2% chance of death. You probably have about as much a chance of dying going scuba-diving or sky-diving - maybe even riding a motorcycle. We're just not that good at comparing risk.'

I was dead right about that last point, and I even managed to exemplify it in the examples I gave, which were, as I later discovered, well off the mark. It turns out that the 2-in-1000 chance you probably have of dying if you get coronavirus before your forties is much higher than the 1-in-about-34 000 chance you have of dying while scuba diving, or the roughly 1-in-100 000 chance you take skydiving. As for riding a motorcycle, it's only if you were doing it in a race that you'd approach the kind of danger you'd face as a thirty-year-old with coronavirus - with a 1 in 1000 chance, you'd only be about half as likely to die as a thirty-year old with Covid-19.

So what kind of thing does give you a 0.2% chance of dying? And, come to that, what sort of activity puts you at something less than a 1% chance of dying, which is starting to look like the best bet for the overall case fatality rate for Covid-19?

The best match for a 2-in-1000 chance (at least on this pretty well-sourced chart I found online) seems to be hang-gliding, whose 1-in-560 risk makes it just slightly less deadly than that. Four bouts as a boxer will get you somewhere near the same amount of risk.

(I'm assuming the chance accumulates when it comes to activities like this, rather than decreasing as it does in the case of multiple coin flips. Dying of a head injury is unfortunately something that increases the more times you get punched, whereas what side a coin lands on isn't affected by what side it fell on last time.)

The closest comparison for the less than 1% risk that a random person with Covid-19 has is Formula 1 racing, which 99 our of every 1000 people are also going to survive. (Of course, those figures likely apply to experienced race-car drivers - if you just jumped into a McLaren and had a go at the next Grand Prix your chances would probably be less than that.) Something that has about twice the risk factor as getting Covid-19 is base-jumping, which has about a 2% fatality rate.

So, if you're under 40, your chances of dying from Covid-19 if you get it are about the same as if you went hang-gliding once (probably assuming you've done it a few times before), or took part in four boxing matches (again, assuming you're not completely untrained). Across all age-groups, getting Covid-19 is about half as likely to kill you as jumping off a cliff or tall building (with a parachute), and presents you with something very like the risk that Lewis Hamilton faces every time he competes.

In conclusion, I clearly under-estimated the risk that even a 0.2% chance of death represents. It's a significantly bigger risk than something like going deep under water with an air-tank strapped to your back. Having said that, that doesn't necessarily mean that it's very dangerous - it may just mean that scuba-diving was even less of a risk than I'd previously thought it was. We don't call off motorcycle racing or Formula 1 because 0.1 or 1% of the competitors are probably going to die (although we have now called them off to stop the spread of Covid-19, interestingly).

Of course, it needs to be noted that these figures represent the (best guess at) the risk for someone who actually contracts the virus, and lots of people won't. (How may will and won't get it over the next few months is something else the experts aren't quite sure about, but I don't think anybody's predicting an 100% infection rate.) It's also worth bearing in mind that these are average figures, and your risk will be higher or lower depending on your other risk factors (especially if you already have other medical conditions).

But, on the whole, it's interesting to think that the overall chance of dying if you get Covid-19 is somewhere between a motorcycle racer's and a base-jumper's. Motorcycle racing and base-jumping aren't activities that we ban, though they're also things that you'd probably want to avoid, unless you were a bit foolhardy. And they, of course, don't really increase the death-risk for anybody else (except maybe rivals you might take out in a crash during your Superbike race).