The health effects of a warming planet

This transcript has been lightly edited for clarity

July of 2023 appears to have been the hottest month globally, since we’ve recorded data. The reality of climate change, and the fact that we expect even hotter months in the future and a steady increase, at least for the next 30 years or so, in heat, even if we take action now, has led us at the Global Burden of Disease, and at IHME, to take a number of steps to try to better incorporate the impact of heat into the Global Burden of Disease, and into our forecasting work.

So for the heat part, what we’ve done is we’ve taken all those places in the world with decent vital registration that report by locality and day, and by course, and then tried to figure out what the shape of the risk curves are for specific causes, for diarrhea, for heart attacks, for drownings, for homicide, for pneumonia, etc. Now, it turns out that understanding the relationship between daily temperature and a disease outcome is rather complex because it’s a U-shaped curve for most, but not all causes.

The U-shaped curve says that on days that are very cold, you have increased risk and days that are hot, you have increased risk. And those risks are quite considerable at either end of the spectrum. Now, of course, we’re worried about heat right now. So it’s the high heat, the high end of the spectrum that matters most.

But the extra complicating factor that we’ve learned through these analyses is that the shape of that curve is different depending on what your annual mean temperature is. So if you live in a place that is commonly temperate, then quite moderate temperature increases can cause harm. If you live in a place that is normally, let’s say, hot, 28, 29 degrees Celsius on average, then it takes a higher degree of increase in temperature to see the large increases in risk.

Now, we’ve already characterized these risk functions and published them, and they have now rolled into the GBD. So we have low and high temperature in the Global Burden of Disease, and we’ve then tried to take that information and build it into our future health scenarios, where we forecast what we think is most likely to happen, what we call a reference scenario, out to 2050, and we’ll soon be publishing to the year 2100. 

We try to trace out what is the increase in temperature, particularly as characterized by the CMIP6 ensemble modeling, as part of the IPCC work. Basically, this is the group of modeling institutions that make macro global climate models. And we take those and then try to take our risk functions and characterize where this increase in risk will be.

Places in India, places in the Sahel are particularly going to be affected by heat. But there are many parts of the world that will see heat effects. There are also, in the temperate climates, in places in Europe, in Russia, Canada, parts of the US, there will also be reduction in cold-related deaths.

And so you get this interesting phenomenon where it might actually, on balance, be a net benefit for a select number of communities that can be quite cold, but it’s the poorer communities in sub-Saharan Africa, in parts of the Middle East, in South Asia, where there’s very little reduction in cold harm, because there isn’t a lot, and there’s a very marked increase in the harms that come from high temperature.

And because there’s a very strong correlation right now between the places that are going to get the hottest and where birth rates are the highest, as we go forward into the future, you get this double effect that where temperature increases will be the largest and potentially the most harmful, are also the places where there will be many more people in the future.

For example, our forecasts of population, and the UN’s forecasts of population, suggest a place like Niger, very much at risk in West Africa, will actually increase their population by five, six, maybe eight-fold over the course of this century, depending on the sort of assumptions that you make about fertility decline.

That’s a dramatic increase in the number of people who are going to be the most vulnerable, because they’re poor and they may stay poor for quite some time, and likely have the most food insecurity, and the largest impacts of temperature. So as we think about forecasting the effects of heat into the future, it’s really this combination of what will happen to population and which diseases are going to be most affected  which will tell us about the balance of the effects.

Now, I mentioned at the beginning that some causes stand out as not having the U-shaped curve. So there are certain things  like drowning, which is just a straight line up: the hotter it is, the more people want to go and swim, and that leads to more drownings. And so there are some causes where it’s pretty much a straight line, heat leads to more adverse outcomes.

Now, there are some diseases for which the data are not great, but many people are concerned that as temperature rises, we may see increases in some parts of the world in vector-borne diseases because the vectors, certain species of mosquito, for example, may start to live in places where they didn’t previously live. And that’s a little harder for us to quantify because we don’t have a rich set of observations with good vital statistics and cause of death data in the places that would be on the edge of having epidemic malaria as vectors spread.

So there are many things that remain poorly understood, poorly quantified. But we do understand that the harm from increased temperature will be greatest in the low-income settings where temperature is expected to go up the most. And that’s parts of India, as well as many parts of the Sahel or near-Sahel regions. Now, when we start to think about it, actually, where we’re trying to build into our future health scenarios the full impact of climate change, it goes well beyond temperature, right?

So we need to factor in things like food insecurity. This notion that has been published now in a number of studies, that there is a certain level of temperature where humans historically have not lived – that is, depending on who you talk to, 29 and a half degrees annual day-night mean temperature. So it’s rather high, but this is in Celsius.

It’s not the peak temperature of the day, it’s the annual mean day-night. There’s very few people who live above 29 and a half. You can push that up, maybe, to 30. And there are very few people, of the 8 billion people in the world, roughly, there may only be 30, 40 million people that live in those zones.

And most of those people are in settings where they have enough income to have air conditioning and other devices to protect themselves from the environment. So as we look into the future and see more and more people, as many as 800 million that might be living in parts of the world where the temperature is above this human niche threshold that has been suggested, what will happen to these people?

Will they be in incredibly difficult circumstances, food insecurity, increases in malnutrition, or will some or most choose to leave? And I think that’s a very big question that we would like to have a better understanding of. When is it so hot, and food supplies so poor, that people choose to pack up and leave in the hopes of finding a better life somewhere else?

We suspect that will be a pretty profound phenomenon, and that may turn out to be as big as or bigger, in terms of human consequences, as the migration driven by heat. And even within in high-income countries, where there are lots of ways to mitigate the effects of heat directly, people may choose to move.

You know, there are now analyses today suggesting that maybe many people in southern Spain will move to northern Spain, as an example. We just don’t know. And that’s an area that really deserves a lot more attention as we think of the myriad ways in which climate change can work through and have effects on people and the environment, as well, of course, as all the things that can be done to mitigate and adapt to these types of changes.

The adaptation part is interesting because if you look in the data of heatwaves, there’s a bigger effect right now in the data that we see of heatwaves in Europe than in the US. And the theory behind that is that the US has much more air conditioning than in general use in Europe. And so it may well be that that things like air conditioning can substantially alter the risk curves that we see.

There’s statistical analysis to back those up as well. So as we think ahead to the sort of temperate high-income world as well, many of the heat effects may not be so large, in terms of mortality, because of things like air conditioning, or changes in lifestyle, or adapting to that climate change. But we expect it’ll be very hard for poor people to adapt in the same way.

So that’s a strong focus for us at IHME, to really try and capture the myriad ways in which climate change and heat may affect select diseases, and affect the well-being of people around the world. So expect more from us in the future as we start to publish some of these studies. And some of this work is going to be greatly aided by much more precision mapping of where people are.

Because for us to be able to understand the effects of heat, we really need to know where people live quite precisely. And fortunately we have a joint project with the satellite company Planet and Microsoft’s AI for Good to try to come up with much more highly resolved and continuously updated maps of where people live, which will be a critical ingredient to a deeper understanding of the current and future impacts of heat and climate change.