Environmental damage kills children

1 10 2019

Yes, childrenairpollutionit’s a provocative title, I agree. But then again, it’s true.

But I don’t just mean in the most obvious ways. We already have good data showing that lack of access to clean water and sanitation kills children (especially in developing nations), that air pollution is a nasty killer of young children in particular, and now even climate change is starting to take its toll.

These aspects of child health aren’t very controversial, but when we talk about the larger suite of indicators of environmental ‘damage’, such as deforestation rates, species extinctions, and the overall reduction of ecosystem services, the empirical links to human health, and to children in particular, are far rarer.

This is why I’m proud to report the publication today of a paper on which I and team of wonderful collaborators (Sally Otto, Zia Mehrabi, Alicia Annamalay, Sam Heft-Neal, Zach Wagner, and Peter Le Souëf) have worked for several years.

I won’t lie — the path to publishing this paper was long and hard, I think mainly because it traversed so many different disciplines. But we persevered and today published the paper entitled ‘Testing the socioeconomic and environmental determinants of better child-health outcomes in Africa: a cross-sectional study among nations* in the journal BMJ Open.

Despite the difficult road, the rationale for embarking on such as study was simple — we seem to have largely failed to convince the wider public that the loss of biodiversity is a bad thing, but if we show empirically to what extent  environmental damage erodes children’s health, then maybe a few more people might actually pull their heads out of their arses and take the Great Dying more seriously. It’s really hard to argue against protecting children’s lives — even the coldest, most heartless, far-right bastard would struggle to defend an anti-child ideology.

childwaterpollutionHaving come to the conclusion many years ago that merely chronicling the demise of the planetary life-support system is not having the desired response from the public and our elected officials, I vowed that I would at least attempt to pose more targeted hypotheses regarding how environmental damage can make human lives more miserable. Linking biodiversity erosion to one’s pocketbook or the compromised health of a loved one to me seems like a more productive way to engender more than just passing empathy for our current extinction emergency.

But posing hypotheses is one thing — finding reliable data and testing them is entirely another.

Testing the macro-determinants of child health is a complex, multi-headed beast, to say the least. It’s not just what children are exposed to, it’s also what they eat, how they’re raised, how they’re treated, how they’re educated, what access to medical services they have, and so on.

And then there’s the question of how you measure ‘health’ itself? Is it merely a death rate? Prevalence of infectious disease (which diseases?), incidence of nutritionally driven stunting? The list of candidate metrics is long, indeed.

So we had to be fairly comprehensive and examine a host of health indicators for child health. Here, we choose to look at children under the age of five, which is the World Health Organization‘s standard for defining a ‘child’. This age group also happens to be the most susceptible to dying from various extrinsic sources.


We also had to choose a region of the world where there was enough spatial variation in child-health performance for the models to have any reasonable chance of picking up trends. This is why we chose the African continent, summarising the data at the national scale. Another reason we chose to examine the patterns in Africa was because we have, as some you might already know, spent a good deal of time and effort measuring the relative environmental performance among nations there.

Combined with a suite of other hypothesised variables, we eventually got to the point where we could conceivably test the hypotheses supporting the rationale.

I won’t go into the methodological detail here (you can delve into the modelling here), but I will report the main findings.

No real surprises for the some of the top-ranked contributors to child-health outcomes in Africa. The top two (depending on what is being tested) were clearly the wealth of a nation (measured crudely by gross domestic product), and access to clean water and sanitation. As a country’s ‘wealth’ increases, children survive better mainly because they have better access on average to improved infrastructure and health care. When they don’t have to drink pollutants, heavy metals, and pathogens, or when they have access to good sanitation (toilets and proper sewerage-treatment amenities), they also do better.

But the next two most influential parameters were a bit of surprise, even to us. It turns out that the higher the average number of children in a household — a more proximal measure of the pressure increased population density has on child health — the faster children die. While from the first principles of density feedback eroding the average fitness of individuals in a population is well-established for many non-human organisms, the evidence is considerably rarer for human populations. But as we’ve recently described for African nations that increasing population density is the main determinant of poor environmental performance, so too does it appear to erode the health of children.

childrenairpollution2And then there was the influence of our index of environmental ‘health’ (a combination of deforestation, cropping, livestock, ecological footprint, water extraction, species threat, emissions, and conservation policy). Just as predicted, and in addition to the other explanatory variables, declining environmental performance translated into worse child-health outcomes. Wow, we thought. This is, in my experience, the first macro-scale, empirical demonstration that environmental damage kills children. Even though it intuitively makes sense, I have to admit that I was initially caught off guard by just how strong the relationship is.

Pretty much every other hypothesis we posed followed expectation, but really only air pollution managed to have any relevant statistical power to explain variation in child health.

I’m sure our results will need some digestion, and there’s bound to be some ensuing discussion. That said, I’m overwhelmed by the fact that we are now starting to see some of the direct effects our rapid slide into mass extinction is having on the most susceptible group of humans — little children. Parents who claim to love and protect their children really cannot logically ignore the environmental crisis any more.

CJA Bradshaw

*No, I’m not excited about the title either, but that format was imposed on us by the journal.




Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: