Wednesday, May 8, 2013

Malaria and climate change - the danger of applying simple answers to complex problems

As the world warms, is the potential range of endemic malaria increasing?  A story in the UK's Guardian suggests it is, and will continue to do so.
Leading health experts are urging the government to take action against the growing threat that mosquito-borne diseases, including potentially fatal malaria, could soon arrive in the UK.
The disturbing recommendation to "act now before it is too late" is being made as a growing body of evidence indicates that what were once thought of as tropical diseases are being found ever closer to the UK.
The story goes on to say that dengue has recently been reported in France and Croatia, and malaria in Greece, so the UK needs to plan public health measures now, as climate change is happening, to thwart the threat.  The story also says that climate change will broaden the range of disease-carrying ticks as well, and that Lyme Disease is increasing in incidence in the UK and Europe, and West Nile virus is poised to spread, further evidence that climate change is going to affect the epidemiologic landscape.
Culex quinquefasciatus; Wikimedia
But, is it that simple?  Lyme Disease is carried by a tick that does just fine in temperate zones, including surviving hard winters.  So, is this tick's habitat increasing because of climate change?  Well, perhaps indirectly. There are now more deer in the UK then in the last 1000 years, which means continual banquet for the tick.  This is thought to be partially due to milder winters, but also to changing landscape in the UK, changing agriculture, and so forth.  So, the spread of infection is a consequence of the spread of the tick, and the spread of the tick is a consequence of the spread of deer, which is partially a consequence of climate change.   

West Nile is not strictly a tropical disease either, and is transmitted by a number of mosquito species, including of the genus Culex.  Culex modestus was recently found in the UK for the first time since 1945, prompting the suggestion that West Nile will soon become a problem.  But if this mosquito was in the UK so long ago, is it right to attribute its return solely to climate change?

In essence, the issue is the casual slippage between observing correlation and assuming causation, one of the most difficult things for the public and even scientists to resist.  Which is not at all to say that climate change won't shake up disease prevalence.  Just that conflating possible effects of climate change with possible consequences of travel or spreading animal habitats and so on that may affect the spread of disease doesn't make understanding and prediction any easier. 

Aedes aegypti; dengue fever vector; Wikimedia
Malaria and dengue fever are another question.  Both mosquito-borne diseases, we think of them as tropical diseases now, but malaria was endemic in the UK and at least southern Europe and the US, and even Canada probably for centuries and dengue was endemic in the southern US and at least sporadic in Europe.  Both diseases have been been controlled in temperate zones (or rather, the rich North) since the end of WWII or thereabouts, primarily because of public health measures like cleaning up standing water, insecticide use and so on.

The map below (originally from Kiszewksi et al., 2004. American Journal of Tropical Medicine and Hygiene 70(5):486-498) shows the location of important malaria mosquitoes, graphically making the point that it's not just the presence of the vector that determines the presence of disease.

Global map of dominant and potentially important malarial vectors; CDC
A 2002 paper in The Lancet on the potential health consequences of climate change points out that the ability to control malaria in areas in which it has long been absent may change with the climate. 
The decline of malaria in Europe and the southern USA in the 20th century was attributable to social and economic development and improved public health services, and not to change in climate. However, this finding does not mean that ambient temperature and rainfall are irrelevant from a health perspective. Climate is one of the fundamental forces behind epidemics, and its effects become evident if adaptive measures falter or cannot be extended to all populations at risk.
Models of the effects of climate change often predict the spread of vector-borne diseases into areas currently clear of the vector and thus of the diseases, but the point of models is to simplify reality.  It's not in fact clear what will happen with diseases for which prevalence depends on a complex of interacting factors, that may not be wholly or adequately taken into account in the models.  For example, the prevalence of transmitting mosquitoes and malaria in a given area is a consequence of social and economic conditions (e.g., conflict that might be impeding public health measures, insufficient funds for public health measures like clearing standing water in public places),  numerous aspects of a country's infrastructure (impassable roads, insufficient health clinics), as well as characteristics of the at-risk population, migration patterns, multiple climate factors, and so on.

So, climate change certainly has the potential to alter disease patterns anywhere and everywhere.  But there are numerous factors that affect vector-borne disease prevalence, and climate is only one of them.  Whether or not it is a major one probably depends on how well controlled these diseases are now, even before climate change throws its wrench into the works.  

We write often about how simple answers in genetics have appeal, even when they don't fit the complex question, and that seems to be true with climate change as well.  The reality is probably going to be quite complex, with a lot of heterogeneity in factors that do or don’t contribute to disease persistence and spread.

Most countries that have been successful in eradicating “tropical diseases” are also wealthy countries; this correlation shouldn’t be ignored. For much of the world, these diseases aren’t really an emerging threat compared to what is already experienced. But perhaps the threat of spreading disease will force more people to realize global realities.

There is another, perhaps related, issue in the sociology of science.  Those who are worried about climate change, or whose political agenda it is to advocate some preventive and corrective responses to it, that might change human behaviors such as consumption of fossil fuels, tend to seize on any fact that might boost their case.  This is only natural, but it exacerbates the problem of assuming that correlations imply causation.  Opponents of climate-change concerns naturally tend to dismiss such evidence as inconclusive, of course.

Finally, there is an important thing to consider with regard to climate change and its potential effects. Already deniers cling to anything they can in order to show that it either isn’t happening or it isn’t something worth worrying about. We should be careful in our predictions so that they too aren’t turned around on us. If malaria doesn’t become endemic in the UK, does that also mean we don’t need to worry about climate change?! Things are likely to change, it’s just really hard to say which ones and in which ways.

This post was contributed to by Anne, Dan and Ken. Please note that Dan and Ken have the cover story, about subjects related to this post, in the current (March/April 2013) issue of the journal Evolutionary Anthropology. The article is viewable without subscription.


Holly Dunsworth said...

Thanks so much for this post.

Can you please help clarify the map? Which mosquitoes are malaria vectors? All? I think most of us know where malaria is most prevalent and can assume the mosquitoes there are vectors, but are ALL the rest of the ones on the map vectors too? Thanks much.

Anne Buchanan said...

I'll try, though it's a bit complex for a comment. I recommend the paper -- Jeffrey Sachs is senior author. It's a 2004 paper proposing an index representing the stability of malaria transmission. The authors include what they consider the most important biologic characteristics of vector mosquitoes, and which these are potentially varies around the world, to determine which contribute most to 'the force of transmission'. These are the proportion of blood meals the mosquito gets from humans, daily survival of the mosquito, and length of the transmission season and of 'extrinsic incubation.' They conclude that these characteristics interact with climate to determine the intensity of transmission and its variation around the world.

"...we derived a spatial index of the stability of malaria transmission based on the most powerful intrinsic properties of anopheline mosquito vectors of malaria that interact with climate to determine vectorial capacity. Because this index examines potential transmission stability, it includes regions where malaria is not currently transmitted, but where it had been transmitted in the past or where it might be transmitted in the future. This index, therefore, includes “anophelism (with as well as) without malaria.”"

Their 'stability index' indicates, they say, that malaria transmission is much more stable in sub-Saharan Africa than anywhere else in the world. That is, while there are potential vectors elsewhere and even climate potentially favorable to the disease, the climate interacting with characteristics of the vector/s in sub-Saharan Africa seem to be optimal. And, when anopheles mosquitoes aren't biting, due to seasonal variation, there are other vectors to take up the slake.

Bottom line: "Due to the superior capacity of many tropical mosquitoes as vectors of malaria, particularly those in sub-Saharan Africa, antimalaria interventions conducted in the tropics face greater challenges than were faced by formerly endemic nations in more temperate climes."

Holly Dunsworth said...

Thanks Anne! I'm glad I asked ... thanks again!