Friday, March 8, 2013

Malaria equilibria: control and contain or eliminate and eradicate?

In at least one previous post I’ve discussed the absence of malaria in regions where it was once endemic.  The U.S. is one such area, where there is little infrastructure designed to maintain our malaria-free status.  Despite this, and despite the fact that it is sometimes imported into the U.S. via travelers coming here from malarious areas, we just don’t have a malaria problem.  Why is that?  

This question is related to another set of questions:  Given that a nation has endemic malaria, should we focus on elimination?  Or is total elimination unattainable, meaning that our efforts are best aimed at control?  These questions have a history.

In 1955 an initiative known as the Global Malaria Eradication Programme (GMEP) was launched.  As its title infers, the goal was to completely rid the world of malaria.  (Note: “Control” refers to maintaining a low level of background malaria; “elimination” means to rid a nation of the disease; and “eradication” is the elimination of malaria from all nations globally.)  At the time spirits were high and the scientists and participating institutions were pretty optimistic.  However, these efforts were confronted with a host of problems.  Just one of these is evolution: provide a strong selective force against mosquito and parasite populations and you’re bound to see some evolution occur.  The optimism of the GMEP soon began to wane and by 1969 the it had largely collapsed.  In sad irony, those places where malaria had already been worst at the start of the GMEP (e.g. Sub-Saharan Africa) retained their malaria problems…

The Strategy: 
The main strategy of the GMEP was the administration of DDT to kill mosquito vectors and chloroquine to kill malaria parasites.  But it isn’t actually necessary to kill all of the mosquitoes or all of the parasites.  The common logic is that there are actually threshold population sizes, above which transmission becomes endemic and below which malaria reaches extinction (elimination or eradication).  

The origin of this logic can be attributed to a guy named Sir Ronald Ross who was something of a game-changer in the fields of malariology and epidemiology (and arguably in mathematics).  He was the first to address vector-borne disease epidemiology in a mathematical way, introducing a deterministic, compartmental model for understanding malaria dynamics (Ross & Hudson, 1917).  Ross essentially split a study population up into people (compartments) who were either susceptible to infection or were infected and mosquitoes that were likewise either susceptible or infected.  

From this model one can calculate something called R0 (r-nought: the reproductive rate).  R0 is also used in other fields such as demography and population ecology, but in the case of epidemiology it basically refers to the average number of infected individuals that will come from the addition of one infected individual in a population.  In the case of malaria, we have to consider not only the mosquito biting rate and transmission probabilities (both from mosquito to human and human to mosquito) but also demographic factors such as birth and death rates.  Lotka showed that when R0 is greater than 1, infection numbers will approach a steady equilibrium point (where malaria is endemic) but that when it is less than 1 infection numbers will eventually fade away altogether (Lotka, 1923).   (MacDonald also worked with R0 and made extensive improvements to these original formulations (MacDonald, Cuellar, & Foll, 1968)).   

So, the goal of the GMEP was to aggressively disrupt the apparently stable equilibrium points (at R0 > 1), by targeting mosquitoes and parasites, so that the disease would be eradicated in target regions.  

The reasons for the failure of the GMEP to reach its ultimate goal are complex, and aren’t entirely the point of this blog post.  What is relevant to this post, however, is the fact that nations which didn’t quite attain eradication subsequently saw malaria case numbers rise to previously high levels (or even higher) whereas nations that did achieve eradication haven’t similarly had to continue control efforts.    

A recent Science paper by Chiyaka et al. specifically addresses these issues (2013).     

The authors note that in those nations where malaria was not eradicated, imported malaria is a real threat to malaria control.  (Some of my dissertation research is actually aimed at addressing this issue along the Thai-Myanmar border).  They also calculated R0 for nations that were both in the original GMEP and had eradicated the disease, finding yearly average R0 values to be much lower than 1 (~0.04).  This finding adds credence to what I and many other malaria researchers have already noticed: malaria importation in regions where malaria has long been eradicated is a very different thing when compared to malaria importation in regions that have never been able to completely eliminate the disease.  

It also addresses a much more important issue: What should we do with our malaria efforts today?  Aggressive malaria elimination programs are expensive, perhaps more so per unit time than malaria control programs.  Furthermore, if an elimination effort doesn’t work, you’re left with malaria control costs anyway.  However, if an elimination project does work, and if what history tells us (and what the above mentioned paper illustrates) is true, then the higher costs associated with elimination projects are more than worth it.  It seems that if you disrupt the malaria transmission cycle in such an aggressive way as to knock it out of its equilibrium state, then it is possible to maintain a malaria free equilibrium with relatively little effort.  

Why aren’t we doing this everywhere?  

Partially because people aren’t sure that this strategy will work.  The failure of the GMEP and other similar efforts must have left at least one generation of malariologists and tropical medicine workers disillusioned.  

Also, evolution is a problem.  Malaria parasites are able to develop resistance to just about everything we throw at them.  And aggressive treatment can actually maximize the fitness of resistant parasites (Read, Day, & Huijben, 2011).  Malaria elimination or even just control must take into account the role of evolution in disease ecology, and the fact that evolution doesn’t always work in a completely intuitive fashion.  But does this mean that mass administrations wouldn’t be able to perturb the malaria equilibrium enough that the relatively few resistant parasites that are left won’t actually matter?  I don’t know and it’s a risky proposition.   
Finally, I increasingly wonder whether the true best approach would be to re-direct tropical disease funding toward universal, free primary health care for all individuals in all nations rather than targeting individual diseases.  I do know that a LOT of money is thrown at individual diseases, but I don’t know if it is enough to re-design national health systems.  However, if it was possible to do so, we would probably see improvements in all-cause mortality and morbidity, regardless of the disease.  But here in the U.S. we won’t even do that for ourselves, so I’m not holding my breath that our tax dollars will ever go toward helping other nations do it either.   

There are no simple answers.  

Chiyaka, C., Tatem, A. J., Cohen, J. M., Gething, P. W., Johnston, G., Gosling, R., Laxminarayan, R., et al. (2013). The Stability of Malaria Elimination. Science, 339(6122), 909–910.

Lotka, A. J. (1923). Contribution to the analysis of malaria epidemiology. I. General part. American Journal of Epidemiology, 3(supp1), 1.

MacDonald, G., Cuellar, C. B., & Foll, C. V. (1968). The dynamics of malaria. Bulletin of the World Health Organization, 38(5), 743–55. Retrieved from

Read, A. F., Day, T., & Huijben, S. (2011). The evolution of drug resistance and the curious orthodoxy of aggressive chemotherapy. Proceedings of the National Academy of Sciences of the United States of America, 108 Suppl 2, 10871–7. doi:10.1073/pnas.1100299108

Ross, R., & Hudson, H. P. (1917). An application of the theory of probabilities to the study of a priori pathometry. Part II. Proceedings of the Royal Society of London. Series A, 93(650), 212.

No comments: