Today marks one of my final days of research in the vacation scholarship program.
I chose to do my research in epidemiology with Dr. Jennifer Flegg, focusing on malaria, and how drug resistance develops and spreads among malaria parasites. Malaria is still one of the biggest child killers on the planet. My goal was to develop models that describe how drug resistance arises, so that better tactics can be developed to delay or prevent it.
At the start of my scholarship, I asked Jen how drug resistance in a disease was modeled. She responded with “well, how would you model it?” So I built a model using differential equations, and changed it over a course of weeks as I learned more.
Once I had a working model, I used it to simulate the emergence of drug resistance in areas with differing levels of malaria. Interestingly, I found that in regions where malaria was more prevalent, drug resistance took a longer time to evolve. This is because in areas where nearly every person has malaria, nearly everyone has a resistance to malaria as well. Because of this, less people will feel sick, and so less people will take drugs. This result echoes historic data – antimalarial drug resistance has always emerged in Southeast Asia, where malaria is less common than it is in Africa.
Using MATLAB, I ran a simulation for 100 years, asking what percentage of treatable people should be given this drug so as to maximise the number of people who can be cured in the long run. This question is not as straightforward as it seems. If everyone is given the drug whenever they feel sick, then drug resistance will evolve faster, lowering the benefit for everyone in the long run. If, instead, nobody is given the drug, then the end results will be the same as if the drug was never developed.
The final answers varied based on the prevalence of malaria in the region, but for a region where malaria’s burden is moderate, the fraction of people who should be given access to treatment turned out to be 26%.
26 percent. That is to say, in order to help the most people over a century, we should ignore almost three quarters of people who are sick today, leaving them to suffer from preventable month-long infections. Is this right? If our goal is to prevent the most suffering, are we expected to resist the temptation to heal the people we see suffering around us, in order to prevent even more suffering in future generations?
And how should policy makers use this information? Should drugs be distributed so as to maximise the number of lives saved over one hundred years? Two hundred?
Or just one election term?
Benjamin Metha, January 29, 2018.
Benjamin Metha was one of the recipients of a 2017/18 AMSI Vacation Research Scholarship.