Understanding Measles’ Rise Through Networks
In Measles Cases Continue to Rise Around the World, New York Times reporter Abdi Latif Dahir chronicles the article’s titular phenomenon. He reports recent measles outbreaks in the Democratic Republic of Congo, Ukraine, Brazil, the Philippines, and the United States, and gives the World Health Organization’s summary explanation of “a deep mistrust of vaccines, gaps in immunization coverage and lack of access to health care facilities or routine checkups” for the increase.
The notion of an epidemic’s spread being proportional to the product of p, the probability of an infected person infecting another, and k, the expected number of people a sick person can infect comes up repeatedly throughout the article—though this model for an epidemic’s spread is never explicitly mentioned. Dahir documents how things which would increase p are common in the affected countries. In these countries, scares over immunization have decreased vaccination rates, making people in affected countries more susceptible to infection. Additionally, he notes that there exist fears that measles might spread to an isolated tribe in Brazil’s Amazon with a low disease resistance. With their high p values, the impact of measles within the tribe would likely be especially dire.
Dahir also notes how increases in k put more people at risk, often with k jumping from near-zero to a higher number when two populations with little previous contact intersected. He notes that “Global health officials are currently investigating whether Orthodox Jewish pilgrims to a rabbi’s grave in the Ukrainian city of Uman may have carried measles to the United States via Israel,” and that there exists a worry in Brazil that Venezuela was the source of the outbreak there.
As morbid as the entire subject is, to see a mathematical model be so relevant is surprisingly awesome, especially when it concerns a more complex social network than a tree-based one.
Source: https://www.nytimes.com/2019/11/26/health/measles-outbreak-epidemic.html