Back

Estimating the maximum risk of measles outbreaks due to heterogeneous fall in immunization rates

Wu, N.; Moon, S. A.; Falk, A.; Marathe, A.; Vullikanti, A.

2023-10-25 health informatics
10.1101/2023.10.24.23297486 medRxiv
Show abstract

Immunization rates for childhoold vaccines, such as MMR, have seen a reduction over the recent years; this fall has only been accentuated after the COVID-19 pandemic. However, there is limited data on where the rates have reduced, and prior work has shown that heterogeneity in the drop in immunization rates has a significant impact on the risk of an outbreak. An important question from a public health perspective is: what is the maximum size of an outbreak in a region, when limited information is available on the fall in immunization rates within the region? This turns out to be a very hard computational problem. We develop a Bayesian optimization based approach for estimating the maximum outbreak size, and use it on a measles model for the state of Virginia. Our results show that the maximum outbreak size is several orders of magnitude higher than estimated in a baseline which assumes homogeneous fall. Even for a 5% reduction in the statewide immunzation rate, the expected outbreak size can be very high. The maximum outbreak size depends crucially on the importation location, i.e., where the disease starts, and importation in an urban region leads to a significantly higher outbreak. The outbreak size remains high even if the drop in immunization is bounded in health service areas in the state.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.