Correction and Addendum to Measles Resurgence in the USA: How International Travel Compounds Vaccine Resistance

In the paper that appeared last week, we wish to correct the list of countries travel from which presents the greatest risk to the U.S. in 2019. The following correction will appear in The Lancet Infectious Diseases:

The list of countries that pose the highest risk of measles to the USA was been corrected to “Ukraine, Mexico, Cuba, Israel, Japan, Thailand, and Philippines.”

This correction has been made to the online version as of 20 May 2019, and will be made in the printed Comment.

The earlier list of countries was incorrect because of a computational error we have now corrected. The error was a mistake in data input.

I would like to make three points:

1. Don McNeil in The New York Times accurately quoted me as saying: “What we did not calculate at all was that it would come from Israel.” It turns out that our risk function does implicate Israel once the correct data are used. This adds to the credibility of the analysis.
2. Several individuals have been skeptical about the importance of Mexico and Cuba. That skepticism arises from our use of suspected cases (as opposed to confirmed cases) to calculate 2019 measles incidence rates around the world. In the case of Mexico, while there are hundreds of suspected cases, there are much fewer confirmed cases. Now, in our data analysis from 2010 to 2018 we do use confirmed cases in our calculations (using WHO data). For 2019, for many countries, suspected cases will take time to be confirmed or disconfirmed. If we restrict the data to confirmed cases, we would grossly underestimate the measles activity in a country. So, we use suspected cases uniformly. Now, the skeptics could be correct that, in the case of Mexico and Cuba, the suspected case count will go down drastically. However, this amounts to special pleading for treating individual countries separately. We need a method applicable to all countries and we are not in a position to make these judgments ourselves. If, for instance, the CDC or WHO provides guidelines on what should be the estimate for each country, we would be happy to incorporate that analysis in our results.
3. Today, a reporter based in New York City told me that the press attention to our paper has encouraged anti-Semitic incidents in the city. We are disgusted by these events and could not condemn them more strongly. By now the roles played by low vaccination rates in some Brooklyn communities of ultra-orthodox Jews and travel to Israel in the large New York outbreak are well known and noted in the New York Times piece (among many other sources); the Times piece also notes that travel from Ukraine was the source of the Israeli cases. We do not go into such detail in our paper but I have also noted it in comments to the press. But it is typical for infectious disease outbreaks to emerge in communities that live in spatially concentrated units and it can be any such community. Using such events for anti-Semitic or racist ends is unconscionable–but perhaps only to be expected in the age of Trump. We urge all researchers like us to work actively against such misuse of all our results.