Back

Genomic epidemiology as a tool for understanding drivers of hepatitis A community outbreaks in Massachusetts and New Hampshire

Krasilnikova, L. A.; Bouton, L.; Brock-Fisher, T. M.; Decker, E.; Godec, M.; Thompson, Z.; Dart, E.; Gao, F.; Gladden-Young, A.; Messer, K. S.; Norville, J.; Specht, I.; Osinski, A.; Li, J.; Lones, C.; DeRuff, K. C.; Siddle, K. J.; Church, D.; Benton, C.; Hansen, K.; Bowen, H.; Bhattacharyya, S.; Epie, N.; Brown, C. M.; Madoff, L. C.; MacInnis, B. L.; Gallagher, G. R.; Smole, S.; Bean, C.; Talbot, E. A.; Burns, M.; Doucette, M.; Fortes, E.; Park, D. J.; Sabeti, P. C.; Wohl, S.

2026-05-19 infectious diseases
10.64898/2026.05.14.26352933 medRxiv
Show abstract

Despite the existence of an effective vaccine, the United States continues to experience outbreaks of hepatitis A, including in Massachusetts (MA) and New Hampshire (NH) in 2018 and again in MA in 2023. To clarify the relationship between these outbreaks and better understand their drivers, we generated hepatitis A virus whole genome sequences from reported cases and analyzed them using open-source genotyping tools developed and released as part of this study. We found that the 2018 and 2023 outbreaks were caused by distinct viral strains, despite affecting individuals with similar demographic characteristics and reported risk factors. Detailed analysis of genomic and epidemiologic data further resolved transmission patterns within and across outbreaks, showing that experiencing homelessness and prior use of drugs were associated with increased transmission while also revealing transmission between individuals with and without these risk factors, as well as spread across state borders. Together, these findings demonstrate the value of broadly accessible genomic tools for understanding hepatitis A outbreaks and illustrate how whole genome analysis can complement epidemiological investigation by resolving transmission patterns and outbreak drivers that can inform public health interventions.

Matching journals

The top 9 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.