Back

Novel multiplex tools in an epidemic panel improve prediction of RSV infection dynamics and disease burden a RESPINOW analysis

Harries, M.; Klett-Tammen, C. J.; Rodiah, I.; Dulovic, A.; Jaeger, V. K.; Krepel, J.; Contreras, S.; Maak, K.; Marsall, P.; Moeller, A.; Heise, J. K.; RESPINOW Study group, ; Castell, S.; Schneiderhan-Marra, N.; Karch, A.; Lange, B.

2024-11-23 epidemiology
10.1101/2024.11.20.24317653 medRxiv
Show abstract

Respiratory Syncytial Virus (RSV) is one of the leading causes of morbidity and mortality among infants and adult risk groups worldwide. Substantial case-underdetection and gaps in the understanding of reinfection dynamics of RSV limit reliable projection estimates. Here, we use a novel RSV multiplex serological assay in a population-based panel to estimate season and age-specific probability of reinfection and combine it with sentinel and notification data to parameterize a mathematical model tailored to project RSV dynamics in Germany from 2020 to 2023. Our reinfection estimates, based on a 20% post-F and a 45% N antibody increase in the assay over consecutive periods, were 5{middle dot}7% (95%CI: 4{middle dot}7-6{middle dot}9) from 2020 to 2022 and 12{middle dot}7% (95%CI: 10{middle dot}5-15{middle dot}2) from 2022 to 2023 in adults. In 2021, 30-39 year olds had a higher risk of reinfection, whereas in 2022, all but the 30-39 age group had an increased risk of reinfection. This suggests age-differential infection acquisition in the two seasons, e.g. due to still stronger public health measures in place in 2021 than in 2022. Model-based projections that include the population-based reinfection estimations predicted the onset and peak for the 23/24 RSV season better than those only based on surveillance estimates. Rapid, age-specific reinfection assessments and models incorporating this data will be critical for understanding and predicting RSV dynamics, especially with changing post-pandemic patterns and new prevention strategies e.g. monoclonal antibody. Helmholtz Association, EU Horizon 2020 research and innovation program, Federal Ministry of Education and Research, and German Research supported this work.

Matching journals

The top 6 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.