# Contact Patterns Drive Age-Structured Transmission Dynamics and Seasonality of Scarlet Fever

## TL;DR

This study shows that contact patterns, especially among school-aged children, drive the spread of scarlet fever in Shanghai, with high-risk groups identified for targeted interventions.

## Contribution

The study introduces an age-structured SIR model with seasonality to quantify scarlet fever transmission dynamics and identify high-risk groups.

## Key findings

- Children aged 7–9 years have the highest force of infection for scarlet fever.
- Transmission seasonality among school-aged groups has a 39% seasonal amplitude.
- The reproduction number R0(t) ranges seasonally between 3.02 and 8.83.

## Abstract

Background: Scarlet fever has seen a sharp increase in its reported incidence in China since 2011, and this study focuses on Shanghai as a representative setting to systematically investigate its transmission dynamics by analyzing age structure. It further identifies high-risk age groups and provides a theoretical foundation for prevention and non-pharmaceutical intervention strategies. Methods: We developed an SIR model that incorporates age structure and seasonality of transmission rate. In parameter estimation, the methodology of the partially observed Markov process framework is employed to derive results based on monthly data. The time-varying reproduction number R0(t) is derived monthly from the next-generation matrix. Age-specific forces of infection are estimated to identify high-risk groups and quantify how school-term-driven contact patterns modulate transmissibility. Results: The force of infection peaked in children aged 7–9 years, whereas the force of infection was highest among adults aged 35–39 years. The seasonal amplitude for transmission among school-aged groups was 39% (95% CI: 37–41%). The estimated R0(t) varied seasonally between 3.02 and 8.83. Conclusions: The transmission rate in Shanghai shows strong age heterogeneity and school-driven seasonality. Children aged 7–9 years are the highest-risk group, and interventions should target them during periods of high R0(t).