Transmission and control of hand, foot and mouth disease in schools

Abstract

Hand, foot and, mouth disease (HFMD) is a common paediatric disease in Asia which sometimes leads to severe complications or even deaths. It is caused by enteroviruses, where three serotypes namely enterovirus 71 (EV-A71), Coxsackievirus A16 (CV-A16) and Coxsackievirus A6 (CV-A6) are most prevalent. HFMD is highly transmissible in community settings such as schools and households, and outbreaks are frequently reported in schools. While some control measures are recommended in schools, their effectiveness has yet to be assessed. The overall objective of the thesis is to study the transmission of HFMD in schools and to evaluate the effectiveness of various control measures.

To achieve the objective, I conducted an observational epidemiological study from May 2015 to February 2016. Symptom profiles of HFMD cases were described. The incubation period distribution of HFMD, defined as the duration from being infected to symptom onset was not available in the literature. Using interval censoring survival analysis, the median incubation period of HFMD was estimated as 5.4 (95% CI 4.4-6.5) days, with a 95% percentile at 14 days. Epidemiological parameters derived from this observational study were used in the transmission in subsequent analysis.

Furthermore, I developed a Susceptible-Exposed-Infectious-Absent-Recovered (SEIAR) model for the transmission of EV-A71, CV-A16, and CV-A6 within a class. The model considered infected cases may develop HFMD or herpangina, or remain asymptomatic, to provide realistic description of the transmission dynamics. Simulations of outbreaks in class were performed under the scenario with no interventions, in both deterministic and stochastic approaches. Most of the infections were symptomatic HFMD for EV-A71 and CV-A16 and herpangina for CV-A6.

The transmission model was further applied for assessing control measures in school class. I considered three non-pharmaceutical interventions (NPIs), namely class suspension, case isolation, and improving hygiene. The percentage reduction in attack rate (ΔAR) between scenarios with or without interventions was used to assess effectiveness. Class suspension was found to be the single most effective intervention (ΔAR: 64% – 91%) for EV-A71 infection, though it would lead to large social burden for childcare. Alternatively, the combined intervention of case isolation and improving overall hygiene (ΔAR: 1% – 100%) was found to be a practical and effective option to interrupt within-class HFMD transmission.

In Hong Kong, citywide school closure has been implemented due to severe influenza epidemics. This provided a unique opportunity to assess the impact of school closure and holidays on the transmission of HFMD. Using HFMD hospital admission data from 2010 to 2019, the transmissibility of HFMD was estimated to reduce slightly by 9.5% (95% CI: 1% – 17.3%) during school closure or holidays. Overall, class-based interventions were found to be more effective and practical control strategies.

In summary, I developed a mathematical model to characterize HFMD transmission in school class, considering the broad spectrum of disease presentation including herpangina and asymptomatic infections. A combined intervention of case isolation and improving hygiene was the most effective and practical strategy to control HFMD in affected schools. My research findings will inform the control of HFMD transmission in the school setting.