Niel Hens: Perspectives on heterogeneity in acquisition of infectious diseases

Abstract:
Frailty models are often used in survival analysis to model multivariate time-to-event data. In infectious disease epidemiology, frailty models have been proposed to model heterogeneity in the acquisition of infection and to accommodate for association in the occurrence of multiple infections.

More traditional frailty models in infectious disease epidemiology rely on the assumption of lifelong immunity after recovery (Farrington et al. (2001). In Abrams and Hens (2014) refinements have been made to account for reinfections with the same pathogen. Farrington et al. (2012) and Unkel et al. (2014) introduced and applied time-varying shared frailty models to paired bivariate serological data. Abrams, Wienke and Hens (submitted) extended the proposed frailty methodology to account for age-dependency in individual heterogeneity through the use of age-dependent shared and correlated gamma frailty models extending also previous work by Hens et al. (2009). More recently, overdispersed frailty models have been investigated (Abrams et al., in prep). In this talk an overview of these developments will be given. The methodology will be illustrated using bivariate current status data on parvovirus B19 and varicella zoster virus, and Hepatitis A and B. In addition several perspectives on the importance of heterogeneity will be given.

References
Abrams, S. & Hens, N. Modeling individual heterogeneity in the acquisition of recurrent infections: an application to parvovirus B19/, 
/2015/, 16/, 129-142
Abrams, S., Wienke, A., Hens, N. Modelling time-varying heterogeneity in recurrent event time data: an application to serological data. JRSS-C In revision
Abrams, S., Aerts, M., Molenberghs, G., Hens, N. Overdispersed frailty models for Type I interval censored data. Biometrics In revision
Farrington, C. P., M. N. Kanaan, and N. J. Gay (2001). Estimation of the basic reproduc- tion number for infectious diseases from age-stratified serological survey data. Journal of the Royal Statistical Society: Series C (Applied Statistics) 50 (3), 251–292.
Farrington, C. P., S. Unkel, and K. Anaya-Izquierdo (2012). The relative frailty variance and shared frailty models. Journal of the Royal 
Statistical Society Series B 74 (4), 1–24.
Hens, N., A. Wienke, M. Aerts, and G. Molenberghs (2009). The correlated and shared gamma frailty model for bivariate current status data: An illustration for cross-sectional serological data. Statistics in Medicine 27 (14), 2785–2800.
Unkel, S., C. P. Farrington, H. J. Withaker, and R. Pebody (2014). Time varying frailty models and the estimation of heterogeneities in 
transmission of infectious diseases. Journal of the Royal Statistical Society Series C 63 (1), 141–158.