Estimating Vaccine Efficacy from Household Data Using Surrogate Outcome and a Validation Sample

• Activity Number: 175
• Type: Contributed
• Date/Time: Tuesday, August 13, 2002 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology*
• Author(s): Xiaohong Davis*+ and Michael Haber
• Affiliation(s): Centers for Disease Control and Prevention and Emory University
• Keywords: vaccine efficacy for susceptibility ; vaccine efficacy for infectiousness ; household data ; semiparametric method ; surrogate outcome ; validation sample

Abstract

Estimation of vaccine efficacy has traditionally focused on vaccine efficacy for susceptibility or VE_S. A vaccine may also lower the infectiousness of a vaccinated person who became infected. This is called vaccine efficacy for infectiousness or VE_I. Estimation of VE_I is challenging because it requires information on exposure to infection and gathering this type of information is often difficult. The problem of estimating VE_S and VE_I is further complicated by the fact that reliable infection outcome is often expensive or impossible to collect across the population under study. Often, a closely related outcome may be used as a surrogate for the infection outcome. We propose a semiparametric method that uses surrogate outcome and a validation sample for household data to estimate VE_S and VE_I. The proposed method shows improved efficiency as compared to using the validation sample only and smaller bias as compared to using surrogate outcome only in estimating VE_S. Using household data is shown to greatly improve the attenuation in VE_S as compared to findings from previous studies. In estimating VE_I, the proposed method performs better when the true VE_I is smaller.