Statistical modelling of solicited symptoms in vaccine clinical trials

Abstract

In many clinical trials, in order to characterize the safety profile of a subject with a given treatment, multiple measurements are taken over time. Mostly, measurements taken from the same subject are not independent. Thus, in cases where the dependent variable is categorical, the use of logistic regression models assuming independence between observations taken from the same subject is not appropriate. In this report, marginal and random effect models that take the correlation among measurements of the same subject into account were fitted. The models were applied to data obtained from a phase-III clinical trial on a new meningococcal vaccine. The goal is to investigate whether children injected by the candidate vaccine have a lower or higher risk for the occurrence of specific adverse events than children injected with licensed vaccine, and if so, to quantify the difference. We extended the random intercept partial proportional odds model and generalized ordered logit model which assumes identical variability for different category levels by introducing category specific random terms. This is very appealing to study the association between different category levels. Since three outcomes (Pain, Redness, Irritability) are measured on the same child, in addition to analyzing a single outcome variable at a time, joint mixed models for a set of different outcomes were studied to see the association between outcomes. Further, whether the new vaccine has identical effect across different outcomes or not, were investigated based on the joint model and non-significant result was obtained. In conclusion, in both marginal and random effects model, significant difference between the two vaccines were found for at least moderate and severe intensity levels of pain adverse event and all and at least moderate intensity levels of redness. For irritability adverse event, significant difference between the two vaccines were not observed.