I would say we need to be careful when we're working with the kind of data we have available. It's not data that supports easily establishing relationships between cause and effect.
While we can measure differences and determine differences and the prevalence of many different conditions, it's difficult to tie that back to specific experiences unless we have a full history of their military occupation. That's a set of information we generally don't have access to when veterans come to work, when they come to VAC for services, when we're talking to veterans or when veterans are filling out questionnaires that provide data to VAC or to Statistics Canada. We don't have that full military career history.