For the use described in your question, the surveys would either need to be published in a peer reviewed journal, or it would need to be designed, conducted & analyzed without sponsor’s influence (s5.10.2). The messaging would be limited to aspects which are non-clinical (so as not to require a clinical trial as evidentiary support) and you’ll need to remove messaging deemed to be disparaging to a product or product class/category. CAVEAT: This is a very general answer to a general question. There are MANY factors which could impact whether a survey is acceptable as evidentiary support. In my experience, things have gone more smoothly when clients first submitted the survey (or at least a thorough description) in a written opinion request (see our fee schedule) along with the types of messages they wished to convey. Sometimes the issue is the reference, but often the issue is the underlying message.