It's an excellent question.
The dilemma, really, is that many of the genetic tests that we discuss with patients are predictive tests. Take the example of Huntington's and someone at the age of 20 who has a test for Huntington's that shows they will get the condition, probably in their forties or fifties. That's going to happen. That's certain information. Take the example of someone who has a cholesterol test as part of their annual physical with their doctor and gets as a result a risk for heart disease, that is a risk that can be influenced by a whole bunch of other things. It can be influenced by what the patient does between then and 10 years from then.
I think the distinction we need to make is not so much how the test is being used. I think that we have to protect all genetic information regardless of in what context it came to light. Because the concern is still there: regardless of whether it was a diagnostic test or a predictive test, that person is still subject to discrimination.