Pairing donors and patients for marrow transplants is difficult enough, given the genetic requirements for a match.
But, as researchers at the University of California San Francisco discovered, it’s even tougher given the nature of surveys.
A team at UCSF recently used a sample of donors from the National Marrow Donor Program to compare different methods of self-identification via survey. A total of 1,752 potential donors were sent questionnaires with multiple measures of identification, as well as a cheek swab for a DNA sample.
The result: No single measure of self-identification best predicted the genetic ancestry identified in the DNA samples. The most accurate survey information came from the geographic origins of the donors’ grandparents, but even then, the data was not perfect.