It would be important to understand how people were recruited to the registry. If based on their healthcare provider alerting them to enroll that results in bias as to who is recruited since based on published data by Ogdie et al, the nonWhite percentages are still low. Maybe it is commented on as a limitation in the article.