In our institution, and in Minnesota, access to patient records for research purposes is not allowed without prior authorization (1). Before a patient visits the Mayo Clinic, a letter is sent wherein we request the patient's authorization to use the medical record for research purposes (2). Patients have the opportunity to consider what is offered in the privacy of their home without the additional influence or "coercion" of a physician or study coordinator. The new Minnesota law does not, however, prohibit contact with an individual to offer an opportunity to participate prospectively in individual research studies, even if the person has not given permission for their medical record information to be used for research purposes. Furthermore, data (such as billing records) contained in administrative databases, which are not part of the medical record, can be used to reach potential research participants.

To optimize care of our patients, our division has instituted disease interest groups. Here a small number of people, or one physician, see most, if not all, patients with a particular disease. These interest groups also perform clinical trials for which enrollment is offered when appropriate. Physicians who want to refer a patient with a specific disease are brought in contact with the disease interest group. This allows concentration of expertise that has served our patients well. Patients can also be referred by using the methods as outlined by Dr. Jones and, recently, by "bringing the bulletin board to the physician" at the time a patient is seen by using Web technology. It is up to the physician and patient to decide whether the patient is referred to the disease interest group. Only after referral to this group and after written informed consent is obtained will the patient be enrolled in any research protocols. Our goal has been to give all patients freedom of choice, a basic tenet of justice in the context of the Belmont principles (3), as well as optimal care. If we rely only on bulletin boards, physician referral, and Web advertising for rare diseases, many patients may be denied the choice of expert care and the opportunity to enroll in clinical trials. We agree with Dr. Jones that scanning of medical records for the sole purpose of identifying research participants is inappropriate without prior patient authorization. As we mentioned in our article, we intend to scan only medical records of patients who have prospectively authorized use of their medical records for research purposes or use other resources to identify patients who may be contacted to provide consent.