Could conditional release of new drugs:2

An alternative source of information is the patient, a suggestion that received criticism, when included in one of the proposals 20 years ago , on the grounds that its value would ‘be outweighed by the potential harm to the doctor-patient relationship’ . Attitudes of patients toward physicians and the health care system have changed during the intervening years, with a desire by many patients for empowerment in theirtreatment. Consequently, the collection of information directly from patients may be more acceptable and successful, and for some of the elements required to evaluate effectiveness, probably essential. Patients are the source of information in McMaster University’s Pharmacy Medication Monitoring Program , which has achieved a good level of response (80% or more), but it is worth noting that the data collection is time-consuming and requires a significant number of staff. The self-monitoring of adverse events by patients has been pioneered with success over the past 10 years by the Centre forMedicationsMonitoring at the University of Texas . The combination of patient and physician data in a conditional release program would provide a comprehensive profile of the use of a drug and its outcome.

In some circumstances, effectiveness could be probed in other ways, such as decreases in hospitalizations or mortality. This may be possible for without the use of which the outcome would be death or illness of sufficient severity normally requiring hospitalization. The increasing availability of population-based health care utilization databases broadens the opportunity for this type of analysis.