Conditional release in Canada is feasible but would not be easy due to the requirement for all being evaluated to be listed in all formularies during the conditional release phase. The implementation would require new legislation (probably); a change in philosophy away from a system concentrated on premarketing evaluation to one with a greater focus on the postmarketing period and an acceptance by health professionals, to whom clinical trials are the ‘gold standard’ that population-based epidemiological methods can be reliable; adequate funding from federal and provincial governments and pharmaceutical companies; enthusiastic cooperation from pharmacists, physicians, health care institutions and patients; collaboration between provincial and federal governments and private insurers; better methods to measure and evaluate effectiveness; and better techniques to evaluate cost effectiveness that truly assess the relations between cost, effectiveness and safety. The availability of good data does not mean better decisions unless one has the appropriate analytical tools.

Given these conditions, conditional release would provide the information to allow real cost effectiveness and safety evaluations in Canada instead of the current inadequate predictions based on efficacy and safety data from clinical trials, small premarketing studies, and the present passive postmarketing surveillance system. Although the ideas proposed here may not meet with universal approval, they should be discussed because it is important that academics, and drug approval and monitoring agencies work together to develop active systems to improve the postapproval evaluation of effectiveness, safety and cost effectiveness of new drugs in Canada, as the recent concern expressed about the fast-tracking of nevirapine and zanamivir demonstrates.