Date and Time
January 11, 2006
1:30 PM - 3:00 PM

Availability
Open to the public
No RSVP required

Speaker
Mark A. Hlatky - Stanford University

Randomized trials are the reference standard for evaluating the effectiveness of clinical interventions. Economic outcomes are now included in many contemporary trials, adding an additional dimension to the assessment of interventions, but data collection and analysis pose methodologic challenges.

Data on medical resource utilization and cost can readily be collected alongside clinical data. The cost of planned interventions can be measured with reasonable accuracy, but costs due to clinical complications are more difficult to measure reliably. The total cost of an intervention often depends critically on these relatively infrequent, yet costly, adverse outcomes. The additional cost of effective therapies may be partially, or even completely, recouped by prevention of complications that are expensive to treat.

Newer therapies are typically more expensive and more effective than older therapies. The most important question in economic assessment is whether patient outcomes are improved sufficiently to justify the extra expense. Cost-effectiveness analysis has been applied to clinical trials to help gauge the value provided by a new therapy. The cost-effectiveness of an intervention compared with an alternative is defined as the ratio between the incremental clinical benefits and the incremental costs. Ironically, it is the measurement of effectiveness that provides the greatest challenge to cost-effectiveness studies in clinical trials. Cost-effectiveness analysis uses life expectancy or quality-adjusted life-years as the measure of effectiveness. This common metric facilitates the comparison of the intervention under study against standard benchmarks for accepted therapies. The follow-up period in most clinical trials is generally long enough to measure the added cost of therapy, but not to capture the full benefits of treatment. The limited time horizon of clinical trials makes it necessary to use a model to extrapolate the observed effect of treatment and project the increase in life expectancy. The resulting cost-effectiveness ratio is sensitive to assumptions about the long-term efficacy of treatment, particularly whether the treatment effect will continue or dissipate over time.

While the use of cost-effectiveness models falls outside the strictly empirical, within-trial analysis framework that is embraced by most clinical trialists, it provides an explicit approach to assessing whether the intervention under study provides a clinically meaningful improvement in outcome that is worthwhile.