Prescribing propensity: influence of life-expectancy gains and drug costs

To determine whether physician willingness to prescribe drugs for primary prevention of cardiovascular disease is influenced by information about the resultant life-expectancy gains (presented in one of two formats) and about drug costs.

Mailed survey (four versions randomly allocated) asking physicians to assess hypothetical preventive interventions with outcomes expressed either as averaged or as stratified gains in life expectancy (e.g., average gain of 15 weeks, versus 5% of treated patients gain 2 to 6 years, 10% gain up to 2 years, and 85% remain unchanged). Both costs and gains were varied to high and low values. The subjects rated their willingness to prescribe treatments on an 11-point scale from "strongly oppose" to "strongly favor."

Internists randomly selected from two Canadian academic centers (n = 330).

231 usable responses were received (76% of the deliverable questionnaires). For low-yield scenarios typical of very effective primary prevention strategies, the physicians gave significantly higher ratings in response to stratified life-expectancy data than to equivalent averaged data (p < 0.0001). The same trend was not observed for high-yield scenarios (p = NS). The ratings were strongly influenced by cost: 34% of the physicians reversed their treatment decisions in response to a tenfold price increase. Despite this, the rankings of the treatments differed from those expected on the basis of cost-effectiveness criteria (p < 0.0001).

Physician enthusiasm for a therapy designed to prolong life expectancy may be influenced by the format in which that life-expectancy gain is presented. Knowledge of drug cost also affects physicians' choices, but their greater focus on treatment effects causes their rankings to depart from those expected with cost-effectiveness criteria.