Cost, uncertainty, and doctors' decisions. The case of thrombolytic therapy.

OBJECTIVE

To examine the prescribing patterns and predictors of physician choice in the selection of intravenous thrombolytic therapy for the treatment of acute myocardial infarction.

DESIGN

Survey of a random sample of 250 California cardiologists between July 1989 and February 1990.

RESULTS

The adjusted response rate was 66%. For a patient presenting within 30 minutes after the onset of an uncomplicated acute anterior-wall myocardial infarction, 98% of respondents reported that they would prescribe a thrombolytic agent, 79% chose tissue plasminogen activator, and 21% chose streptokinase. Users of tissue plasminogen activator were nine times more likely than streptokinase users to perceive tissue plasminogen activator as superior for early coronary artery recanalization, although most users of tissue plasminogen activator and streptokinase perceived no difference between the two agents for improvement in ejection fraction and mortality. Estimates of side effects did not distinguish the two groups. Users of streptokinase were eight times more likely to practice in a health maintenance organization than were users of tissue plasminogen. For a self-paying patient, 36% of users of tissue plasminogen activator said that they would switch to streptokinase, and 27% would switch for a patient insured by Medicaid.

CONCLUSIONS

These results indicate that physicians place great emphasis on surrogate end points. Physicians vary in their willingness to use more expensive therapies over cheaper alternatives, even when their perceptions of the relative risks and benefits are similar. The patient's insurance status and the provider's practice setting may exert an important influence on doctors' clinical choices.