New Tool Predicts Patient Risk of Heart Procedures

A statistical model that will predict a heart patient's risk of death from coronary artery interventions such as balloon angioplasty or stenting has been developed by a team of eight hospitals coordinated by the University of Michigan Health System (UMHS). The first study of the model was published in the July 2001 issue of Circulation.

The study showed that unlike previous estimates of such risk, based on general information, the new model allows doctors to customize risk assessment based on a patient's specific medical history. Using a standardized form, the researchers collected information on nearly 20,000 patients, including data on gender, age, medications taken, other medical conditions such as hypertension, and prior operations. Next, they scored the questions, giving points based on responses. The total was then compared to the statistical model, and patients were categorized into certain risk groups based on their score. Data on the processes of care are collected, the patients are followed, and their outcomes after coronary intervention are recorded.

The researchers emphasize that patient data should never be used to deny care but, rather, to give both the doctor and the patient more information on which to base a treatment decision. Patient data will also translate into improved safety for patients because doctors will be able to identify those at risk for certain complications and tailor treatment to diminish those risks. The collaborative effort has resulted in a much larger database and better pool of data than any one center could achieve on its own.

"The consortium is using the extensive database to identify best practices for coronary interventions, disseminate information about them, and foster quality and efficiency improvement among cardiovascular centers,” said Mauro Moscucci, M.D., UMHS director of interventional cardiology and the study's principal investigator.




Related Links:
Univ. of Michigan