Contacts:
Sarah Goodwin

Kathi Ovnic
Holly Korschun
November 1, 1999
EMORY BIOSTATISTICIAN RECEIVES $360,000 GRANT FROM HEART ASSOCIATION

A biostatistics researcher at the Rollins School of Public Health of Emory University has been awarded a three-year, $360,000 grant from the American Heart Association (AHA) to study "Methods for Evaluating Noninvasive Diagnostics Tests."

Andrzej S. Kosinski, Ph.D., associate professor of biostatistics, received one of 12 grants awarded nationally as part of a special $5.6 million Patient Care and Outcomes Research Program set up the American Heart Association Pharmaceutical Roundtable, a partnership between the AHA and a strategic coalition of leaders from the pharmaceutical industry.


The grant will expand statistical methods for evaluating the performance of
diagnostic tests when the true disease status is not available for all patients undergoing a diagnostic test. The decision to perform the gold standard test to establish the presence or absence of disease (for instance, angiography for detecting clogged arteries) is often influenced by the results of less invasive or expensive diagnostic tests and other risk factors. If only data from patients who received the gold standard test are used (i.e. if patients with definite verification of the disease status were used to assess the test performance), the commonly used measures of a diagnostic test performance, sensitivity and specificity, are likely to be biased. This bias is called a verification bias.


Estimates of test sensitivity will often be higher and estimates of test specificity lower than the true values. This is because the population consisting only of patients with verified disease status will likely be sicker than the population in which the diagnostic test is routinely used. Without adjustment for the verification bias one may possibly introduce into the medical practice a diagnostic test with apparent but not true high sensitivity. …

The research funded by the grant will provide clinical investigators with the much needed and improved statistical tools to better design studies evaluating new diagnostic tests and to correctly analyze data from such studies.

Huiman X. Barnhart, Ph.D., associate professor of biostatistics, W. Dana Flanders, M.D., D.Sc., professor of epidemiology, William S. Weintraub, M.D., professor of medicine, and Leslee J. Shaw, Ph.D. associate professor of medicine, are co-investigators on the grant.

The Patient Care and Outcomes Research Program of the AHA Pharmaceutical Roundtable funds a broad range of research that will guide future strategies for reducing cardiovascular disease and stroke, the No. 1 and No. 3 killers in the nation. Funded projects cover a broad range of inquiries into ways to reduce the toll of heart disease, including patient and professional provider behaviors, cost-benefit analysis and patient patient care delivery, as well as more traditional areas of heart research like cardiology and emergency medicine.

###

For more general information on The Robert W. Woodruff Health Sciences Center, call Health Sciences Communication's Office at 404-727-5686, or send e-mail to hsnews@emory.edu Copyright ©Emory University, 1999. All Rights Reserved.