A New Prescription for Drug Studies
What can avatars tell medical researchers about the health of real people? The same information, it turns out, that testing large human populations would, says UWM researcher Peter Tonellato.
A professor of public health, Tonellato uses avatars, mathematical models and simulations to develop reliable medical guidance at a fraction of the cost of a large-scale human study. Currently he is testing optimal dosage levels of the common blood clot-preventing drug warfarin, using a representative population of virtual patients created from real-life health records.
Tonellato and his colleagues in UWM’s Laboratory for Public Health Informatics and Genomics created the avatars to find the drug’s “sweet spot,” a dosage that prevents clots while minimizing the risk of side effects. The work is being done in conjunction with Aurora Health Care’s Patient-Centered Research division.
“This type of research teases out information about the intertwined impact of genetics and environment,” says Tonellato. For example, genetic research has shown that African Americans generally metabolize warfarin more quickly than other ethnic groups. Asian Americans generally metabolize it more slowly.
Not every individual in those groups metabolizes the drug the same way, so it’s still important to consider individualized genetic testing in certain situations, says Tonellato. But improved modeling can help identify the likely value of the genetic test on improved health care outcomes and, as a result, reduce the need for this comparatively expensive option.
Public health decisions must include not only science, but also cost, policy and health care outcomes, says Tonellato. With his research, Tonellato aims to provide guidance that takes all of these factors into account.
“We can then have a discussion about not only what is scientifically proven and clinically validated, but also what is feasible and practical for the community to use effectively.”