Boston researchers have developed a mathematical model that uses analytics to predict adverse drug events that could be discovered in the future, according to a study published in the journal Science Translational Medicine, Modern Healthcare reports (McKinney, Modern Healthcare, 12/22/11).
Methodology
For the study, researchers at Children's Hospital Boston collected 2005 data on 809 medications and 852 adverse drug events. The data were available on several websites and the reference database Lexicomp, which contains medication risk information from drug packaging inserts.
Researchers then ran the 2005 drug data through the mathematical model and compared the results with the information known about the drugs by 2010 (Shurkin, U.S. News & World Report, 12/23/11).
Study Findings
The study found that the mathematical model successfully predicted 42% of the adverse drug event relationships that were unknown in 2005 but were identified over the next five years (Modern Healthcare, 12/22/11).
The model also helped researchers predict which adverse drug events would not occur with 95% accuracy (U.S. News & World Report, 12/23/11).
Implications
Ben Reis -- study co-author and leader of the predictive medicine group at Children's Hospital Boston's Informatics Program -- said, "This approach allows us to make the important transition from detection to prediction." He added, "We can potentially identify a dangerous drug side effect early on, instead of having to wait for sufficiently many patients to be affected by it in order to allow for detection" (Modern Healthcare, 12/22/11).