New ways to assess drug benefits can help cut health care costs

(American Statistical Association) Newly approved drugs are compared to either a placebo or to one standard of care, with little information about their effectiveness to the other available treatment options. The statistical performance of a commonly used method to address this challenge has yet to be extensively studied or reported. David Cheng of Harvard's T.H. Chan School of Public Health and researchers from Analysis Group have identified conditions under which this method is valid.