Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

In October 2004 the FDA directed manufacturers of many antidepressant drugs to include special warning labels regarding risks of side effects leading to suicidality among children and adolescents. This directive provides a useful natural experiment to examine the impact of negative information and black box warning labels on prescription drug use. Using Medicaid state drug utilization data from the Centers for Medicare and Medicaid Services (CMS), we examine the timing and magnitude of effects of the warning labels on antidepressant use. These effects are of interest because the warning label came after a long period of clinical information streaming into the physician and pharmacy market. Thus, if the warning label per se affects drug sales this may be attributed to information consolidation, physician over-reaction or to patient inputs. We also test for effects of the mandatory label on market shares of drugs within the class. The warning label affects a broad set of antidepressant drugs, even though clinical information about suicidality is very unevenly distributed across the drugs. Thus, we examine the relative importance of the warning label and negative clinical information in determining drug market share. Our analysis is undertaken using drug by state by quarter data for all classes of mental health drugs as defined by the Lexicon database produced by Multum, over the period 2001:1 through 2005:4. Using these data and after controlling for a number of additional influences, we are able to ascertain the effects of the black box warning label on the use of this important class of drugs.