Several high-profile prescription drugs have been withdrawn from the U.S. market in the last decade, yet there is no direct evidence of how a prescription drug withdrawal affects consumers' use of remaining drugs within the same therapeutic class. In theory, remaining drugs in the therapeutic class could enjoy competitive benefits or suffer negative spillovers from the withdrawal of a competing drug. Using the Medical Expenditure Panel Survey, we test for spillovers following prescription drug withdrawals in six therapeutic classes between 1997 and 2001. Results vary, but we find stronger evidence of negative spillovers than competitive benefits. We conclude with a discussion of the characteristics of drugs and classes that may influence how remaining drugs are affected by a withdrawal in the class.