The effects of direct-to-consumer drug advertising on statin use depends on the person's experience with the statin
Research Activities, February 2012, No. 378
Direct-to-consumer (DTC) advertising of statins has different effects on product switching depending on whether the person is just being prescribed a statin, has been on a statin for only a short time, or has been using a statin for a long time, concludes a new study. Economic models of the impact of DTC ads over time have failed to include drug quality, which can be observed directly for statins by the degree of reduction over time of blood low-density lipoprotein (LDL) levels, according to the researchers. Using clinical data together with national and local costs for advertising each of the three major statins (Lipitor®, Pravachol®, and Zocor®), the researchers tested econometric models of beginning a statin after diagnosis and of switching between the statins as a function of length of product use.
They found that the longer a patient is on one of the drugs, the weaker is the effect of DTC advertising in providing information on the drug's experience characteristics. In other words, the patient substitutes his or her own experience for that conveyed in the ads. In contrast, the FDA-required verbal listing of possible side effects increase the likelihood of the patient switching away from the statin in use over time, because the patient has to rely on information from DTC ads on negative side effects. Finally, the researchers confirmed the importance of statin quality (decline of LDL-cholesterol levels from baseline to recommended levels) in reducing the likelihood of a patient deciding to switch to another lipid-lowering drug.
The researchers developed the econometric models based on data on more than 600,000 patients extracted from the electronic medical records of the Practice Partner Research Network (90 primary care practices in 33 cities across the United States). Local and national monthly DTC advertising expenditures were obtained for each of the three brands for 1998 through 2004. The study was funded in part by the Agency for Healthcare Research and Quality (HS11326).
More details are in "Can credence advertising effects be isolated? Can they be negative? Evidence from pharmaceuticals," by W. David Bradford, Ph.D., and Andrew N. Kleit, Ph.D., in the Southern Economic Journal 78(1), pp. 167-190, 2011.