Factors Influencing Physicians' Willingness to Substitute Generics For Brand-Names when Prescribing Antimicrobial Drugs

Physicians often continue toprescribe brand-name drugs to theirpatients even when less expensivegeneric equivalents are available. Ina 1994 study, Judith Hellersteinadvances two hypotheses to explainthis behavior. First, doctors mayconsciously conclude that certainbrand-name drugs impart a relativetherapeutic benefit that outweighstheir higher cost. Second, physiciansmay choose to prescribebrand-name drugs without evidenceof therapeutic superiority if neitherthey nor their insured patients bearthe increased cost of these drugs.The second hypothesis implies thatmoral hazard is evident in physicians'prescribing behavior. Hellerstein'sfindings support neither hypothesis,but her estimation equation does notexplicitly capture the effects ofbrand-name/generic pricedifferentials and information diffusionon the probability of genericprescription. The author adaptsHellerstein's theoretical model to amodified estimation equation thatincorporates these effects and uses itto create new estimates based ondata on antimicrobial prescriptionsfrom the 1994 National AmbulatoryMedical Care Survey (NAMCS).Unexpectedly, the results appear toaffirm both hypotheses. Theevidence for moral hazard isparticularly strong, as self-payingpatients are significantly more likelythan patients with Medicare orprivate insurance to be prescribedthe generics that are cheapestrelative to their brand-namecounterparts. The author also findsthat certain popular antimicrobialdrugs such as amoxicillin andsulfamethoxazole/trimethoprim areprescribed in the same form (genericor brand-name) by most doctors tomost patients. The market powerexhibited by these preferred formsleads the author to conclude thatthey are "brands" in the economicsense.