Effects of prescription drug warning messages on perceived risk : an elaboration likelihood approach

Responding to relaxed regulation, pharmaceutical manufacturers recently have instituted direct-to-consumer (DTC) prescription drug-advertising, bypassing the physician. The resulting debate over the propriety of such DTC advertising has centered, in part, on the effects of DTC advertising upon untrained consumers, particularly the effectiveness with which those ads convey risk information to readers/viewers. This study examined relations among several important risk communication variables, within the framework of the Elaboration Likelihood Model of persuasion. How are the complexity of risk information in an advertisement and the reader's prior product ·knowledge related to important "central" responses such as product-relevant thought, perceived risk and attitude toward the product? How do "peripheral" elements such as the reader's attitude toward the advertisement itself and perceived communicator credibility affect processing? Data were collected from 151 student subjects in an experimental setting with prior product knowledge and message complexity manipulated for a fictitious prescription drug. Through Linear Structural Relations (LISREL) modeling, causal effects among the variables were estimated. Low-knowledge readers perceived greater risk to themselves as the complexity of the risk information increased, though high-knowledge readers showed no relation between complexity and risk. High-knowledge readers, however, were more likely to think about the product as their attitudes toward the ad vehicle grew more positive, a relation not manifested by low-knowledge readers. Knowledge had no effect upon some relations, with positive attitude toward the ad causing positive brand attitudes, high perceived credibility causing reduced perceived risk and high perceived risk causing negative brand attitudes. The ELM was supported by the data, with low-knowledge readers showing stronger effects through peripheral processing routes in the causal model and high-knowledge readers showing stronger effects through central routes. However, high-knowledge readers' attitude toward the ad appeared to determine processing intensity, a finding not predicted by the ELM. Alternative interpretations of "Attitude Toward the Ad" are discussed in light of this finding. These results have important implications for policy makers and other parties involved with DTC prescription drug advertising. Messages should be crafted so they are not unnecessarily complex, given the effect of complexity upon perceived risk. Attractive ad vehicles appear to enhance product-relevant thought among high-knowledge vi readers, while they lead to positive attitudes through peripheral routes for low-knowledge readers.