Can credence advertising effects be isolated? Can they be negative? Evidence from pharmaceuticals

We explore the relative impact of informative versus the credence effects of direct-to-consumer advertising in the market for prescription drugs. In particular, we examine how advertising for statin medications affects the delay between diagnosis and pharmacological treatment for patients with elevated cholesterol, as well as the time between treatment initiation and therapy switching. Duration models of delay to treatment indicate that over some ranges, advertising increases the likelihood of treatment, while over other ranges greater advertising is associated with lower likelihood of treatment. Results for the switching behavior models indicate that on net, advertising tends to both discourage switching between drugs and encourage therapy continuation. However, the component of the advertising that represents credence effects is generally positive, which indicates that as patients' spell of drug usage lengthens, the Food and Drug Administration-required warnings in advertisements induce consumers to switch away from the pharmaceuticals they are using.