Direct-to-Consumer Advertising of Pharmaceuticals: An Integrative Review - Innovation and Marketing in the Pharmaceutical Industry - page 627

level, from 2002 to 2004 using a hierarchical Bayesian negative binomial model.
They fi nd that DTCA expenditures have a positive and persistent effect on the num-
ber of visits to physicians by newly diagnosed hyperlipidemia patients. In particu-
lar, they report that in this category an additional expenditure of $221 on DTCA
generates one additional newly diagnosed patient visit. Liu and Gupta ( 2011 ) con-
sider alternative explanations for the contradictory fi ndings. These include the con-
jecture that having gained experience with DTCA, fi rms may be using advertising
more effectively in the later period considered by Liu and Gupta ( 2011 ), or differ-
ences in model specifi cations and data (regional vs. national) between the two stud-
ies may be driving the different results.
Using data pooled across 151 drug classes, Iizuka and Jin ( 2005 ) report a positive,
statistically signifi cant effect of DTCA on patient visits. In particular, they fi nd that
every $28 increase in DTCA leads to an additional drug visit within 1 year. The
lower amount relative to Liu and Gupta ( 2011 ) is possibly due to averaging across
a large number of drug classes that differ in effectiveness of DTCA. Cantor ( 2010 )
collected individual measures of advertising exposure to print advertising and specifi es
a model in which the variation in ad exposure used to estimate the key parameters
is orthogonal to individual unobservable variables that may drive both ad exposure
and healthcare seeking behaviors. The study confi rms that DTC advertising
does have some infl uence on the likelihood of an individual to visit a healthcare
practitioner.
Using data from the Medical Care Expenditure Panel Survey of individuals who
have no previously diagnosed medical condition, Hosken and Wendling ( 2010 ) fi nd
that drug advertising is an important determinant of an individual's decision to go for
a “check-up” visit, which by defi nition is designed to diagnose conditions that patients
may be unaware they have. The authors interpret their fi nding as affi rming the “infor-
mative” role of advertising, i.e., advertising provides information to consumers and
thereby increases demand for the category. They also fi nd large differences in the
effects of advertising across demographic groups. For instance, women are more
responsive than men, blacks and whites are more responsive than Hispanics, and
the highly educated are more responsive than the less educated.
Bradford et al. ( 2006 ) found that DTC advertising of COX-2 inhibitors Vioxx
and Celebrex increased the fl ow of osteoarthritis patients to physician offi ces each
month. More recently, Bradford et al. ( 2010 ) examined using a duration model how
DTC advertising affects the delay between diagnosis and pharmacological treatment
for patients suffering from a common chronic disease. They fi nd that on average
television advertising had the effect of lengthening the time patients wait to begin
therapy after being diagnosed. However, this average is composed of reduction in
waiting time among patients who are good clinical candidates for the therapy and
an increase in time for poor candidates. This fi nding supports the idea that advertising
matches patients with treatments, and therefore is informative.
Overall, recent evidence for DTCA's effect on increasing the number of patients
seeking diagnoses and treatment, as refl ected in number of patient visits to physician
offi ces, is quite strong and robust across therapeutic categories and studies.