Pharmaceutical Detailing Elasticities: A Meta-Analysis - Innovation and Marketing in the Pharmaceutical Industry - page 533

earlier meta-analyses (e.g., 0.16 in Bijmolt et al. 2005 , and 0.29 in Tellis 1988 ).
These results are summarized and discussed below.
The results with respect to the two market-type characteristics are as follows.
Product life cycle ( PLC ) stage . A key advantage of detailing is face-to-face exchange
between physician and sales rep that can address the former's questions and objec-
tions. This advantage tends to be more pronounced in the case of newer pharma
products. For example, Narayanan et al. ( 2005 ) found that sales calls for new phar-
maceuticals are more informative and persuasive , resulting in higher average per-
sonal selling elasticity values in the launch phase than those in the later stages of the
life cycle.
In our analysis, we dummy code the PLC stage variable as being 1 if the product
is in the early stage of the PLC and 0 if it is in the late stage of the PLC. Coders were
instructed to look for explicit statements in the paper about whether a product was
in the early or late stage of the PLC. If such statements were not provided, the cod-
ers used a rubric where they inferred the stage of the PLC based on the date of
launch of the drug. If detailing elasticities were estimated using data which were
predominantly generated over a time horizon less than 5 years after date of launch
of the product (the date of launch is generally provided in the paper), the PLC stage
was coded as early; otherwise coded as late (see also Footnote 2). Table 18.3 indi-
cates a shift in setting from late to early stage has a signifi cant main effect ( b = 0.742,
p <0.01) as well as interaction effects with several other variables on estimated
detailing elasticity. Therefore, all of these effects must be interpreted together to
assess the overall effect of a shift in the market setting from late to early stage. We
accomplish this by calculating the difference in detailing elasticities between early
and late stages of the product life cycle, holding all other moderators at their mean
value. We fi nd that, holding all moderators at their average level, detailing elastici-
ties at the early stage are higher than those in the late stage by 0 . 125 .
Geographic setting — USA vs. Europe . In pharma promotion, Europe is more heav-
ily reliant on information provided through sales forces than in the USA where
direct-to-consumer advertising is allowed (Fischer and Albers 2010 ). There is also
more saturated sales force coverage in the US pharmaceutical market, e.g.,
Chintagunta and Desiraju ( 2005 ). Therefore, we expected that detailing elasticities
will be lower in US settings than in European settings.
Accordingly, we dummy code the geographic setting variable as being 1 if the
elasticity stemmed from the USA and 0 if the elasticity pertained to Europe. As we
see in Table 18.3 , we found no main effect for geographic setting, i.e., detailing
elasticities between Europe and USA do not appear to be different ( b = 0.019, not
signifi cant (ns)). However, we found a signifi cant interaction between stage of the
PLC and geographic setting. Since we found other interaction effects, we calculated
the difference in detailing elasticities between the USA and Europe, holding all
other moderators at their mean value. We fi nd that detailing elasticities in Europe
are higher than those in the USA by 0 . 21 .
One possible reason for this is the difference in the regulatory profi les of coun-
tries in Europe vs. the USA. According to a classifi cation based on fi ve regulatory