Effects of DTC Advertising on R&D Effort

Background: The economic cost of pharmaceutical research and development (R&D) currently exceed $1B per New Chemical Entity (NCE) generated.  Profit maximizing firms continually cull and supplement their research portfolio to increase expected future returns.  In 1997 the regulations with respect to Direct-to-Consumer (DTC) advertising were relaxed and led to rapid increase in DTC advertising – $5.4B by 2006 (although the industry has reduced DTC substantively since then).   Moreover, DTC spending typically has been concentrated in select therapeutic categories.  While literature has examined the role of DTC on patient demand, physician prescription activity and firm revenues, little attention has focused on whether firms shifted their R&D effort more so into therapeutic categories that were amenable to DTC relative to “non-DTC” categories.  

Objectives: This study aims to estimate the elasticity of R&D (as measured by compounds in development by phase or overall) at either the industry or firm level as a function of lagged (or concurrent) DTC expenditure.  We hypothesize that the ability to utilize DTC in select therapeutic categories should led to relatively more compounds under investigation in those categories as DTC serves as a mechanism to increase demand in these select categories, but not others.

Data: DTC advertising data are from TNS Media and are disaggregated by product for 1997-2003 and reported in $US.  R&D data, measured by number of compound in development, were from Citeline PharmaProjects and include 15 therapeutic categories (some with relatively high levels of DTC and others with very little) from 1995-2005 and are further disaggregated by firm and R&D phase (I,II,III).  We classify products from both sources on the basis of World Health Organization (WHO) therapeutic categories.  Additional data on market size and other potential relevant covariates will also be incorporated in future regression models. 

Methods:  At the industry level we estimated the elasticity of DTC spending (lagged by two years) on total number compounds across all phases for each therapeutic category in a negative binomial count regression.  Coefficients reflect incident relative risk.  At the industry-level we have 105 year-therapy cases. Additional analyses will estimate specification at the firm level to determine under what circumstances DTC may serve as a deterrent or motivator for additional R&D effort by firm.  Naturally, the sample sizes will be substantially larger for the firm specific estimates.

Results: The therapeutic category fixed effect specification suggests that for every additional $100 Million in lagged DTC (2 year lag) there are 13% more compounds in development at the industry level (Exponentiated IRR Coefficient = 1.13) .  We consider this an upper bound as future models will account for the time trend which likely will moderate the effects.

Conclusions:  The initial result suggests that category level-DTC may likely have an effect on firm decisions with respect to R&D decisions.