A newly published study in PLoS Medicine provides very useful data on a long-contested question:
Marc-Andre Gagnon and Joel Lexchin teamed up to determine how much the U.S. drug industry actually spends on marketing as compared to R&D. They dismiss, as do most knowledgeable outsiders, the industry's own reports that it spends more on R&D than on marketing.
Let me provide some personal background. When I wrote HOOKED, I puzzled over the conflicting figures that Gagnon and Lexchin juggle with. I teamed up with two health economists and we spent a couple of years trying to develop a complete financial accounting of all the relationships between medicine and Pharma (ultimately failing). Nonetheless, bits and pieces emerged that seemed strongly to me to suggest that the "official" figures--usually those provided by IMS Health--must be far too low. (Gagnon and Lexchin used 2004 as their data year, and in 2004, IMS reported the total expenditure for pharmaceutical marketing in the US to be $27.7 billion.) For example, a few typical drug reps provided me with estimates of their annual budgets for physician gifts. If one were to multiply those figures by the total number of drug reps in the US, the result markedly exceeded any IMS estimates.
Therefore, when Marcia Angell claimed in her 2004 book, The Truth About the Drug Companies, that the real marketing costs were as high as $54 billion, I was quite ready to believe her estimates. In HOOKED, I hedged and estimated within the $40B range, lest I be thought to be too rabidly anti-industry.
If all you want is the bottom line, Gagnon and Lexchin estimate the total at $57.5 billion, which is 24.4% of total drug sales; R&D expenses account for 13.4% of sales by contrast. This works out to $61,000 spent on promotion for every US physician.
The details--Gagnon and Lexchin were able to contrast the IMS data with data from another firm, CAM, that gets information direct from physicians (while IMS gathers its data from the firms). Moreover, CAM has had the advantage of being able to compare its figures with industry data--the industry has been willing to open its books, at least partially, to CAM, which is quite an unusual situation. That allows CAM to estimate how much promotional activity is not captured in its physician data, and to make corrections accordingly.
Gagnon and Lexchin found that for 2004, the latest year for which they could get data from both companies, IMS and CAM agreed on how much was spent on direct-to-consumer ads ($4B) and medical journal ads ($0.5B). They disagreed markedly on the amount spent on detailing to physicians--$7.3B according to IMS, $20.4B according to CAM. The major difference seemed to be that IMS counted as the cost of a detail visit the rep's salary plus transportation. CAM added in the expenses of the entire management apparatus that backed up the rep, plus the costs of the promotional materials that the rep distributed. It seems obvious that the true costs of detailing are better represented by the CAM figures.
By contrast, CAM came up with a markedly low estimate for the value of "free" drug samples distributed to physicians, $6.3B vs. the $15.9B per IMS. Here the explanation was simple-- CAM used the average wholesale price while IMS used the retail cost of the same drugs. CAM also likely underreported the quantity of samples. Gagnon and Lexchin elected to use the IMS figures for two reasons--IMS got the data on quantity of samples direct from the companies; and the companies themselves, when they report on drugs that are given away for charity programs, use the retail price as the value of the drug. Since the industry itself favors the higher figure, Gagnon and Lexchin elected to go with it.
Angell based her very high estimate on the annual reports that the large drug firms make to the SEC. Gagnon and Lexchin offer several reasons why those reports are probably not accurate for our purposes. For instance, "marketing" costs for SEC purposes includes the costs of packaging and shipping the manufactured drugs from the plant to the pharmacy.
I should add that the estimate I provided in HOOKED for the cost of drug marketing per US physician was around $13,000. This is a many-years-old figure and so was most certainly too low. The $61,000 figure seems much more probable.
A final comment--think about what it means that the debate over the amount of money spent on marketing by Pharma has been going on for many years, and we are still trying to get the accurate figures on these expenses. There's a huge debate today about the price of gasoline; but I am not aware that there is any mystery over what a gallon of crude oil cost on today's market, or what Exxon's profits were in the most recent quarter. The fact that so many of these basic data points are simply not known, or are still contested, highlights the lack of transparency within which the drug industry does it business.