Critique of the DiMasi/Tufts Methodology and Other Key Prescription Drug R&D Issues

  • DiMasi’s figure is the average R&D cost for "self-originated new chemical entities" (NCEs) – which are the most expensive class of new drugs. DiMasi admitted that none of the 68 drugs in his study received any government financial support. This makes his sample non-representative of all drugs brought to market. (Not all new drugs brought to market are NCEs and fewer still are "self-originated NCEs," which means they were created entirely in-house by a drug company.)
  • The R&D costs for all new drugs are detailed in Section II of Public Citizen’s July report. Public Citizen found that the R&D costs for all new drugs brought to market between 1994 and 2000, based on PhRMA’s data, ranged from $71 million to $118 million. The R&D costs for NCEs were $150 million, on average.
  • DiMasi’s new $802 million estimate for self-originated NCEs (like his original $231 million figure) does not represent what companies actually spend to discover and develop new molecular entities. It includes the expense of using money for drug research rather than other investments (known as the "opportunity cost of capital"). DiMasi also does not account for huge tax deductions that companies get for R&D. Therefore, he substantially overestimates net expenditures on R&D.
  • The opportunity cost of capital amounts to 50 percent of DiMasi’s total figure. In its analysis of the 1991 DiMasi study, the OTA subtracted opportunity cost to get a pre-tax R&D cash outlay of $127 million for every new drug (including failures). In DiMasi’s new study, the pre-tax cash outlay is $403 million per drug.
  • It should be noted that five of the seven previous R&D cost studies that DiMasi references in his 1991 study did not include opportunity cost of capital in their calculations.
  • According to the OTA’s 354-page report on pharmaceutical R&D: "The net cost of every dollar spent on R&D must be reduced by the amount of tax avoided by that expenditure." The tax deduction reduces the cost of R&D by the amount of the corporate marginal tax rate (currently 34 percent). This means, in effect, that every dollar spent on R&D costs $0.66.
  • After subtracting tax deductions and the opportunity cost of capital, OTA found that DiMasi’s after-tax R&D cash outlay for a new NCE was $65.5 million (in 1990 dollars). That is the estimate of how much the drug companies in DiMasi’s 1991 study actually spent on new chemical entities, including failures. Public Citizen inflated this figure to year 2000 dollars and found that actual after-tax cash outlay for NCEs (including failures) was $110 million – based on DiMasi’s data.
  • Applying the same methodology to the new DiMasi study, the average after-tax cash outlay for a self-originated NCE is approximately $240 million.
  • Clinical trial costs account for the largest portion of DiMasi’s new estimate. DiMasi’s new study puts out-of-pocket clinical trial costs at $282 million, based on the NCEs in his study. His estimate is four times more than an estimate of clinical trial costs ($75 million) published by the Congressional Research Service in April 2001.
  • Moreover, DiMasi’s estimate of clinical trial costs greatly exceeds the drug industry’s own data on the subject. PhRMA’s own survey of 1999 R&D expenditures states that clinical trial costs account for 29 percent of all R&D costs. (See Table 6, "Domestic U.S. R&D By Function" in "Pharmaceutical Industry Profile 2001.") Yet DiMasi’s study says clinical trials account for 70 percent of all R&D costs ($282 million out of $403 million total out-of-pocket expenditures for each drug).
  • Evidence suggests that the time required to conduct clinical trials on new drugs is also decreasing. A January 2000 report by the Tufts Center for the Study of Drug Development stated that clinical testing time declined by 19 percent for drugs approved in 1996-1998 when compared with drugs approved in 1993-1995.
  • The advent of technologies such as genomics and combinatorial chemistry, has led, according to investment analysts at Lehman Brothers, "to a growing school of thought that the cost of discovering new biological targets and the cost of creating drug leads is falling." The Boston Consulting Group predicts that drug companies will increase the number of new drugs (NCEs) they produce annually by five-to-tenfold by the year 2003.
  • Industry R&D risks and costs are often significantly reduced by taxpayer-funded research, which has helped launch the most medically important drugs in recent years and many of the best-selling drugs. According to the NIH, taxpayer-funded scientists conducted 55 percent of the research projects that led to the discovery and development of the top five selling drugs in 1995 (see Section III of report). PhRMA has not challenged the NIH document.
  • Drug industry R&D does not appear to be as risky as companies claim. In every year since 1982, the drug industry has been the most profitable in the United States, according to Fortune magazine’s rankings. During this time, the drug industry’s returns on revenue (profit as a percent of sales) have averaged about three times the average for all other industries represented in the Fortune 500. It defies logic that R&D investments are highly risky if the industry is consistently so profitable and returns on investments are so high. (See Section V)
  • Drug industry R&D is made less risky by the fact that only about 22 percent of the new drugs brought to market in the last two decades were innovative drugs that represented important therapeutic gains over existing drugs. Most were "me-too" drugs, which often replicate existing successful drugs. (See Section VI)
  • In addition to receiving research subsidies, the drug industry is lightly taxed, thanks to tax credits. The drug industry’s effective tax rate is about 40 percent less than the average for all other industries, according to the Congressional Research Service. (See Section VII)
  • Drug companies receive a huge financial incentive for testing the effects of drugs on children. This incentive, called pediatric exclusivity and which Congress recently voted to reauthorize, amounts to $592 million in additional profits per year for the drug industry – and that’s just to coax companies to test the safety of several hundred drugs for children. It is estimated that the cost of such tests is less than $100 million a year. (See Section VIII)