[Ip-health] Reporting on $802 number -- clinical trials costs

[James Love]

There are lots of issues that one could look at in today's press reports
of the Tufts estimate of $802 million as the "average" cost of
developing a new drug.  One quick point is that almost no one questioned
directly whether or not the data from the 10 companies would be a sound
statistical basis for an "average" cost estimate, and few reports
mentioned the paucity of information that was released to support the
estimate.  Of particular interest to CPTech was the way that reporters
handled the issue of the costs of clinical trials.  Every news report
below cites the rising costs of clinical trials as the primary reason
for the escalating costs of drug development.  *None* of the news
reports mentioned that the Tufts/Merck/PhRMA/Boston Consulting event
refused to disclose how much money clinical trials cost.  The AP,
Reuters (in some editions) and the Washington Post mentioned the
contrary evidence from the IRS, that companies were reporting
expenditures of only $7.9 million on clinical trials for orphan drugs
(per approval) on income tax returns, less than 1 percent of the $802
million estimate.  The New York Times and the Boston Globe ignored this,
even though the Tufts researchers said the cost of clinical trails was
the primary reason for the high cost of drug development.   

The most interesting quote was from Gilmartin, the CEO of Merck, who
said that the "typical" clinical trial involves 4,000 patients, quoted
in the Washington Post.  This was one of the few empirical figures, and
it did not come from the Tufts researchers themselves.   The 4,000 is
more than 3 times the average number of patients in trials cited by the
FDA for HIV drugs, and nearly 7 times the size of the trials for the
1999 orphan drugs.  It may be "typical" for some products, but not at
all typical for others, and in any case, it is difficult to see how even
a 4,000 patient trial number gets you even close to $802 million, unless
one assumes huge per patient costs that are completely out of line with
the independent data from CROs, who will freely quote clinical trials
costs.   At even $10 k per patient for a trial -- two to three times a
typical CRO estimate and 50 percent higher than the NCI's reported
average costs, this would be $40 million, or 5 percent of the $802
number.  Doubled for Phase I, II and III risks (say .2. .45 and .7),
this still isn't that large, and of course, it certainly isn't an
"average."   (And of course, for orphans, the US government pays for 50
percent of the cost through the Orphan Tax credit).

   Jamie

Mentioned IRS data on clinical trials costs

 NYT    	no
 WP     	yes
 AP		yes
 Reuters  	in late editions
 Boston Globe	no


 http://www.nytimes.com/2001/12/01/business/01DRUG.html
NYTimes December 1, 2001
Research Cost For New Drugs Said to Soar
By ROBERT PEAR

Dr. Joseph A. DiMasi, the principal author of the study, attributed much
of the increase in the cost of drug development, beyond inflation, to
the rising costs of testing new drugs in humans. Drug companies recruit
thousands of people around the world to participate in such clinical
trials before drugs are marketed in the United States.

   [does not mention contrary data from IRS on costs of clinical trials]

http://www.washingtonpost.com/wp-dyn/articles/A42120-2001Nov30.html
Price Tag for a New Drug: $802 Million
Findings of Tufts University Study Are Disputed by Several Watchdog
Groups 
By Ceci Connolly
Washington Post Staff Writer
Saturday, December 1, 2001; Page A10 

The study's author, Joseph DiMasi, said that high failure rates, the
exorbitant cost of clinical trials and the amount of time needed to win
government approval -- about 12 years -- were the primary reasons drug
development has become such a pricey business.

 ....

Facing an increasingly skeptical public, the pharmaceutical industry is
quick to note the financial risks inherent in bringing new medications
to market. One in five drugs is approved for sale in the United States,
according to several experts. The typical clinical trial involves 4,000
people today, compared with 1,300 in the 1980s, Gilmartin said.

    [does mention contrary evidence from IRS on costs of clinical
trials]


http://biz.yahoo.com/rf/011130/n30295250_1.html
Friday November 30, 6:19 pm Eastern Time
Drug development costs $802 mln on average--study
(UPDATE: adds consumer group, Merck comments paragraphs 14-15, 17)
By Lisa Richwine 
(Reuters)

The major reason for the jump was a significant rise in costs for
clinical trials in people.

``The size and complexity of the clinical trials has increased,''
Kenneth Kaitin, director of the Tufts Center, said in an interview.

He said new regulatory requirements added costs. Also attempts to sell
more products globally required marketing approval in multiple markets.
More companies are also testing drugs for chronic diseases, requiring
longer testing.

    [Late editions do mention contrary evidence from IRS on costs of
clinical trials]

http://dailynews.yahoo.com/h/ap/20011130/hl/drug_development_costs_1.html
Friday November 30 5:31 PM ET 
Activists Slam Drug Dev. Cost Study
By THERESA AGOVINO, AP Business Writer 

DiMasi says that clinical trials are the main reason for the price hike.
He says more drugs are being tested which creates competition for
patients so recruiting is more expensive. 

    [does mention contrary evidence from IRS on costs of clinical
trials]

http://www.boston.com/dailyglobe2/335/business/R_D_costs_for_drugs_skyrocket_study_says+.shtml
R&D costs for drugs skyrocket, study says
Tufts Center estimates amount up threefold from a decade ago
By Naomi Aoki, Globe Staff, 12/1/2001

Kaitin attributes the staggering increase in the cost of developing a
drug in large part to the soaring costs of human clinical testing. The
size of clinical trials has steadily increased in the past two decades
at the same time that volunteers have become more scarce. As researchers
learn more about the potential hazards posed by drugs, companies are
also required to run a growing battery of safety tests.

    [does not mention contrary evidence from IRS on costs of clinical
trials]