[Ip-health] How to get to $2.6 billion in 2014 , comment on the Tufts cost study

Jamie Love james.love at keionline.org
Tue Nov 18 11:54:31 PST 2014


I fixed a couple of typos.

On Tue, Nov 18, 2014 at 2:46 PM, Jamie Love <james.love at keionline.org>
wrote:

>
> From :  http://keionline.org/node/2127
>
>

DiMasi estimates an "Average out-of-pocket cost of $1,395 million" but what
this figure represents is an estimate of the risk adjusted outlays on drug
development, including the actual out of pocket costs claimed for his
secret sample of drugs, adjusted for risk, then automatically adding 44.5
percent of the risk adjusted number of pre-clinical expenses. To this he
adds "time costs (expected returns that investors forego while a drug is in
development) of $1,163 million".

The words "average out-of-pocket" are misleading, for the following reason.
They give the impression that a company actually spent $1.4 billion on some
drug, and it was risky. But no company spent $1.4 billion a drug. The
number is a combination of what they spent on the drug that was approved,
and money spent on projects that failed, at least in theory. Here is my
guess for what the elements of the $1.4 billion look like, based upon the
slides that DiMasi presented today. I believe a reported outlay of $1 on
clinical testing was adjusted for risk, so it is now counted as $3.5. To
this, DiMasi adds and automatic 44.5 percent more, or $1.57, for
pre-clinical spending. Now the $1 in clinical testing is counted as $5.06,
to reflect the risk adjusted outlays on both clinical and pre-clinical
testing. So, the $1.395 billion in "out-of-pocket" expenses probably looks
something like $276 million for clinical trial expenses, $690 million is
risk adjustments for the clinical investment, bringing the total to $966
million. Add in 44.5 percent, or $430 million for pre-clinical expenses,
which may or may not have been incurred, and you have $276+$690+$430 =
$1.395 billion (rounding suppressed)

So which number sounds more impressive? "$276 million, plus adjustments for
risk?" Or $1,395 billion in "average out of pocket" expenses? This is one
way that the study was designed to confuse people who do not have the
capacity or time to unpack the numbers.

Then there is the issue of the cost per patient. If the average number of
patients in trials is 5,000, the average cost per-patient will be $55,200
per patient, a number that is not credible as an average, based upon data
from third party sources. But now I'm guessing. Joe DiMasi knows what these
numbers actually are. Eventually, we will see them, after the $2.6 billion
number is burned into our brains. Tufts says the full study will be
available, but not until some time next year.

But what else can I say about $276 million as an average for clinical
testing? The entire IRS outlay for the orphan drug tax credit was less than
$650 million in 2010, and that year the FDA gave 195 Orphan designations
and 14 marketing approvals. If clinical testing outlays were really as high
as DiMasi says, the IRS outlays, while growing, would have been nearly an
order of magnitude larger.

And finally, there is the final boost in the numbers from the "time costs."
This is $1.163 out of 2.558 billion. How does the cost estimate now look?
You $1 in clinical testing costs (at some huge per-patient costs that are
secret for now) becoming $3.5, after adjustments for risk. Then you add in
the $1.57 for automatically assumed pre-clinical costs (44.5 percent of the
$1.57), which adds to $5.06. Then you add in $4.22, for finance charges
(long estimated development times, financed at inflation PLUS 10.5
percent), and you get $9.28. That how you get to $2.6 billion, in 2014.



More information about the Ip-health mailing list