Transcript Berndt et al. - Berkeley Law

March 16, 2009
Beyond the Ivory Tower:
Making Practical Choices
Stephen M. Maurer
Goldman School of Public Policy & Berkeley Law School
Introduction
Advocate’s View:
“Everything’s Great (and We Can Fix It)”
This Course (Agnostic)’s View:
“Whatever Costs Less”
Today’s Lecture:
Illustrating the “Sharp Pencil Approach”
A Work – and Hopefully a Conversation –
In Progress
Roadmap
Prizes vs. PPPs
Overpayment Under E2E Prize
Overpayment Under PPPs*
Designing a PaYG System
PPPs, Access, and Leveraging Resources
*PPPs can offer prizes!
Prize Overpayments
Will Sponsors Overpay?
Usual Viewpoint: Reward Amounts “Should
be More Transparent.”
This Lecture: Smart People Working Very,
Very Hard!
Prize Overpayments
Naïve Calculation: DiMasi et al.
$803m ± $115m  14% Average
Overpayment
DiMasi et al., “The Price of Innovation,”
Journal of Health Economics 22: 151-185 (2003)
Prize Overpayments
A Better Calculation
R&D Cost = Net Revenue – Marketing Costs
What are the error bars?
Berndt et al., “Advanced Markets for a
Malaria Vaccine: Estimating Costs and
Effectiveness” (2005)
Prize Overpayments
What is Net Revenue?
$2.84bn
Limited Data
Revenue is
Not a Single Number…
Uncertainty ~ 10%??
Circularity?
Grabowski, et al. “Returns on Research and Development for
1990s New Drug Introductions.” Pharmacoeconomics 20 Supp.
3:11-29 (2002)
Prize Overpayments
R&D Cost = Net Revenue – Marketing Costs
What is Marketing Cost?
Published Estimates: 15 - 36%
 R&D Expenditure is $1.8bn - $2.4bn.
 Implied overpayment = 14%.
What Number Do Berndt et al. use?
Prize Overpayments
10%
Rationale: Drug companies will spend less on
promotion (including free samples).
 $2.56bn
18% Overpayment
What number do Berndt et al. use?
Prize Overpayments
$3.0bn.
Rationale: “[A] malaria vaccine may be more difficult to
Develop than the typical new chemical entity.”
$25% Overpayment
Folklore: $3.3bn.
 39% Overpayment
Prize Overpayments
Caveats:
Assumes Past R&D Costs
Tied to Rich Nation R&D Effort
What Does “Overpayment” Mean?
Finessing Our Ignorance?
Prize Overpayments
Caveats: Assumes Past R&D Costs
Data are often 15 years old.
Real drug discovery costs have risen
7-8%/year since 1970s.
And: Clinical grew from 6.1% in 1970s to
11.8% in the 1980s.
What does Pharma believe today?
Prize Overpayments
Caveats: Tied to Rich Nation R&D Effort
Berndt et al.: “Would need to pay out $2.56bn to
match the average revenue brought in by NCEs.”
Global Alliance: TB vaccine would cost $115 - $240m.
8% of Berndt et al. Estimate!!!
Global Alliance for TB Drug Development: Economics of TB Drug
Development (2001)
http://www.tballiance.org/downloads/publications/TBA_Economics_R
eport.pdf
Prize Overpayments
Caveats: Tied to Rich Nation R&D Effort
PPP estimates show “enormous variability” in
input costs, should be approached with “great
caution.”
Early estimates are being revised…
Towse et al., “Estimates of the Medium Term Financial Resource Needs for
Development of Pharmaceuticals to Combat ‘Neglected Diseases’” in Widdus
and White, “Combating Diseases Associated With Poverty” (Initiative on
Public-Private Partnerships for Health: 2004)
Prize Overpayments
Caveats: What Does Overpayment Mean?
Higher Rewards  Higher Effort
Not as Bad as “Ordinary” Waste
But: Diminishing returns & “Me-Too” drugs.
Downside
Losing An “Extra” Drug Program
For 25% overpayment 
“Buy 3 get one free…”
Prize Overpayments
Caveats: Finessing Our Ignorance
Hollis: Racing for Shares in a Pooled Reward
(QALYs  Reduced Benefits?)
Adjusting the Prize Over Time?
Reward Can’t Grow Faster than Internal
Rate of Return (11%).
Assume Historic 7% Rate…
Assume 11.8% Rate…
Can’t go below rich nation reward as long
as pharma is internally funded!
PPP Overpayments
Basic Picture: Prizes + Competitive Bidding =
Unambiguous Improvement
PPP Overpayments
Purchasing Power:
Enforcing the Best Price
Example: Compound Development
15-20% Accounting Profits in Early 1990s
Cut by ~ 5% in late 1990s
Typical of outsourced costs?
Counterexamples:
Manufacturing in Asia is 5% cheaper…
Does Pharma Know How to Do Trials in
LDCs?
A much smaller penalty than prizes!
But: Are PPPs “less efficient…?”
PPP Overpayments
Competence and Adverse Selection?
Evidence (1)
Virtual Pharmas
Evidence (2)
Pasteur Institute, Fort Detrick, etc.
Probes both efficiency and reduced R&D
intensity.
Are Foundations Good Shareholders?
The Big Unknown: Foundation culture
Design Issues
Multiple Missions, Lack of Competition.
Designing a PaYG
System
Designing PaYG
PaYG includes Prizes!
Where Do Prizes Make Sense?
1. Setting R&D Priorities?
Does Pharma Know More Than the Sponsor…
About Disease Burden?
About WHO Approvals/Purchasing
Procedures?
About Uptake?
Designing PaYG
2. Pre-Clinical Drug Discovery
Ideas Are Cheap
A Small Community?
Widely Dispersed Knowledge?
Proprietary Data Inside Pharma?
Early Stage Developments
~25% of Total Costs.
Best Efforts Prizes?
Designing PaYG
3. Clinical Trials
75% of Costs
What Does Pharma Know About
Developing World Trials?
Is The “Near-Term” Prize Inverted?
4. Approvals and Distribution
Sponsors Know More than Pharma…
Adding Patents
Adding Patents
“PPPs have increased drug development”
- Suerie Moon
Adding Patents
How Much Will the Private Partner Invest?
Adding Patents
Estimating the Prize Amount (Revisited)
- Amount of Required Reward
- Future Cost
Outside LDCs
To Gates, Governments
- Do PPPs Care?
Overpayment Revisited
- Extra Effort  Crowding Out
Adding Patents
Side Issue: Silver Bullet Technologies
- If Rich Nations Didn’t Invent This Technology …
Nanotechnology?
- Counterexample: Amyris
Rich nations don’t make medicines from bark…
March 16, 2009
Beyond the Ivory Tower:
Making Practical Choices
Stephen M. Maurer
Goldman School of Public Policy & Berkeley Law School