Transcript Economics of Oncology Drug Development

Unresolved Issues in the
Globalization of Clinical Trials
Robert M Califf MD
Vice Chancellor for Clinical Research
Globalization of Clinical Trials
 Current State
Money
Ethics
Cultural environment
Genetic variation
 Envisioning a positive future
• Since 2002, the number of FDA investigators outside
the US has grown by 15% annually, while the number
inside the US has declined by 5.5%.
• One-third of phase 3 trials of the 20 largest US
pharmaceutical companies are being conducted
solely outside the US.
• For those same firms and studies, a majority of study
sites (13,521 of 24,206) are outside the US.
Source: Glickman, SW et al. NEJM 2009
The Globalization of Clinical Investigators
Percent of Total
1572s Filed
100%
US-Based
5%
9%
12%
9%
Western Europe
13%
19%
Rest of World
25%
10%
75%
29%
11%
13%
14%
50%
86%
80%
77%
70%
62%
25%
57%
0%
1997
Sources: Tufts CSDD
1999
2001
2003
2005
2007P
Growth in Numbers of Active
FDA-Regulated Investigators
2001
2003
2006
5-yr Growth
Rate
Most Recent
3-yr Growth
Rate
Russia
176
317
623
28.8%
25.3%
Poland
215
314
322
8.4%
.8%
Brazil
139
296
292
16%
-.5%
Costa Rica
12
20
22
12.9%
3.2%
Argentina
174
305
462
21.6%
14.8%
China
80
120
307
30.9%
36.8%
India
64
145
464
48.6%
47.4%
Sources: Tufts CSDD
The test of a first-rate intelligence is
the ability to hold two opposed ideas in
the mind at the same time, and still
retain the ability to function.
F. Scott Fitzgerald, "The Crack-Up" (1936)
US novelist (1896 - 1940)
Creative Tension
 Globalization of clinical research is a very good
thing for the US and for all of us in a flat world
We need more evidence to guide effective practice
in the US
So does every other country in the world
 Driving globalization because of poor efficiency
in the conduct of research in the US is a bad
thing for the US
By the way, the same thinking could be applied to
any economically advantaged country that begins to
think it is “above it all”
Dollars and Sense: Factors Pushing
Clinical Research Out of the U.S.
American Federation for Medical Research
(AFMR)
April 14, 2009
Kevin Schulman, MD
Director, Health Sector Management Program
The Fuqua School of Business
Director, Center for Clinical and Genetic Economics
Duke University Medical Center
Globalization of Clinical Trials—The Good
 Larger sample sizes are needed
Modest treatment effects predominate
Subgroups must be validated
Will become even more important in “personalized” or
“stratified” medicine
 Competition spurs improvement
 More research will be done
 Collaboration leads to shared learning
Practice patterns
Ethics
Will ensure transparency on results reporting
 Adequate sample sizes/study designs to understand genetic
heterogeneity
Globalization: A Necessity
 Proof of concept trials—there are
pharmacogenomic differences
 Efficacy trials—the context of clinical practice
matters
 Effectiveness trials—the relative costs and
balance of risk and benefits are context
dependent
The question is not whether we globalize, it is
why we do it, how we handle cultural
differences and how we lump or split findings
Economic Proposition to Industry From
Globalization
$
duration*
$1 – $10 billion*
Pre-Launch
Potential local market
opportunity
No requirement to match
study populations to
target markets
Return
speed*
fall-off*
ROI
maximize area*
$800 – $880 mil (2001)
0
Patent
Launch
Decrease costs or time
Patent
Expiration
The Good– the US Needs to Wake Up!
 Rapid movement to moving clinical trials away
from US
 US is inferior in:
Study cost index
Population indicators (tx naïve patients)
Business environment
 US is better, but losing ground quickly in:
Clinical research personnel qualification
Study site organization
Thought leader quantity
Image Source: http://www.pandora.ca/pictures21/900666.jpg
The Demise of Empires
 Dominance at a point in time
 Arrogance about superiority
 Failure to pay attention to quality of work
 Leaders content to “ride the wave”
 Entrenched interests can buy stability through
controlling laws and regulations
 Inability to create or respond to innovation
 Cost of transactions exceeds cost of
actually doing the work!
Data gathered by the Tufts Center for the
Study of Drug Development (Tufts CSDD)
 >90% of all clinical trials delayed due to over-ambitious
timelines and difficulty with patient enrollment
 Top reasons for delays in trials:
Protracted budget negotiations
Slow IRB review and approval
Poor patient recruitment and retention
 Estimated 20% of PIs fail to enroll a single patient and
30% under-enroll in a given trial
 Between 2000 and 2005 38% of PIs who participated in
clinical trials in a given year did not return in a
subsequent year through 2008
Up from 26% in previous 5 years
Patient Recruitment and Retention
Percent Screened who
were Randomized
Percent Randomized
who Completed the
Study
Screen : Complete
1999-2002
2003-2006
75%
59%
69%
48%
50%
25%
•Performance data on 57 phase II and III (across TAs) protocols provided
by five pharmaceutical companies
Source: Tufts CSDD
Protocol Designs More Complex and Burdensome
Annual Growth Rate
(2000-2006)
12.1%
10.5%
8.7%
6.5%
Compensation per
Procedure
Number of Unique
Procedures
Frequency of
Procedures
Execution Burden
Number of
Eligibility Criteria
-7.9%
Represents 10,038 industry protocols; provided by Fast Track Systems
Work effort values based on Medicare’s RVU methodology
Sources: Tufts CSDD; Getz et al. Assessing the Impact of Protocol Design
Change on Clinical Trial Performance. American Journal of Therapeutics. 2008
15(5); 450 - 457
Performance ‘Impact’ of protocol complexity
 Compared US-based pivotal trial protocols
executed 1999-2002 (lower complexity) and
2003-2006 (higher complexity):
 Number of CRF pages rose to an average of
180 pages vs. 55
 Controlling for treatment duration, cycle times
increased substantially across all measures
 Enrollment rates worsened
Source: Tufts CSDD
Cycle Time Metrics
69% - 75% Increase
780
Median Days Elapsed
714
460
12% – 20% Increase
115
129
Protocol Ready to
FPFV
120
413
143
Protocol Ready to
Drug Available
1999-2002
Protocol Ready to
LPLV
Protocol Ready to
Data Lock
2003-2006
Source: Tufts CSDD
Clinical Trial Cost Estimates
$450
$400
Total
$350
Coordinating Center
$300
Site Payments
$250
Other
$200
$ In US 2007 Millions
$150
$100
$50
$0
Full Cost
Industry
Streamlined
Industry
More
Streamlined
Pharmacogenomics
Source: Roses AD. Nature 2000;405:857-865.
Randomized Trial of Genotype-Guided
Dosing of Warfarin Therapy:
Influence of Genotype on Warfarin Dose?
Genotype
(N = 188)
Prevalence
% Enzyme
Activity
S/R Warfarin
(mg/L)
Daily
Doses
(mg)
Clearance/LB
W
(ml/min/kg)
2C9 *1/*1
64%
100%
0.45
(0.11)
5
(2.5)
0.065 (0.025)
2C9 *1/*2
*2/*2
21%
50-70%
0.69
(0.28)
3
(1.5)
0.041 (0.021)
2C9 *1/*3
*2/*3 *3/*3
15%
10%
1.43
(0.63)
2
(1.0)
0.020 (0.011)
Herman et al, The Pharmacogenomics J 4:1-10. 2005
(From L. Lesko)
McClain et al, ACMG Presentation, October 30, 2006 (In Press)
Randomized Trial of Genotype-Guided
Dosing of Warfarin Therapy
Boxplots of distribution of warfarin dose by
CYP2C9 and VKORC1 genotype
Sconce et al. Blood 2005; 106:2329
-NEJM 360; 2009
Cytochrome P-450 Polymorphisms and
Clopidogrel Response
-NEJM 360; 2009
Cytochrome P-450 Polymorphisms and
Clopidogrel Response
-NEJM 360; 2009
GENEVA, SWITZERLAND—World Health Organization officials expressed disappointment Monday at the
group's finding that, despite the enormous efforts of doctors, rescue workers and other medical
professionals worldwide, the global death rate remains constant at 100 percent.
Death, a metabolic affliction causing total shutdown of all life functions, has long been considered
humanity's number one health concern. Responsible for 100 percent of all recorded fatalities worldwide,
the condition has no cure.
"I was really hoping, what with all those new radiology treatments, rescue helicopters, aerobics TV
shows and what have you, that we might at least make a dent in it this year," WHO Director General Dr.
Gernst Bladt said. "Unfortunately, it would appear that the death rate remains constant and total, as it
has inviolably since the dawn of time."
The Good--More Research will be Done
 Broader sources of funding
Many governments now giving tax breaks for
industry funded research
 Do the research where costs are less
 Do the research where barriers to trial initiation
are less
ACC/AHA Clinical Practice
Guidelines
ACC/AHA Guidelines: Level of
Evidence of Recommendations
19.0%
LoE A
36.3%
LoE B
LoE C
44.9%
ACC/AHA Guidelines: More Guidelines
No Improvement in Proportion with High Quality!
70%
60%
50%
40%
30%
20%
10%
0%
A B C
Atrial
Fibrillation
(2001-2006)
A B C
Heart
Failure
(2001-2006)
A B C
Stable
Angina
(1999-2002)
A B C
Unstable
Angina
(2000-2007)
A B C
A B C
PCI
(2001-2005)
Pacemaker
(1998-2002)
The Bad
 Differences in practices can be found as a
function of geography
 Differences in outcomes can be found as a
function of geography
 Differences in treatment effect can be found as
a function of geography
Current Practices in ACS Care
USA vs Rest of GRACE
Andrzej Budaj
for the GRACE Investigators
Postgraduate Medical School, Grochowski Hospital, Warsaw Poland
In-hospital Management
All ACS
USA
Rest of GRACE
AA (%)
95
95
B-blockers (%)
88
B-blockers iv (%)
27
ACE-I (%)
61
ARB (%)
7
Statins (%)
Other lipid lowering (%)
65
8





82
9
66
5
66
3
Temporal trends
USA vs Rest of GRACE
%
8
Death in hospital
All ACS
7
6
6.3
5.7
4.9
5
4
USA
Rest of GRACE
4.4
4.8
4.7
4.2
4.0
3
2
2000
2002
2004
2006
Death and/or MI at 30 Days
Placebo
Eptifibatide
Total n
Male
16.88
13.89
6103
Female
13.68
14.88
3357
0.5
Eptifibatide
Better
9808LB11, 38
PT-E2-7
1
1.5
Placebo
Better
Death and/or MI
Male
NA Female
Male
WE
Female
Male
EE
Female
Male
LA
Female
0.2
Eptifibatide
Better
9808LB11, 39
PT-E2-8
1
Placebo
16.16
Eptifibatide
12.35
Total n
2499
12.89
16.03
12.10
20.35
19.05
10.57
13.03
15.52
20.43
21.78
1328
2542
1154
814
727
21.43
5.63
16.39
15.58
248
148
5
Placebo
Better
Timing of Angiography
100
Male
Female
75
%50
25
0
72
7
30
Hours Days Days
Overall
9808LB11, 40
PT-C-17
72
7
30
Hours Days Days
72
7
30
Hours Days Days
72
7
30
Hours Days Days
72
7
30
Hours Days Days
North
America
Western
Europe
Eastern
Europe
Latin
America
Study Undertaken by FDA
statisticians to evaluate
possibility of systematic
regional differences
 Major cardiovascular outcome studies
evaluated over the last 10 years
 Overall study result statistically
positive, ie. demonstrated overall
effect
 Region never pre-specified as a factor
to be evaluated statistically
 16 independent studies
Estimates and confidence intervals for difference
between US and Non-US treatment effects for each
study
In 13 of 16 , US log hazard
above 0
Study
% US
1
31
2
45
3
27
4
5
9
4
6
19
7
43
8
38
9
10
4
74
11
74
12
9
13
3
14
15
29
17
16
90
-1.5
-1.0
-0.5
0.0
0.5
difference of log-hazard ratios
J. Lawrence
1.0
1.5
A figure
From the
label
Goal
External Validity
X
Internal Validity

Article cites numerous examples of trials, which produce
parameter estimates of questionable value to Western
European decision makers and highlights differences
between emerging regions and Western Europeans, which
suggest the two are mutually incomparable.

Proposed that these differences may confound study
outcomes, decision-making parameter estimates and data
pertaining to the incidence of adverse drug interactions

Further research should be undertaken in order to explore
the relationship between geographical variance and
external validity, particularly where safety data derived
from relatively drug naïve regions are assumed to pertain to
a maximally treated populations elsewhere in the world
Source: Wathal. Outsourcing Clinical Trials www.samedanltd.com

Extrinsic factors such as medical practice, disease definition and
study population may influence applicability of foreign data to an
EU setting

Global drug development doesn’t necessarily support approval of
unrestricted indications in an EU population


Consider and discuss possible influence of extrinsic
factors on interpretation of results and wording of
indications
In depth, prospective analysis of potential ethnic factors when
conducting a clinical trial in a certain region.

Depending on the outcome of analyses can be
decided whether certain clinical trials conducted in a
specific area of the world would be relevant to EU
setting or if there are reasons to perform additional
clinical trials within the EU
Source: European Medicines Agency, Pre-Authorization Evaluation of Medicines for Human Use. Feb. 19, 2009

Verification at time of evaluation of marketing
authorization application that trials have been conducted in
accordance with GCP and ethical standards


Greater transparency of this process and its
outcome should be described in the European
Public Assessment Report (EPAR)
Increased GCP inspection including further extension of
GCP policy on increasing numbers of routine inspections as
part of the need for greater supervision of the conduct and
ethical standards of clinical trials performed outside the
EEA
Source: European Medicines Agency, EMEA strategy paper: acceptance of clinical trials conducted in the third countries, for evaluation in
Marketing Authorization Applications. Feb. 5, 2008
Source: Reed SD, Drug Information Journal, 2007
Source: Reed SD, Drug Information Journal, 2007
Source: Reed SD, Drug Information Journal, 2007
The Ugly
 Differences in reported adverse events can be
found in different countries
Is there a relationship between relative financial
incentive and lower rates of side effects?
 Differences in adherence can be found in
different countries
Is there a relationship between lack of access to
health care outside of the trial and adherence?
The Ugly
UK
US
NonAdherenc
e
39.7%
35.5%
Withdrawa Withdrawa Withdrew
l:AE’s
l: Subject Consent
preference
12%
17%
55
7%
16%
166
Poland
China
Argentina
14.8%
6.1%
13.6%
2%
5%
2%
9%
5%
8%
2
1
1
The Obscene
 When the per patient reimbursement exceeds
reasonable levels, the human experimentation
entrepreneurs will be tempted
 When the FDA cannot inspect, the cheaters will
figure out how to get the data looking real
good!
Source: http://www.tampabay.com/news/business/article934677.ece
Source: http://www.tampabay.com/news/business/article934633.ece
Source: http://www.tampabay.com/opinion/essays/article934654.ece
Ethical Concerns in clinical trials in India:
an investigation
• Lapatinib, GlaxoSmithKline
– The majority of breast cancer patients in India cannot
afford proper treatment. This trial required seriously ill
patients who had not received treatment for their
condition.
– Their economic vulnerability forces patients in India to
take part in trials in order to get access to treatment
and to disregard the potential risks that participating in
clinical trials entails. By carrying out this clinical trial in
India GlaxoSmithKline (GSK) took advantage of the
vulnerable position of breast cancer patients.
Source: Srinivasan S, et al. Ethical concerns in clinical trials in India: an investigation. Center for Studies in Ethics and Rights, Mumbai, India. Feb. 2009
Ethical Concerns in clinical trials in India: an
investigation
• Risperidone, Johnson & Johnson
– Patients in trial were suffering from an acute
attack of a psychiatric condition that would
have caused them much distress.
– They were harmed because they were taken
off all treatment before they were put on either
the active drug or a placebo.
– Those on the placebo were also harmed
because they were deprived of an effective
treatment
Source: Srinivasan S, et al. Ethical concerns in clinical trials in India: an investigation. Center for Studies in Ethics and Rights,
Mumbai, India. Feb. 2009
Ethical Concerns in clinical trials in India: an
investigation
• Quetiapine fumurate extended release,
AstraZeneca
– These two placebo-controlled trials of
quetiapine were conducted on patients with
schizophrenia. An immediate release
formulation of the drug had already been
approved and these trials were of an
extended release version of the drug.
– A patient in one of the quetiapine trials
committed suicide after 173 days of being on
placebo.
Source: Srinivasan S, et al. Ethical concerns in clinical trials in India: an investigation. Center for Studies in Ethics and Rights,
Mumbai, India. Feb. 2009
Factors Pulling Research Offshore
•
•
•
Cost: cost per patient may be 1/10 the cost in
the US (Garnier, JP).
Availability: untreated or under-treated
patients may speed recruitment of patients to
clinical trials reducing the time costs of
research.
Regulations: local processes may be more
variable (and potentially less restrictive) than
in the US.
Source: Garnier JP. Rebuilding the R&D engine in big pharmacy. Harv Bus Rev 2008;86:68-76.
Factors Pushing Research Offshore
•
•
Costs (Clinical Care)
Costs (Administrative)
•
•
•
•
Systems that developed to govern single site
studies are inefficient in oversight of multicenter,
multinational studies
IRB (redundant site level process)
Contracting (redundant site level process)
Compliance (research as a potential criminal
issue)
Administrative Barriers
Source: Dilts, DM et al. Journal of Clinical Oncology Oct. 2006
IRB Approval Timeline for a Focus Group
discussing Health Insurance with Latinos
Event
Time
Comments
Study Submitted to IRB as
Exempt
Aug 2007
IRB rejected exempt status
Discussions with IRB about
rejection of exempt status
Aug-Sep 2007
Focus groups may not qualify
for IRB exempt status
Study rewritten and submitted
as expedited
Oct 2007
IRB requested extensive
changes
Initial study approved by IRB
Feb 2008
Local cultural center requests
major changes to study
Apr 2008
Continued negotiations between
local cultural center and IRB
Incorporate relevant cultural
concerns in initial study design
May-Sep 2008
Rewritten study submitted to
IRB
Oct 2008
Final Study Approved
Jan 2009
IRB requested additional
changes
Contracting for Clinical Research
“…the system would be better served if
there were universally accepted
contractual language …
Such language would help safeguard the
integrity of the research process.”
-JEFFREY M. DRAZEN
Editor-in-Chief, New England Journal of Medicine
Source: Drazen, JM. NEJM Oct. 24, 2002
(1) I am familiar with, and will
comply with, applicable federal
regulations and guidance for the
protection of human subjects:
HHS regulations at 45 FR 46 and
associated guidance; FDA
regulations at 21 CFR Parts 50,
54, 56, 312, 314, 601, 812, and
814 and associated guidance;
the HIPAA privacy regulations at
45 CFR Parts 160 and 164 and
associated guidance; the DUHS
Federal-Wide Assurance; and
relevant institutional policies and
procedures for the protection of
human research subjects.
Source: DUHS Principal Investigator Agreement form. Version 6/25/08
Source: Glickman, SW et al. NEJM 2009.
Selection of Patients in Multinational Trials
• Problem: Research in communities that are not
intended to be major markets for the products
under testing can be ethically problematic
• Solutions
– Sponsors should describe how trial populations match
intended markets
– Create target enrollment according to region on the basis
of intended use of product, similar to target enrollment of
women and minorities
Source: Glickman, SW et al. NEJM 2009.
Transparency of Clinical Trial Results in
Developing Countries
• Problem: Protection of publication rights and access
to trial data for investigators is necessary to
preserve the integrity of research
• Solutions
– Publish all data regardless of research location, and
reinforce requirements according to FDA
Amendments Act of 2007
– Preserve publication rights globally through legal
agreements at onset of trial
– Create trial leadership that incorporates
representatives of countries involved in study
Source: Glickman, SW et al. NEJM 2009.
Regulatory Oversight of International
Clinical Research
• Problem: Regulatory agencies have little
information on trials conducted outside their
countries
• Solutions
– Mechanism for sharing regulatory oversight among
government agencies worldwide
– Public registry of IRBs and inventory of country-specific
provisions for ethical oversight
– Comprehensive study of the globalization of clinical
research by IOM or WHO
– Central statistical monitoring system to find unusual data
patterns suspicious for fraud
Source: Glickman, SW et al. NEJM 2009.
Training and Experience of Clinical
Investigators
• Problem: Investigators in developing countries are
typically less experienced than investigators in
developed countries
• Solutions
– Formal training programs for clinical research for
investigators in developing countries to expand global
clinical research leadership capacity and improve
collaboration worldwide
– Mechanism for tracking investigators who are trained
to conduct clinical trials and those who have been
prohibited from conducting trials
Source: Glickman, SW et al. NEJM 2009.
Genomic Information in Drug
Development
• Problem: Lack of pharmacogenomic data for
subjects limits confidence in generalizability of
results
• Solutions
– Expand FDA Voluntary Genomic Data Submissions
program to international regulatory agencies
– Develop global data warehousing and data analysis
capabilities
Source: Glickman, SW et al. NEJM 2009.
IRB Quality and Inefficiency
• Problem: Redundancy in review process may harm
patient safety by requiring diversion of effort to
unnecessary procedures and practices
• Solutions
– Greater use of centralized IRBs (eg, Central IRB
Initiative, European Union Clinical Trials Directive)
– Mutual acceptance of proposal review in consortia
(eg, Biomedical Research Alliance of New York)
– Streamlined best practices to reduce unnecessary
work for investigators (eg, Clinical Trials
Transformation Initiative)
Source: Glickman, SW et al. NEJM 2009.
Payment Compliance
• Problem: Increased costs and delays in payment
for research subjects divert financial support
from research to administration and make
research less attractive to investigators because
of risk of criminal penalties
• Solutions
– Establish nonpunitive mechanism for reconciliation of
payment
– Expand mechanisms to pay for usual care services (eg,
within Medicare and Medicaid)
Source: Glickman, SW et al. NEJM 2009.
Commercial Contracts
• Problem: Variety of contracting practices
brings complexity and delays to research
• Solutions
– Adopt standard contract language for clinical
research agreements
Source: Glickman, SW et al. NEJM 2009.
Confidentiality Agreements in
Commercial Contracts
• Problem: Confidentiality agreements reduce the
transparency and efficiency of clinical research
• Solutions
– Adopt standard confidentiality language for clinical
research agreements
Source: Glickman, SW et al. NEJM 2009.
A collaborative effort to find solutions
 In light of these issues the U.S. FDA’s Office of
Critical Path Programs and Duke University
joined together as founding members of a publicprivate partnership:
The Clinical Trials Transformation Initiative (CTTI)
 All stakeholders are involved in this initiative
including government, industry, academia,
patient advocates, clinical investigators,
professional societies, and others
Mission
To identify practices that through broad
adoption will increase the quality and
efficiency of clinical trials
Scope
CTTI will conduct projects in support of
its mission to identify practices that will
increase the quality and efficiency of
clinical trials
“Quality” - the ability to effectively answer
the intended question about the benefits and
risks of a medical product (therapeutic or
diagnostic) or procedure, while assuring
protection of human subjects
Scope (continued)
CTTI will generate evidence about how
to improve the design and execution of
clinical trials
Projects about design will address
principles generally applicable to clinical
trials to assure that they are fit to
accomplish their intended purpose.
Scope (continued)
While CTTI focuses on clinical trials, it
may study other types of clinical
research (e.g., registries) that can
provide data to regulatory agencies.
Although CTTI will concentrate initially
on the design and conduct of clinical
trials in the United States, it seeks to
identify practice improvements that can
be applied internationally.
Executive Committee
 FDA: (Rachel Behrman, OC, Co-chair; Bob Temple,
CDER, Bram Zuckerman, CDRH)
 Duke: (Rob Califf, Co-chair)
 NIH liaison: (Amy Patterson)
 Industry: (Glenn Gormley, Jay Siegel, Susan Alpert,
Alberto Grignolo)
 Academia: (David DeMets)
 Patient representative: (Nancy Roach)
 At-large representative: (Ken Getz)
 Non-US regulatory liaison: (Hans-Georg Eichler, EMEA)
 Steering Committee co-chairs, ex officio (James
Ferguson, Briggs Morrison)
Member Organizations
Category
# organizations
Pharmaceutical companies
7
Clinical research organizations
7
Academic institutions
7
Professional societies
7
Device companies
6
Biotechnology companies
5
Government
4 (FDA, CMS, NIH, OHRP)
Clinical investigator groups
3
Trade organizations
3
Patient representatives
2 (TBD)
Private equity firm
1
Regulatory law firm
1
*Began recruiting members May
2008
Projects
 Priority areas defined by Executive Committee:
Design principles
Data quality and quantity (including monitoring)
Study start-up
Adverse event reporting
 Information about the process for submission,
review, and approval of projects available at
CTTI Web site:
www.trialstransformation.org/projects
Life Expectancy at Birth: Developed
and Developing Countries, 1955-2002
Source: World Health Report 2003