IOM/NRC/NAS September 30, 2008
Economic Impact of Disease and the Case for Surveillance
Bruce Lee, MD MBAAssistant Professor of Medicine, Epidemiology, and Biomedical
InformaticsUniversity of Pittsburgh
IOM/NRC/NAS September 30, 2008
Overview
• The Value of Surveillance• Cost of Disease • Public Health Response
Costs• Direct Health Care Costs• Productivity Losses• Additional Economic Costs• Conclusions
IOM/NRC/NAS September 30, 2008
Value of Surveillance
• The Value of Surveillance• Cost of Disease • Public Health Response
Costs• Direct Health Care Costs• Productivity Losses• Additional Economic Costs• Conclusions
IOM/NRC/NAS September 30, 2008
Economic Value of an Intervention
Net Returns = -Savings from
Outcomes Averted
Cost of Intervention
Economic Value of Influenza Vaccine
Cost of Influenza
Outcomes Prevented
Cost of Influenza Vaccine
= -
IOM/NRC/NAS September 30, 2008
Surveillance as “Insurance”
2009 2010 2011 2011 2012 2013 2014 2015 2016 2017 2018 2019
Disease Costs
Surveillance Costs
?
?
? ?
IOM/NRC/NAS September 30, 2008
Actuarially Fair Premium
Value of “Insurance”
= Reduction of Loss Probability
Value of Avoidable Loss
X
IOM/NRC/NAS September 30, 2008
What can be done during this time to alter “history”?
Value of Time and Reduction in Loss
Surveillance Detection of
DiseasePublic Health Manifestation
of Disease
Value of Surveillance
IOM/NRC/NAS September 30, 2008
Early Response can Be Cost-Savings
Start of Post-attack Treatment (days)
Savi
ngs
($
bill
ion
)Anthrax Tuleremia Brucellosis
20
15
10
5
0
0 1 2 3 4 5 6 0 1 2 3 4 5 6 7 14 23 56
Kaufmann AF, Meltzer MI, Schmid GP. Emerg Infect Dis. 1997
IOM/NRC/NAS September 30, 2008
Economic Value of BioWatch
• Cost-benefit model of biological surveillance
• Reduces time to treatment to 48 hours
• Economic benefit: $1.11 billion to $50.74 billion
• Depends on nature of release and value of statistical life assigned.
• Costs of BioWatch justified when probability of biological threat >1.26 percent.
Schneider, J Environ Health. 2005.
IOM/NRC/NAS September 30, 2008
Early Information Interrupts Chain
Disease
Morbidity
Mortality
Productivity Loss
Productivity Loss
Health Care Costs
InfectedOutbreak
Decrease transmission
Alter the course of disease
Alter outcomes of disease
Reallocate replacement resources
IOM/NRC/NAS September 30, 2008
Potential Interventions
Decrease Transmission
Alter Course of Disease
Alter Outcomes of Disease
Reallocate Replacement Reosurces
Vector Control
?
Supportive Care
Always Possible
West Nile Virus
VaccineNPI
Antivirals
Supportive Care
Antivirals
Always Possible
Influenza
NPI
?
Supportive Care
Always Possible
SARS
Environmental Containment
Antibiotics
Supportive Care
Always Possible
Anthrax Animal-Human
Human-Human
Environment-Human Combined
IOM/NRC/NAS September 30, 2008
Cost of Disease
• The Value of Surveillance• Cost of Disease• Public Health Response
Costs• Direct Health Care Costs• Productivity Losses• Additional Economic Costs• Conclusions
IOM/NRC/NAS September 30, 2008
Types of Economic Studies
Prospective
Active data collection
during outbreak or
epidemic
Validity
Requires surveillance
systemGeneralizability
Retrospective
Review of records,
interviews, and surveys
Validity Data captureGeneralizability
Modeling and Simulation
Mathematical/computer
model
GeneralizabilityScenario analysis
Sensitivity analyses
Needs to be grounded in data
Study Methods Advantages Disadvantages
IOM/NRC/NAS September 30, 2008
Time Frame or Time Horizon
Time
Cost
Study Time Frame
IOM/NRC/NAS September 30, 2008
Components of Economic Costs
67%
33%
Productivity Loss
Medical Costs
Total: $96.2 millionAbout $240 per person
• Retrospective analysis of Milwaukee Cryptosporidiumoutbreak
• Corso, et al. 2003
• Societal perspective
• 4 month horizon
IOM/NRC/NAS September 30, 2008
Public Health Response Costs
• The Value of Surveillance• Cost of Disease • Public Health Response
Costs• Direct Health Care Costs• Productivity Losses• Additional Economic Costs• Conclusions
IOM/NRC/NAS September 30, 2008
2002 Louisiana West Nile Virus Epidemic
$20.1 Million 329 of 4,156 Cases
June 2002 to February 2003Medical
$4.4 MNon-Medical
$6.5 M
Public Health
Response $9.2 M
Zohrhabian, EID, 2004
IOM/NRC/NAS September 30, 2008
Cost of Public Health Response
• One case of measles in Iowa, 2004
• 2525 hours of personnel time for contact tracing and quarantine
• Estimated cost: $142,452
(Dayan, Pediatrics, 2005)
IOM/NRC/NAS September 30, 2008
Direct Health Care Costs
• The Value of Surveillance• Cost of Disease • Public Health Response
Costs• Direct Health Care Costs• Productivity Losses• Additional Economic Costs• Conclusions
IOM/NRC/NAS September 30, 2008
Hospitalization Costs
Pneumonia
Influenza
Respiratory Failure
Bacteria Infection
3.69 +/-0.07%
0.99 +/-0.13%
19.73 +/-0.35%
2.74 +/-0.37%
62.11 +/-0.42%
77.47 +/-0.93%
33.13 +/-0.67%
69.64 +/-1.79%
$5,329
$3,415
$12,260
$5,881
Deaths Routine DischargeMedian
$8,127 +/- $111
$5,341 +/- $186
$21,298 +/- $533
$12,419 +/- $748
Mean
Source: National Inpatient Sample (NIS) from Healthcare Utilization Project (HCUP)
IOM/NRC/NAS September 30, 2008
Total Costs from Hospitalizations
$0
$2 Million
$4 Million
$6 Million
$8 Million
$10 Million
$12 Million
$14, Million
1 99 197 295 393 491 589 687 785 883 981Hospitalizations
Cos
t
IOM/NRC/NAS September 30, 2008
Underestimation of Hospitalization Costs
Common Diagnosis
Infinite Hospital Capacity
Infinite Resources
No In-Hospital Transmission
Linear Curve
Testing/Diagnostic Costs
Transfer Costs
Decreased efficiency
Isolation, Quarantine, and Decontamination Costs
Complex Curve
No Re-Admissions
Equivalent Acuity
Repeat Hospitalizations
Treatment Costs
No Additional Procedures Procedural Costs
IOM/NRC/NAS September 30, 2008
Secondary Hospitalizations
Infected Patients
Worried Well
Misdiagnosed Patients
Stress-Induced Co-Morbidity
Exacerbations
IOM/NRC/NAS September 30, 2008
Outpatient Costs
Initial Clinic Visit
Follow-Up Clinic Visit
Chest X-Ray
*Assuming Cost-to-Charge Ratio of 0.463
99214
99213
71020
CPT Code
$95.70
$63.69
$36.64
Charge
$44.31
$29.49
$16.96
$90.76
Cost*
IOM/NRC/NAS September 30, 2008
Cost of Outpatient Visits
$-
$50,000
$100,000
$150,000
$200,000
$250,000
$300,000
$350,000
$400,000
$450,000
$500,000
1 473 945 1417 1889 2361 2833 3305 3777 4249 4957Patients
Cos
t
IOM/NRC/NAS September 30, 2008
Underestimation of Outpatient Costs
Common Diagnosis
Infinite Clinic Capacity
Infinite Resources
No In-Clinic Transmission
Linear Curve
Testing/Diagnostic Costs
Transfer Costs
Decreased efficiency
Isolation, Quarantine, and Decontamination Costs
Complex Curve
Minimal Follow-Up
Equivalent Acuity
Additional Visits
Treatment Costs
No Additional Procedures Procedural Costs
IOM/NRC/NAS September 30, 2008
Secondary Clinic Visits
Infected Patients
Worried Well
Misdiagnosed Patients
Stress-Induced Co-Morbidity
Exacerbations
IOM/NRC/NAS September 30, 2008
Productivity Losses
• The Value of Surveillance• Cost of Disease • Public Health Response
Costs• Direct Health Care Costs• Productivity Losses• Additional Economic Costs• Conclusions
IOM/NRC/NAS September 30, 2008
Productivity Losses
+
+
# Deaths
# Hospitalizations
# Outpatients
NPV Future Earnings
8 Hours
4 Hours
X
X
X
Average LOS
# Visits/Patient X X
X X Average Wage
Average Wage
IOM/NRC/NAS September 30, 2008
Age Distribution
0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
6.00%
7.00%
8.00%5
5-9
10-1
4
15-1
9
20-2
4
25-2
9
30-3
4
35-3
9
40-4
4
45-4
9
50-5
4
55-5
9
60-6
4
65-6
9
70-7
4
75-7
9
80-8
4
> 85
Age
IOM/NRC/NAS September 30, 2008
Life Expectancy Distribution
0
10
20
30
40
50
60
70
80
900 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80 84 88 92 96 100
104
108
Age
Life
Exp
ecta
ncy
IOM/NRC/NAS September 30, 2008
Wage Increases Since 1990Year Index Increase
1990 21,027.981991 21,811.60 3.7%1992 22,935.42 5.2%1993 23,132.67 0.9%1994 23,753.53 2.7%1995 24,705.66 4.0%1996 25,913.90 4.9%1997 27,426.00 5.8%1998 28,861.44 5.2%1999 30,469.84 5.6%2000 32,154.82 5.5%2001 32,921.92 2.4%2002 33,252.09 1.0%2003 34,064.95 2.4%2004 35,648.55 4.6%2005 36,952.94 3.7%2006 38,651.41 4.6%
Average IncreaseMedian Increase
3.9%4.3%
IOM/NRC/NAS September 30, 2008
Net Present Value of Lifetime Earnings
NPV of 35 Year Old Working Until 65 Years
Old
NPV of 35 Year Old Working Until 75 Years
Old
=
=
Σ (Wage)Age-35 / (1.03)Age-35
Σ (Wage)Age-35 / (1.03)Age-35
=
=
$1.67 million
$2.4 million
NPV of 65 Year Old Working Until 75 Years
Old= Σ (Wage)Age-35 / (1.03)Age-35 = $0.5 million
IOM/NRC/NAS September 30, 2008
Productivity Losses from Mortality
$-
$100 Million
$200 Million
$300 Million
$400 Million
$500 Million
$600 Million
1 21 41 61 81 101 121 141 161 181 200Deaths
Prod
uctiv
ity L
oss
General Population to 75
General Population to 65
Older Population
IOM/NRC/NAS September 30, 2008
Productivity Losses from Hospitalizations
$-
$100,000
$200,000
$300,000
$400,000
$500,000
$600,000
$700,000
1 211 421 631 841 1051 1261Hospitalizations
1471 1681 1891
Prod
uctiv
ity L
oss
IOM/NRC/NAS September 30, 2008
Productivity Losses from Clinic Visits
$50,000
$100,000
$150,000
$200,000
$250,000
$300,000
Prod
uctiv
ity L
osse
s
$350,000
$400,000
$450,000
1 523 1045 1567 2089 2611 3133 3655 4177 4699Clinic Visits
IOM/NRC/NAS September 30, 2008
Underestimation of Productivity Loss
Maximum efficiency and no queues
No test waiting timesMinimal travel
Maximum efficiency and no delaysDischarge to home and no recovery time.
Wages are a perfect proxyNo operational concerns
Minimal length of stay.
Perfect insurance.
Hospital Time
Perfect insurance
Immediate scheduling
Clinic Time
Productivity Losses
IOM/NRC/NAS September 30, 2008
Additional Economic Costs
• The Value of Surveillance• Cost of Disease • Public Health Response
Costs• Direct Health Care Costs• Productivity Losses• Additional Economic Costs• Conclusions
IOM/NRC/NAS September 30, 2008
Congressional Budget Office: Pandemic Influenza
• Lost production as a result of work absence.
• “Mild” scenario: would cost about 1% of GDP while a
• “Severe” scenario would cost 4.25%of GDP.
• Costs of $130-550 billion • Did not consider threshold and
compounding effects • Did not include other important
economic effects.
IOM/NRC/NAS September 30, 2008
Blackouts Disrupted Operations
• On August 14, 2003, 50 million people, from Cleveland to Toronto to New York City lost electricity for almost a day. Significant disruptions.
• 1967 blackout in New York City resulted in looting, arson, and mayhem.
• Infectious disease outbreak could be worse.
IOM/NRC/NAS September 30, 2008
Major Economic Disruptions
Tourism
Stock Market
Consumer Confidence
IOM/NRC/NAS September 30, 2008
Threshold at which Costs Explode
Costs
% of Population Infected
IOM/NRC/NAS September 30, 2008
Economic Trends
Increased Specialization
Just in Time Inventory
Maximal Operating Capacity
Lave, Casman, and Lee, 2008
Decreased Interchangeability and Adaptability
Decreased Reserves
Minimal Surge Capacity
Shift Failure Curve to the
Left
10%
IOM/NRC/NAS September 30, 2008
Conclusions
• The Value of Surveillance• Cost of Disease • Public Health Response
Costs• Direct Health Care Costs• Productivity Losses• Additional Economic Costs• Conclusions
IOM/NRC/NAS September 30, 2008
Summary
$4,000 to $13,000 Per Hospitalization
$200 Per Clinic Visit
$1-2 Million Per Death
$26,000 to $150,000 Per Case
Additional Treatment Costs Follow-Up Care
Additional Treatment Costs Follow-Up Care
Operational Costs
Communications Costs
Investor and Consumer Confidence
Follow-Up Care
Transaction Costs
Legal Costs
IOM/NRC/NAS September 30, 2008
Conclusions
• Economic studies and models are likely to produce optimistic underestimates of the cost-of-disease.
• Costs do not scale linearly• Economic trends increasing the
potential costs of outbreaks and epidemics
• Anything that affects more than 10% of a local population may cause an explosion in costs.
• Positive Externalities: Facilitate research and understanding in seasonal and endemic infectious diseases (e.g., influenza, mononucleosis, Lyme disease)
Direct Health
Care Costs
Public Health Response and Productivity Losses