Slide show - Department of Health and Human Resources

Report
Community Interventions for Pan Flu—
Lessons from History and Modeling
Cathy Slemp, MD, MPH
WV Bureau for Public Health
November 2006
Influenza Pandemic Viruses
Requirements:
– A new influenza A subtype that can infect humans
AND
– Causes serious illness
AND
– Spreads easily from human-to-human
H5N1 meets the first two prerequisites,
but not the last
Next pandemic virus may or may not be due to a
variation of current H5N1 virus
Update:
H5N1 in Humans – 2003-2006
• As of October 31, 2006: 256 cases, 152 deaths (~60%)
– Ten countries
• Sporadic, with occasional clusters
• Most had close contact with sick poultry
• Few cases of probable, limited human-to-human
transmission
• All lived in countries with poultry outbreaks
Update
As of 10/31/06: 256 cases; 152 deaths; most poultry related
Flu Pandemics Happen.
(10 in last 300 years)
Impacts of Past Pandemics
Pandemic
Deaths in
the US
Deaths
Worldwide
Population Affected
Spanish Flu (H1N1)
1918-1919
500,000
40 million
Persons 20-40 years
old
Asian Flu (H2N2)
1957-58
70,000
1-2 million
Infants, elderly
Hong Kong Flu (H3N2)
1968-69
36,000
700,000
Infants, elderly
IMPACT CAN BE DRAMATIC
Daily Deaths in Ohio - 1918
Brodrick OL. Influenza and pneumonia deaths in Ohio in October and November, 1918.
The Ohio Public Health Journal 1919;10:70-72.
Sample Estimate of Pandemic
Morbidity/Mortality, West Virginia*
•
•
•
•
•
•
•
Characteristic Moderate (1957-68-like) Severe (1918-like)
Illness
540,000 (30%)
540,000 (30%)
Outpatient
270,000 (50%)
270,000 (50%)
Hospitalization
5,314
60,813
ICU Care
791
9,123
Ventilators
399
4,558
Deaths
1,284
11,690
* Based upon DHHS U.S. estimates applied to WV population numbers.
These are in the absence of potential interventions.
What we don’t know … about
the next pandemic
• When will it occur?
• Which virus will cause it, H5N1 or another?
• Who will be most at risk (Elderly and infants?
Other?)
• How severe an illness will it cause?
• Will there be multiple waves?
• Will antiviral medication work?
• How long until we have a vaccine?
• What are the best control measures?
Planning Pandemic Control
Measures
Community-Based Interventions
1. Delay outbreak peak
2. Decompress peak burden on hospitals / infrastructure
3. Diminish overall cases and health impacts
#1
Pandemic outbreak:
No intervention
#2
Daily
Cases
Pandemic outbreak:
With intervention
#3
Days since First Case
What can we learn from
historical analysis?
Excess Mortality /
100,000 Population
Cumulative Excess Mortality
by Location in 1918
1000
900
800
700
600
500
400
300
200
100
0
0
10
20
30
40
Location
McLaughlin AJ. Epidemiology and Etiology of Influenza. Boston Medical and Surgical Journal, July 1920.
Deaths Rates / 100,000 Population
(Annual Basis)
1918 Death Rates: Philadelphia v St. Louis
16000
Philadelphia
St. Louis
14000
12000
10000
8000
6000
4000
2000
0
22 /22 /22 /22 /22 /22 /22 /22 /22 /22 /22 /22 /22 /22 /22 /22
/
15 /22 /29 0/6 /13 /20 /27 1/3 /10 /17 /24 2/1 2/8 /15 /22 /29
/
1
9
9
9
1
1
1
11
11
11
10
10
10
12
12
12
Date
Weekly mortality data provided by Marc Lipsitch (personal communication)
/1
8
/1
8
/1
8
/1
8
Date
/1
8
8/
18
12
/2
140
1/
18
160
12
/2
4/
18
12
/1
Peak
147
52
12
/7
0/
18
11
/3
Wash DC
Seattle
3/
18
11
/2
6/
18
11
/1
/1
8
Deaths / 100,000 Population
100
11
/9
/1
8
11
/2
6/
18
10
/2
9/
18
10
/1
2/
18
10
/1
10
/5
9/
28
9/
21
9/
14
Washington DC v Seattle
Seattle
Washington, DC
120
Aggregate (1918)
550
335
80
60
40
20
0
Thinking Through
Control Measures
Influenza Transmission
Facemasks, cough etiquette
Viruses:
Cleaning, handwashing
Leave original host
Survive in transit
Social distance, cohorting
Facemasks, handwashing
Vaccination
Delivered to a susceptible host
Reach a susceptible part of the host
Escape host defenses
Multiply and cause illness
Potential Tools in Our Toolbox
• Our best countermeasure – vaccine – will probably be
unavailable during the first wave of a pandemic
• Antiviral treatment may improve outcomes but will have
only modest effects on transmission
• Antiviral prophylaxis may have more substantial effects
on reducing transmission
• Infection control and social distancing should reduce
transmission, but strategy requires clarification
Ro
R0 = 12
Effect of Increasing Social Distance
on Epidemic Dynamics
Exponentiation
Suppression
Ro = 2.0,
Ro = 0.67,
Progression = 1:2:4:8:16
Progression = 1:2:4:3:2
Increasing “Social Distance”

“Community Shielding” Measures







Close or alter high risk transmission environments e.g.
schools, daycare centers if supported by epidemiology
Cancel large public gatherings (concerts, theaters)
Minimize other exposures (market, church, public transit)
Encourage ill and exposed to stay home (I & Q)
Worksite adaptations (e.g., telecommuting, etc.)
Scaling back transport services (holiday schedule)
Consider additional community measures


COOP to minimize economic impact
Surgical masks, barrier precautions, hand hygiene
What Does Disease Modeling
Suggest?
Examining the Potential
of Combined Interventions
%Population Infected
80%
70%
60%
50%
40%
30%
20%
QP
Q,T
PT
P
T
Q
None
N
on
e
Sc
ho
ol
Ki
ds
Sc
Ad
ho
ul
ts
ol
,A
du
Ki
ds
l ts
,A
du
Sc
Sc
l ts
ho
ho
ol
ol
,
,K
Ki
id
ds
s,
A
du
lts
0%
Q,P,T
10%
Value of combining strategies –
Longini model
70
60
50
40
30
20
10
0
Clinical attack rate
Antiviral stockpile needed
Base case (Ro=1.9)
Generic social distancing
School closure
School closure + generic social distancing
60% Case treatment + 60% household prophylaxis
60% Case treatment + 60% household prophylaxis + 60% social prophylaxis (60% TAP)
60% TAP + School closure + generic social distancing
Conclusions
• Models suggest that partially effective interventions,
when used in a layered manner, may be highly effective
in controlling the spread of influenza in a community.
• Mitigation strategies appear to be most effective when
implemented in a uniform manner early in an outbreak.
• When used as part of a layered strategy, models suggest
that social distancing measures can have a significant
impact on disease transmission, even if one assumes
low rates of compliance and effectiveness.
What are limits of this data?
• Observational data from 1918; data
incomplete; cannot link cause and effect
• Modeling impact of interventions useful, but
– Doesn’t yet incorporate people’s behavioral
responses to flu itself or to our interventions
– Doesn’t incorporate secondary consequences of
interventions (e.g., effects of school closure on
education, workforce, etc.)
• Does help shape discussion.
Community Mitigation Strategies Carry
Consequences That Should Be Anticipated and
Incorporated into Pandemic Planning
• Economic impact and potential disruption of services due to
absenteeism
• Issues associated with sequestration of children
• Home-based care
• Disproportionate impact on certain populations
• Administration of antiviral medications
– As treatment without rapid diagnostics
– As prophylaxis to household contacts of ill persons
These and other consequences may occur in the absence of
community-wide interventions, as a result of spontaneous
action by the public.
Residences
Workplace / Classroom Social Density
Offices
Hospitals
7.8 feet
Elementary
Schools
11.7 feet
16.2 feet
3.9 feet
http://buildingsdatabook.eren.doe.gov/docs/7.4.4.xls
Spacing of people: If homes were like
schools
*Based on avg. 2,600 sq. ft. per single family home
Spacing of people: If homes were like
schools
*Based on avg. 2,600 sq. ft. per single family home
Households in the United States
Single adult (no children)
2 or more adults (no children)
35.4%
Tw o parents w ith children<18
Single parent w ith children<18
37 million
26.6%
28 million
12 million
28 million
26.3%
11.7%
Source: U.S. Census Bureau, Population Division, Current Population Survey, 2003 Annual Social and Economic Supplement
http://www.census.gov/population/www/socdemo/hh-fam/cps2003.html
Labor Status of Parents
Households with no children<18
Households with children>12
Households with children<12 and non-working adult
Working couple with children<12
62.0%
Single working parent with children<12
66 million
18 million
5 million
8 million
9 million
16.6%
4.5%
8.0%
8.9%
Source: U.S. Census Bureau, Population Division, Current Population Survey, 2003 Annual Social and Economic Supplement
http://www.census.gov/population/www/socdemo/hh-fam/cps2003.html
Macroeconomic Analysis
• Preliminary macroeconomic analyses of the impact of
community-wide interventions have been performed,
using several economic models
• These models predict supply-side impacts that range
from a decrease in overall economic impact as a result
of community-wide interventions, to a modest increase in
impact
• These estimates do not incorporate the costs associated
with lives lost during a severe pandemic
• If an economic value is assigned to lives lost during a
severe pandemic, community-wide interventions result in
a 5-10 fold decrease in overall cost
A Targeted and Layered
Approach
So, Recent Analyses Suggest That Community
Actions May Significantly Reduce Illness and
Death Before Vaccine is Available
Early and uniform implementation of such measures as:
•
•
•
•
School closure
Keeping kids and teens at home
Social distancing at work and in the community
Encouraging voluntary home isolation by ill individuals and
voluntary home quarantine by their household contacts
• Treating the ill and providing targeted antiviral prophylaxis
to household contacts
• Implementing measures early and in a coordinated way
A Layered Approach
Individual / Household /
Agency
Community
Isolation of ill
Hand hygiene
Treatment of ill
Cough etiquette
Quarantine of exposed
Infection control
Prophylaxis of exposed
Living space control
School closure
Isolation of ill
Protective sequestration
Designated care provider
of children
Facemasks
Social distancing
- Community
- Workplace
Liberal leave policies
International
Containment-at-source
Support efforts to reduce
transmission
Travel advisories
Layered screening of
travelers
Health advisories
Limited points of entry
Epidemiology Drives Approach
(Targeted)
Mild
Moderate
Severe
Case Fatality Rate
≤ 0.1%
0.1 - 0.5%
≥ 0.5%
Isolation
Yes
Yes
Yes
Treatment
Yes
Yes
Yes
Quarantine
No
???
Yes
Prophylaxis
High-risk individuals
High-risk individuals
Yes
School Closure
Reactive
Punctuated ???
Proactive
Protective sequestration High-risk individuals
High-risk individuals
Children
Community social
distancing
High-risk individuals
Encouraged
Encouraged +
selective closures
Workplace protections
Encourage good
hygiene
Social distancing
Aggressive social
distancing
Liberal leave policies
Confirmed influenza
Influenza-like illness
ILI and/or sick family
members
Things to consider in choosing
strategies
• Disease severity
• Information on the disease (e.g., are there high risk
subgroups? How effective are antivirals? etc.)
• Ability to practically implement the control measure
• Public acceptability of the control measure
• Secondary impacts of the measure—are we doing
more harm than good?
• What should be implemented by communities and
what centrally? Is a common approach important?
• Ethical considerations
What Can Communities Do Now?
• Education of leadership about the need for
cross-sectoral planning
• Engagement of non-health communities:
education, private sector, labor, faith
communities, NGO’s, the public
• Development of Community-wide plans
• Scenario-based discussions of
implementation
• Plan how to support and protect staff
What does this take?
(Now and when the time comes)
Leadership
Imagination
Resiliency
of Individuals, Agencies, and Communities
Contributors to Historical
Analysis and Modeling
HSC/NSPI Writing Team
Richard Hatchett, MD
Carter Mecher, MD
Laura McClure, MS
CDR Michael Vineyard
HSC
Rajeev Venkayya, MD
Ken Staley, MD, MPA
NSC
Rita DiCasagrande, MS
CEA
Steven Braun, PhD
CDC
David Bell, MD
Martin Cetron, MD
Rachel Eidex, MD
Lisa Koonin, MN, MPH
Anthony Marfin, MD
Modelers
Joshua Epstein, PhD
Stephen Eubank, PhD
Neil Ferguson, PhD
Robert Glass, PhD
Betz Halloran, PhD
Nathaniel Hupert, MD
Marc Lipsitch, MD
Ira Longini, PhD
NIH
James Anderson, PhD
Irene Eckstrand, PhD
Peter Highnam, PhD
Ellis McKenzie, PhD
RTI
Philip Cooley, PhD
Diane Wagener, PhD
NVPO
Bruce Gellin, MD
Ben Schwartz, MD
Department of Education
Camille Welborn, MS
Department of Labor
Suey Howe, JD
Department of the Treasury
Nada Eissa, PhD
Chris Soares, PhD
John Worth, PhD
Department of Finance Canada
Steven James
Timothy Sargent
University of Michigan
Howard Markel, MD
Get Informed,
Be Prepared!
RESOURCES






WVBPH: Div Threat Prep or DSDC
Your Emergency Management Agency and
Local Health Department
http://www.wvflu.org
http://www.pandemicflu.gov
ASTHO (www.astho.org) and NACCHO
(www.naccho.org) Websites
CDC website (www.cdc.gov)

similar documents