Introduction Model Outcome Measures Results Conclusions

Optimizing Influenza Vaccine Distribution

Jan Medlock

Clemson UniversityDepartment of Mathematical Sciences

03 August 2009

Introduction Model Outcome Measures Results Conclusions

12 MAY 2006 VOL 312 SCIENCE www.sciencemag.org854

POLICYFORUM

References and Notes

1. WHO, “Global Polio Eradication Initiative: 2004 Annualreport” (WHO/VB/05.05, WHO UNICEF, IVB DocumentCentre, Geneva, 2005) (http://whqlibdoc.who.int/hq/2005/WHO_POLIO_05.03.pdf).

2. Funding update (www.polioeradication.org/fundingback-ground.asp).

3. F. Fenner, D. A. Henderson, I. Arita, Z. Jezek, I. D. Ladnyi,Smallpox and Its Eradication (WHO, Geneva, 1988).

4. R. B. Aylward, R. W. Sutter, D. L. Heymann, Science 310,625 (2005).

5. O. M. Kew, R. W. Sutter, E. M. de Gourville, W. R. Dowdle,M. A. Pallansch, Annu. Rev. Microbiol. 59, 587 (2005).

6. A. Nomoto, I. Arita, Nat. Immunol. 3, 205 (2002).

7. M. Arita, et al., J. Virol. 79, 12650 (2005).8. J. D. Sachs, in The End of Poverty: Economic Possibilities

for Our Time (Penguin, New York, 2005), chap. 10.9. WHO [press release] (www.who.int/mediacentre/

news/releases/2005/pr49/en/index.html).10. “Can infectious diseases be eradicated? A report on the

International Conference on the Eradication of InfectiousDiseases,” Rev. Infect. Dis. 4 (5), 916 (1982).

11. I. Arita, in The Eradication of Infectious Diseases, W. R.Dowdle, D. R. Hopkins, Eds. (Wiley, New York, 1998),chap. 15.

12. The World Bank, World Development Indicators 2004(World Bank, Washington DC, 2004).

13. “Finance and economics: Recasting the case for aid.”

Economist, 22 January 2005, p. 67.14. D. E. Bloom, D. Canning, M. Weston, World Econ. 6 (3),

15 (2005)15. D. L. Heymann, R. W. Sutter, R. B. Aylward, Nature 434,

699 (2005).16. WHO, UNICEF, “GIVS global immunization vision and

strategy 2006–2015” (WHO/VB/05.05, WHO UNICEF, IVBDocument Centre, Geneva, 2005; (www.who.int/vaccines/GIVS/english/english.htm).

17. We are grateful for the advice from D. A. Henderson, T. Miyamura, and T. Nakano.

10.1126/science.1124959

Rather than thinking only about saving themost lives when considering vaccine rationingstrategies, a better approach would be tomaximize individuals’ life span andopportunity to reach life goals.

Who Should Get Influenza Vaccine

When Not All Can?Ezekiel J. Emanuel* and Alan Wertheimer

PUBLIC HEALTH

The potential threat of pandemic influenza

is staggering: 1.9 million deaths, 90 mil-

lion people sick, and nearly 10 million

people hospitalized, with almost 1.5 million

requiring intensive-care units (ICUs) in the

United States (1). The National Vaccine Advisory

Committee (NVAC) and the Advisory Com-

mittee on Immunization Policy (ACIP) have

jointly recommended a prioritization scheme that

places vaccine workers, health-care providers,

and the ill elderly at the top, and healthy people

aged 2 to 64 at the very bottom, even under

embalmers (1) (see table on page 855). The pri-

mary goal informing the recommendation was to

“decrease health impacts including severe mor-

bidity and death”; a secondary goal was minimiz-

ing societal and economic impacts (1). As the

NVAC and ACIP acknowledge, such important

policy decisions require broad national discus-

sion. In this spirit, we believe an alternative ethi-

cal framework should be considered.

The Inescapability of Rationing

Because of current uncertainty of its value, only

“a limited amount of avian influenza A (H5N1)

vaccine is being stockpiled” (1). Furthermore, it

will take at least 4 months from identification of

a candidate vaccine strain until production of

the very first vaccine (1). At present, there are

few production facilities worldwide that make

influenza vaccine, and only one completely in

the USA. Global capacity for influenza vaccine

production is just 425 million doses per annum,

if all available factories would run at full capac-

ity after a vaccine was developed. Under cur-

rently existing capabilities for manufacturing

vaccine, it is likely that more than 90% of the

U.S. population will not be vaccinated in the

first year (1). Distributing the limited supply

will require determining priority groups.

Who will be at highest risk? Our experience

with three influenza pandemics presents a com-

plex picture. The mortality profile of a future

pandemic could be U-shaped, as it was in the

mild-to-moderate pandemics of 1957 and 1968

and interpandemic influenza seasons, in which

the very young and the old are at highest risk.

Or, the mortality profile could be an attenuated

W shape, as it was during the devastating 1918

pandemic, in which the highest risk occurred

among people between 20 and 40 years of age,

while the elderly were not at high excess risk

(2, 3). Even during pandemics, the elderly

appear to be at no higher risk than during inter-

pandemic influenza seasons (4).

Clear ethical justification for vaccine prior-

ities is essential to the acceptability of the pri-

ority ranking and any modifications during the

pandemic. With limited vaccine supply, uncer-

tainty over who will be at highest risk of infec-

tion and complications, and questions about

which historic pandemic experience is most

applicable, society faces a fundamental ethical

dilemma: Who should get the vaccine first?

The NVAC and ACIP Priority Rankings

Many potential ethical principles for rationing

health care have been proposed. “Save the most

lives” is commonly used in emergencies, such

as burning buildings, although “women and

children first” played a role on the Titanic. “First

come, first served” operates in other emergen-

cies and in ICUs when admitted patients retain

beds despite the presentation of another patient

who is equally or even more sick; “Save the

most quality life years” is central to cost-effec-

tiveness rationing. “Save the worst-off ”

plays a role in allocating organs for transplan-

tation. “Reciprocity”—giving priority to people

willing to donate their own organs—has been

proposed. “Save those most likely to fully

recover” guided priorities for giving penicillin

to soldiers with syphilis in World War II. Save

those “instrumental in making society flourish”

through economic productivity or by “con-

tributing to the well-being of others” has been

proposed by Murray and others (5, 6).

The save-the-most-lives principle was

invoked by NVAC and ACIP. It justifies giving

top priority to workers engaged in vaccine pro-

duction and distribution and health-care work-

ers. They get higher priority not because they

are intrinsically more valuable people or of

greater “social worth,” but because giving them

first priority ensures that maximal life-saving

vaccine is produced and so that health care is

provided to the sick (7). Consequently, it values

all human life equally, giving every person

equal consideration in who gets priority regard-

less of age, disability, social class, or employ-

ment (7). After these groups, the save-the-most-

lives principle justifies priority for those pre-

dicted to be at highest risk of hospitalization and

dying. We disagree with this prioritization.

Life-Cycle Principle

The save-the-most-lives principle may be justi-

fied in some emergencies when decision

urgency makes it infeasible to deliberate about

priority rankings and impractical to categorize

individuals into priority groups. We believe that

a life-cycle allocation principle (see table on

page 855) based on the idea that each person

should have an opportunity to live through all

Department of Clinical Bioethics, The Clinical Center,National Institutes of Health, Bethesda, MD 20892–1156,USA.

The opinions expressed are the authors’ and do not reflectthe policies of the National Institutes of Health, the PublicHealth Service, or the Department of Health and HumanServices.

*Author for correspondence. E-mail: eemanuel@nih.gov

Published by AAAS

Science 2006

• Should value people “on the basis of the amount the personinvested in his or her life balanced by the amount left to live.”

• Then vaccinate the most-valued people!• Misses epidemiology: Transmission, Case mortality, Vaccine

efficacy

Introduction Model Outcome Measures Results Conclusions

12 MAY 2006 VOL 312 SCIENCE www.sciencemag.org854

POLICYFORUM

References and Notes

1. WHO, “Global Polio Eradication Initiative: 2004 Annualreport” (WHO/VB/05.05, WHO UNICEF, IVB DocumentCentre, Geneva, 2005) (http://whqlibdoc.who.int/hq/2005/WHO_POLIO_05.03.pdf).

2. Funding update (www.polioeradication.org/fundingback-ground.asp).

3. F. Fenner, D. A. Henderson, I. Arita, Z. Jezek, I. D. Ladnyi,Smallpox and Its Eradication (WHO, Geneva, 1988).

4. R. B. Aylward, R. W. Sutter, D. L. Heymann, Science 310,625 (2005).

5. O. M. Kew, R. W. Sutter, E. M. de Gourville, W. R. Dowdle,M. A. Pallansch, Annu. Rev. Microbiol. 59, 587 (2005).

6. A. Nomoto, I. Arita, Nat. Immunol. 3, 205 (2002).

7. M. Arita, et al., J. Virol. 79, 12650 (2005).8. J. D. Sachs, in The End of Poverty: Economic Possibilities

for Our Time (Penguin, New York, 2005), chap. 10.9. WHO [press release] (www.who.int/mediacentre/

news/releases/2005/pr49/en/index.html).10. “Can infectious diseases be eradicated? A report on the

International Conference on the Eradication of InfectiousDiseases,” Rev. Infect. Dis. 4 (5), 916 (1982).

11. I. Arita, in The Eradication of Infectious Diseases, W. R.Dowdle, D. R. Hopkins, Eds. (Wiley, New York, 1998),chap. 15.

12. The World Bank, World Development Indicators 2004(World Bank, Washington DC, 2004).

13. “Finance and economics: Recasting the case for aid.”

Economist, 22 January 2005, p. 67.14. D. E. Bloom, D. Canning, M. Weston, World Econ. 6 (3),

15 (2005)15. D. L. Heymann, R. W. Sutter, R. B. Aylward, Nature 434,

699 (2005).16. WHO, UNICEF, “GIVS global immunization vision and

strategy 2006–2015” (WHO/VB/05.05, WHO UNICEF, IVBDocument Centre, Geneva, 2005; (www.who.int/vaccines/GIVS/english/english.htm).

17. We are grateful for the advice from D. A. Henderson, T. Miyamura, and T. Nakano.

10.1126/science.1124959

Rather than thinking only about saving themost lives when considering vaccine rationingstrategies, a better approach would be tomaximize individuals’ life span andopportunity to reach life goals.

Who Should Get Influenza Vaccine

When Not All Can?Ezekiel J. Emanuel* and Alan Wertheimer

PUBLIC HEALTH

The potential threat of pandemic influenza

is staggering: 1.9 million deaths, 90 mil-

lion people sick, and nearly 10 million

people hospitalized, with almost 1.5 million

requiring intensive-care units (ICUs) in the

United States (1). The National Vaccine Advisory

Committee (NVAC) and the Advisory Com-

mittee on Immunization Policy (ACIP) have

jointly recommended a prioritization scheme that

places vaccine workers, health-care providers,

and the ill elderly at the top, and healthy people

aged 2 to 64 at the very bottom, even under

embalmers (1) (see table on page 855). The pri-

mary goal informing the recommendation was to

“decrease health impacts including severe mor-

bidity and death”; a secondary goal was minimiz-

ing societal and economic impacts (1). As the

NVAC and ACIP acknowledge, such important

policy decisions require broad national discus-

sion. In this spirit, we believe an alternative ethi-

cal framework should be considered.

The Inescapability of Rationing

Because of current uncertainty of its value, only

“a limited amount of avian influenza A (H5N1)

vaccine is being stockpiled” (1). Furthermore, it

will take at least 4 months from identification of

a candidate vaccine strain until production of

the very first vaccine (1). At present, there are

few production facilities worldwide that make

influenza vaccine, and only one completely in

the USA. Global capacity for influenza vaccine

production is just 425 million doses per annum,

if all available factories would run at full capac-

ity after a vaccine was developed. Under cur-

rently existing capabilities for manufacturing

vaccine, it is likely that more than 90% of the

U.S. population will not be vaccinated in the

first year (1). Distributing the limited supply

will require determining priority groups.

Who will be at highest risk? Our experience

with three influenza pandemics presents a com-

plex picture. The mortality profile of a future

pandemic could be U-shaped, as it was in the

mild-to-moderate pandemics of 1957 and 1968

and interpandemic influenza seasons, in which

the very young and the old are at highest risk.

Or, the mortality profile could be an attenuated

W shape, as it was during the devastating 1918

pandemic, in which the highest risk occurred

among people between 20 and 40 years of age,

while the elderly were not at high excess risk

(2, 3). Even during pandemics, the elderly

appear to be at no higher risk than during inter-

pandemic influenza seasons (4).

Clear ethical justification for vaccine prior-

ities is essential to the acceptability of the pri-

ority ranking and any modifications during the

pandemic. With limited vaccine supply, uncer-

tainty over who will be at highest risk of infec-

tion and complications, and questions about

which historic pandemic experience is most

applicable, society faces a fundamental ethical

dilemma: Who should get the vaccine first?

The NVAC and ACIP Priority Rankings

Many potential ethical principles for rationing

health care have been proposed. “Save the most

lives” is commonly used in emergencies, such

as burning buildings, although “women and

children first” played a role on the Titanic. “First

come, first served” operates in other emergen-

cies and in ICUs when admitted patients retain

beds despite the presentation of another patient

who is equally or even more sick; “Save the

most quality life years” is central to cost-effec-

tiveness rationing. “Save the worst-off ”

plays a role in allocating organs for transplan-

tation. “Reciprocity”—giving priority to people

willing to donate their own organs—has been

proposed. “Save those most likely to fully

recover” guided priorities for giving penicillin

to soldiers with syphilis in World War II. Save

those “instrumental in making society flourish”

through economic productivity or by “con-

tributing to the well-being of others” has been

proposed by Murray and others (5, 6).

The save-the-most-lives principle was

invoked by NVAC and ACIP. It justifies giving

top priority to workers engaged in vaccine pro-

duction and distribution and health-care work-

ers. They get higher priority not because they

are intrinsically more valuable people or of

greater “social worth,” but because giving them

first priority ensures that maximal life-saving

vaccine is produced and so that health care is

provided to the sick (7). Consequently, it values

all human life equally, giving every person

equal consideration in who gets priority regard-

less of age, disability, social class, or employ-

ment (7). After these groups, the save-the-most-

lives principle justifies priority for those pre-

dicted to be at highest risk of hospitalization and

dying. We disagree with this prioritization.

Life-Cycle Principle

The save-the-most-lives principle may be justi-

fied in some emergencies when decision

urgency makes it infeasible to deliberate about

priority rankings and impractical to categorize

individuals into priority groups. We believe that

a life-cycle allocation principle (see table on

page 855) based on the idea that each person

should have an opportunity to live through all

Department of Clinical Bioethics, The Clinical Center,National Institutes of Health, Bethesda, MD 20892–1156,USA.

The opinions expressed are the authors’ and do not reflectthe policies of the National Institutes of Health, the PublicHealth Service, or the Department of Health and HumanServices.

*Author for correspondence. E-mail: eemanuel@nih.gov

Published by AAAS

Science 2006

• Should value people “on the basis of the amount the personinvested in his or her life balanced by the amount left to live.”

• Then vaccinate the most-valued people!• Misses epidemiology: Transmission, Case mortality, Vaccine

efficacy

Introduction Model Outcome Measures Results Conclusions

12 MAY 2006 VOL 312 SCIENCE www.sciencemag.org854

POLICYFORUM

References and Notes

1. WHO, “Global Polio Eradication Initiative: 2004 Annualreport” (WHO/VB/05.05, WHO UNICEF, IVB DocumentCentre, Geneva, 2005) (http://whqlibdoc.who.int/hq/2005/WHO_POLIO_05.03.pdf).

2. Funding update (www.polioeradication.org/fundingback-ground.asp).

3. F. Fenner, D. A. Henderson, I. Arita, Z. Jezek, I. D. Ladnyi,Smallpox and Its Eradication (WHO, Geneva, 1988).

4. R. B. Aylward, R. W. Sutter, D. L. Heymann, Science 310,625 (2005).

5. O. M. Kew, R. W. Sutter, E. M. de Gourville, W. R. Dowdle,M. A. Pallansch, Annu. Rev. Microbiol. 59, 587 (2005).

6. A. Nomoto, I. Arita, Nat. Immunol. 3, 205 (2002).

7. M. Arita, et al., J. Virol. 79, 12650 (2005).8. J. D. Sachs, in The End of Poverty: Economic Possibilities

for Our Time (Penguin, New York, 2005), chap. 10.9. WHO [press release] (www.who.int/mediacentre/

news/releases/2005/pr49/en/index.html).10. “Can infectious diseases be eradicated? A report on the

International Conference on the Eradication of InfectiousDiseases,” Rev. Infect. Dis. 4 (5), 916 (1982).

11. I. Arita, in The Eradication of Infectious Diseases, W. R.Dowdle, D. R. Hopkins, Eds. (Wiley, New York, 1998),chap. 15.

12. The World Bank, World Development Indicators 2004(World Bank, Washington DC, 2004).

13. “Finance and economics: Recasting the case for aid.”

Economist, 22 January 2005, p. 67.14. D. E. Bloom, D. Canning, M. Weston, World Econ. 6 (3),

15 (2005)15. D. L. Heymann, R. W. Sutter, R. B. Aylward, Nature 434,

699 (2005).16. WHO, UNICEF, “GIVS global immunization vision and

strategy 2006–2015” (WHO/VB/05.05, WHO UNICEF, IVBDocument Centre, Geneva, 2005; (www.who.int/vaccines/GIVS/english/english.htm).

17. We are grateful for the advice from D. A. Henderson, T. Miyamura, and T. Nakano.

10.1126/science.1124959

Rather than thinking only about saving themost lives when considering vaccine rationingstrategies, a better approach would be tomaximize individuals’ life span andopportunity to reach life goals.

Who Should Get Influenza Vaccine

When Not All Can?Ezekiel J. Emanuel* and Alan Wertheimer

PUBLIC HEALTH

The potential threat of pandemic influenza

is staggering: 1.9 million deaths, 90 mil-

lion people sick, and nearly 10 million

people hospitalized, with almost 1.5 million

requiring intensive-care units (ICUs) in the

United States (1). The National Vaccine Advisory

Committee (NVAC) and the Advisory Com-

mittee on Immunization Policy (ACIP) have

jointly recommended a prioritization scheme that

places vaccine workers, health-care providers,

and the ill elderly at the top, and healthy people

aged 2 to 64 at the very bottom, even under

embalmers (1) (see table on page 855). The pri-

mary goal informing the recommendation was to

“decrease health impacts including severe mor-

bidity and death”; a secondary goal was minimiz-

ing societal and economic impacts (1). As the

NVAC and ACIP acknowledge, such important

policy decisions require broad national discus-

sion. In this spirit, we believe an alternative ethi-

cal framework should be considered.

The Inescapability of Rationing

Because of current uncertainty of its value, only

“a limited amount of avian influenza A (H5N1)

vaccine is being stockpiled” (1). Furthermore, it

will take at least 4 months from identification of

a candidate vaccine strain until production of

the very first vaccine (1). At present, there are

few production facilities worldwide that make

influenza vaccine, and only one completely in

the USA. Global capacity for influenza vaccine

production is just 425 million doses per annum,

if all available factories would run at full capac-

ity after a vaccine was developed. Under cur-

rently existing capabilities for manufacturing

vaccine, it is likely that more than 90% of the

U.S. population will not be vaccinated in the

first year (1). Distributing the limited supply

will require determining priority groups.

Who will be at highest risk? Our experience

with three influenza pandemics presents a com-

plex picture. The mortality profile of a future

pandemic could be U-shaped, as it was in the

mild-to-moderate pandemics of 1957 and 1968

and interpandemic influenza seasons, in which

the very young and the old are at highest risk.

Or, the mortality profile could be an attenuated

W shape, as it was during the devastating 1918

pandemic, in which the highest risk occurred

among people between 20 and 40 years of age,

while the elderly were not at high excess risk

(2, 3). Even during pandemics, the elderly

appear to be at no higher risk than during inter-

pandemic influenza seasons (4).

Clear ethical justification for vaccine prior-

ities is essential to the acceptability of the pri-

ority ranking and any modifications during the

pandemic. With limited vaccine supply, uncer-

tainty over who will be at highest risk of infec-

tion and complications, and questions about

which historic pandemic experience is most

applicable, society faces a fundamental ethical

dilemma: Who should get the vaccine first?

The NVAC and ACIP Priority Rankings

Many potential ethical principles for rationing

health care have been proposed. “Save the most

lives” is commonly used in emergencies, such

as burning buildings, although “women and

children first” played a role on the Titanic. “First

come, first served” operates in other emergen-

cies and in ICUs when admitted patients retain

beds despite the presentation of another patient

who is equally or even more sick; “Save the

most quality life years” is central to cost-effec-

tiveness rationing. “Save the worst-off ”

plays a role in allocating organs for transplan-

tation. “Reciprocity”—giving priority to people

willing to donate their own organs—has been

proposed. “Save those most likely to fully

recover” guided priorities for giving penicillin

to soldiers with syphilis in World War II. Save

those “instrumental in making society flourish”

through economic productivity or by “con-

tributing to the well-being of others” has been

proposed by Murray and others (5, 6).

The save-the-most-lives principle was

invoked by NVAC and ACIP. It justifies giving

top priority to workers engaged in vaccine pro-

duction and distribution and health-care work-

ers. They get higher priority not because they

are intrinsically more valuable people or of

greater “social worth,” but because giving them

first priority ensures that maximal life-saving

vaccine is produced and so that health care is

provided to the sick (7). Consequently, it values

all human life equally, giving every person

equal consideration in who gets priority regard-

less of age, disability, social class, or employ-

ment (7). After these groups, the save-the-most-

lives principle justifies priority for those pre-

dicted to be at highest risk of hospitalization and

dying. We disagree with this prioritization.

Life-Cycle Principle

The save-the-most-lives principle may be justi-

fied in some emergencies when decision

urgency makes it infeasible to deliberate about

priority rankings and impractical to categorize

individuals into priority groups. We believe that

a life-cycle allocation principle (see table on

page 855) based on the idea that each person

should have an opportunity to live through all

Department of Clinical Bioethics, The Clinical Center,National Institutes of Health, Bethesda, MD 20892–1156,USA.

The opinions expressed are the authors’ and do not reflectthe policies of the National Institutes of Health, the PublicHealth Service, or the Department of Health and HumanServices.

*Author for correspondence. E-mail: eemanuel@nih.gov

Published by AAAS

Science 2006

• Should value people “on the basis of the amount the personinvested in his or her life balanced by the amount left to live.”

• Then vaccinate the most-valued people!• Misses epidemiology: Transmission, Case mortality, Vaccine

efficacy

Introduction Model Outcome Measures Results Conclusions

Problem Setup

• For influenza• Age structure but not risk or occupation• Given an outcome measure• How to distribute limited vaccine doses?• Nonlinear constrained optimization

Introduction Model Outcome Measures Results Conclusions

Model

URUS UIUE

VRVIVS VE

νU

γλ

νV

(1− ε)λ

τ

τ γ

Age structured (0, 1–4, 5–9, 10–14, 15–19, . . . , 70–74, 75+)No birth or natural death

Introduction Model Outcome Measures Results Conclusions

2007 US Population Age StructureNu

mber

Age (years)

0M

1M

2M

3M

4M

5M

0 20 40 60 80 100

Sources: US Census, US Census.

Introduction Model Outcome Measures Results Conclusions

Parameters

Parameter Ages Value RefLatent period, 1/τ all 1.2 d [1]

Infectious period, 1/γ all 4.1 d [1]

Vaccine efficacy 0–64 0.80 [2, 3]

against infection, εa 65+ 0.60Vaccine efficacy 0–19 0.75against death 20–64 0.70 [4, 2]

65+ 0.60[1] Longini et al, Science, 2005; [2] Galvani, Reluga, & Chapman, PNAS, 2007;[3] CDC, ACIP, 2007; [4] Meltzer, Cox, & Fukuda, Emerg Infect Dis, 1999.

Introduction Model Outcome Measures Results Conclusions

Death RateInflu

enza

deathrate

(per

day)

Age (years)

1957, unvaccinated1957, vaccinated

1918, unvaccinated1918, vaccinated

0.000

0.002

0.004

0.006

0.008

0.010

0 20 40 60 80

Sources: Serfling, Sherman, & Houseworth, Am J Epidemiol, 1967; Luk, Gross, & Thompson, Clin Infect Dis, 2001;Glezen, Epidemiol Rev, 1996.

Introduction Model Outcome Measures Results Conclusions

Contacts

Social Contacts and Mixing Patterns Relevant tothe Spread of Infectious DiseasesJoel Mossong

1,2*, Niel Hens

3, Mark Jit

4, Philippe Beutels

5, Kari Auranen

6, Rafael Mikolajczyk

7, Marco Massari

8,

Stefania Salmaso8

, Gianpaolo Scalia Tomba9

, Jacco Wallinga10

, Janneke Heijne10

, Malgorzata Sadkowska-Todys11

,

Magdalena Rosinska11

, W. John Edmunds4

1 Microbiology Unit, Laboratoire National de Sante, Luxembourg, Luxembourg, 2 Centre de Recherche Public Sante, Luxembourg, Luxembourg, 3 Center for Statistics,

Hasselt University, Diepenbeek, Belgium, 4 Modelling and Economics Unit, Health Protection Agency Centre for Infections, London, United Kingdom, 5 Unit Health Economic

and Modeling Infectious Diseases, Center for the Evaluation of Vaccination, Vaccine & Infectious Disease Institute, University of Antwerp, Antwerp, Belgium, 6 Department of

Vaccines, National Public Health Institute KTL, Helsinki, Finland, 7 School of Public Health, University of Bielefeld, Bielefeld, Germany, 8 Istituto Superiore di Sanita, Rome,

Italy, 9 Department of Mathematics, University of Rome Tor Vergata, Rome, Italy, 10 Centre for Infectious Disease Control Netherlands, National Institute for Public Health

and the Environment, Bilthoven, The Netherlands, 11 National Institute of Hygiene, Warsaw, Poland

Funding: This study formed part ofPOLYMOD, a European Commissionproject funded within the SixthFramework Programme, Contractnumber: SSP22-CT-2004–502084.The funders had no role in studydesign, data collection and analysis,decision to publish, or preparationof the manuscript.

Competing Interests: The authorshave declared that no competinginterests exist.

Academic Editor: Steven Riley,Hong Kong University, Hong Kong

Citation: Mossong J, Hens N, Jit M,Beutels P, Auranen K, et al. (2008)Social contacts and mixing patternsrelevant to the spread of infectiousdiseases. PLoS Med 5(3) e74. doi:10.1371/journal.pmed.0050074

Received: August 8, 2007Accepted: February 15, 2008Published: March 25, 2008

Copyright: � 2008 Mossong et al.This is an open-access articledistributed under the terms of theCreative Commons AttributionLicense, which permits unrestricteduse, distribution, and reproductionin any medium, provided theoriginal author and source arecredited.

Abbreviations: BE, Belgium; DE,Germany; FI, Finland; GB, GreatBritain; IT, Italy; LU, Luxembourg; NL,The Netherlands; PL, Poland

* To whom correspondence shouldbe addressed. E-mail: joel.mossong@lns.etat.lu

A B S T R A C T

Background

Mathematical modelling of infectious diseases transmitted by the respiratory or close-contactroute (e.g., pandemic influenza) is increasingly being used to determine the impact of possibleinterventions. Although mixing patterns are known to be crucial determinants for modeloutcome, researchers often rely on a priori contact assumptions with little or no empirical basis.We conducted a population-based prospective survey of mixing patterns in eight Europeancountries using a common paper-diary methodology.

Methods and Findings

7,290 participants recorded characteristics of 97,904 contacts with different individualsduring one day, including age, sex, location, duration, frequency, and occurrence of physicalcontact. We found that mixing patterns and contact characteristics were remarkably similaracross different European countries. Contact patterns were highly assortative with age:schoolchildren and young adults in particular tended to mix with people of the same age.Contacts lasting at least one hour or occurring on a daily basis mostly involved physicalcontact, while short duration and infrequent contacts tended to be nonphysical. Contacts athome, school, or leisure were more likely to be physical than contacts at the workplace or whiletravelling. Preliminary modelling indicates that 5- to 19-year-olds are expected to suffer thehighest incidence during the initial epidemic phase of an emerging infection transmittedthrough social contacts measured here when the population is completely susceptible.

Conclusions

To our knowledge, our study provides the first large-scale quantitative approach to contactpatterns relevant for infections transmitted by the respiratory or close-contact route, and theresults should lead to improved parameterisation of mathematical models used to designcontrol strategies.

The Editors’ Summary of this article follows the references.

PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e740381

PLoSMEDICINE

PLoS Med 2008

Surveyed 7,290 Europeans for daily contacts

Introduction Model Outcome Measures Results Conclusions

Contacts

0–4

5–9

10–14

15–19

20–24

25–29

30–34

35–39

40–44

45–49

50–54

55–59

60–64

65–69

70+

Age (years)

0–45–910–1415–1920–2425–2930–3435–3940–4445–4950–5455–5960–6465–6970+

Age(years)

10

100

Contactr

ate(per

person

perd

ay)

Introduction Model Outcome Measures Results Conclusions

R0

• R0 = 1.4 for Swine Flu (Fraser et al, Science, 2009)

• R0 = 2.0 for 1918 Pandemic (Mills et al, Nature, 2004)

• We considered R0 = 1.4 and also R0 = 1.2, 1.6, 1.8, 2.0

Introduction Model Outcome Measures Results Conclusions

Outcome Measures

Map outcome (number infected, dead, etc) to objective• Total Infections• Total Deaths• Years of Life Lost: Using expectation of life (NCHS, US Life Tables, 2003)

• Contingent Valuation: Indirect assessment of value of lives ofdifferent ages

• Total Cost: Converts deaths, infections, etc into dollars

Introduction Model Outcome Measures Results Conclusions

Contingent Valuation

• Survey asked about20, 30, 40, 60 yearolds and fit

va = aω−1 exp (−ψaω)

(Cropper et al, J Risk Uncertain,

1994)

• Alternative:wage–risk marketdata, but only forworking-aged adults

Relativ

edisutility

ofdeath

Age (years)

0.0

0.2

0.4

0.6

0.8

1.0

0 20 40 60 80

Introduction Model Outcome Measures Results Conclusions

Total Cost

• Monetary cost ofillness (Meltzer, Cox, &

Fukuda, Emerg Infect Dis, 1999)

• Monetary cost ofdeath

• Future lifetimeearnings (Haddix et

al, 1996)

• Alternatives:Include value ofnon-work time

Future

lifetim

eearnings

Age (years)

0.0

0.5

1.0

1.5

0 20 40 60 80

Introduction Model Outcome Measures Results Conclusions

Outcome MeasuresRe

lativ

edisutility

ofdeath

Age (years)

Total DeathsYears of Life Lost

Contingent ValuationTotal Cost

0.0

0.2

0.4

0.6

0.8

1.0

0 20 40 60 80

Introduction Model Outcome Measures Results Conclusions

No VaccinationNu

mberinfected

Time (days)

19571918

0M

2M

4M

6M

8M

10M

0 60 120 180 240 300 360

Introduction Model Outcome Measures Results Conclusions

Current Vaccination

CDC estimate• 84M doses used in

2007• 100M+ doses

annually• 600M doses for Swine

FluVa

ccinecoverage

Age (years)

0%

20%

40%

60%

0 20 40 60 80

Sources: CDC, ACIP, 2008; NHIS, 2007.

Introduction Model Outcome Measures Results Conclusions

EradicationEradica

tiondo

ses

R0

19571918

0

25

50

75

100

125

150

1 1.2 1.4 1.6 1.8 2

Introduction Model Outcome Measures Results Conclusions

1957-like Mortality

0M

20M

40M

60M

InfectsDeaths

YLLCV Cost

InfectsDeaths

YLLCV Cost

InfectsDeaths

YLLCV Cost

Numbero

fdoses

20M Doses 40M Doses 60M Doses

5–910–1415–19

20–2430–3435–39

45–4965–6975+

Introduction Model Outcome Measures Results Conclusions

1918-like Mortality

0M

20M

40M

60M

InfectsDeaths

YLLCV Cost

InfectsDeaths

YLLCV Cost

InfectsDeaths

YLLCV Cost

Numbero

fdoses

20M Doses 40M Doses 60M Doses

5–910–1415–19

20–2430–3435–39

45–4965–6975+

Introduction Model Outcome Measures Results Conclusions

1957-like MortalityInfections

0.0

0.5

1.0

Deaths

0.0

0.5

1.0

YLL

0.0

0.5

1.0

CV5–9

10–1415–1920–2425–2930–34

35–3945–4965–6970–7475+

0.0

0.5

1.0

0M 20M 40M 60M

Cost

Vaccine doses

0.0

0.5

1.0

0M 20M 40M 60M

Introduction Model Outcome Measures Results Conclusions

1918-like MortalityInfections

0.0

0.5

1.0

Deaths

0.0

0.5

1.0

YLL

0.0

0.5

1.0

CVVaccine doses

5–910–1415–19

20–2430–3435–39

0.0

0.5

1.0

0M 20M 40M 60M

Cost

Vaccine doses

0.0

0.5

1.0

0M 20M 40M 60M

Introduction Model Outcome Measures Results Conclusions

R0 = 2.0, 1957-like MortalityInfections

0.0

0.5

1.0

Deaths

0.0

0.5

1.0

YLL

0.0

0.5

1.0

CVVaccine doses

0.0

0.5

1.0

0M 50M 100M

Cost

Vaccine doses

0.0

0.5

1.0

0M 50M 100M

Introduction Model Outcome Measures Results Conclusions

R0 = 2.0, 1918-like MortalityInfections

0.0

0.5

1.0

Deaths

0.0

0.5

1.0

YLL

0.0

0.5

1.0

CV0

5–910–1415–1920–24

25–2930–3435–3940–44

0.0

0.5

1.0

0M 50M 100M

Cost

Vaccine doses

0.0

0.5

1.0

0M 50M 100M

Introduction Model Outcome Measures Results Conclusions

Sensitivity Analysis

• Reduced vaccine efficacy against infectionShifts to protecting at risk

• Reduced vaccine efficacy against deathReduced susceptibility in elderlyReduced infectious period for vaccineesReduced infectiousness for vaccineesLittle change for 50% reduction

Introduction Model Outcome Measures Results Conclusions

1957-like Mortality, 40M Doses

0%

20%

40%

60%

InfectsDeaths

YLLCV Cost

Redu

ction

OptimalCurrentUniform

Former CDCSeasonalPandemic

Ages 5–19

Introduction Model Outcome Measures Results Conclusions

1918-like Mortality, 40M Doses

0%

20%

40%

60%

InfectsDeaths

YLLCV Cost

Redu

ction

OptimalCurrentUniform

Former CDCSeasonalPandemic

Ages 5–19

Introduction Model Outcome Measures Results Conclusions

Conclusions

• 65M doses prevents an R0 = 1.4 epidemic• 135M doses prevents an R0 = 2.0 epidemic• Can improve vaccination policies• Infections: Vaccinate transmitters, children (5–19) & parents

(30–39)• Deaths, YLL, Contingent, & Cost:

• When vaccine limited, vaccinate those at risk of death• When vaccine plentiful, vaccinate transmitters• Transition varies between outcome measures• Deaths averted transitions last

• Joint work with Alison GalvaniFunded by NSF grant SBE-0624117