Evolution of ACIP Influenza Vaccination Recommendations:

Promise and Challenge

Overview of the Path to Expanded

Recommendations

Anthony Fiore, MD, MPH

Influenza Division, NCIRD, CDC

NACCHO Meeting

AtlantaJuly 14, 2008

• ACIP recommendations up to 2008 and vaccine coverage

• Rationale and decision process: Expanding vaccine recommendations to school age children

• Challenges in measuring impact

Presentation Overview

The findings and conclusions in this presentation are those of the

author and do not necessarily represent the views of

the Centers for Disease Control and Prevention

Recommendation Changes for Influenza Vaccination: Milestones through 2007

Before 2000: Persons aged 65 or olderPersons with chronic medical conditions that make them

more likely to have complications of influenza Pregnant women in the second or third trimesterContacts (household and out of home caregivers) of the above groupsHealthcare workers

2000: Adults 50 and older

2004: Children aged 6--23 monthsContacts (household and out of home caregivers) of

children aged 0--23 monthsWomen who will be pregnant during influenza season

2006: Children aged 6--59 monthsContacts (household and out of home caregivers) of

children aged 0-59 months

0

10

20

30

40

50

60

70

80

90

100

Year

Co

vera

ge L

evel (%

)

>=65 yrs

Healthy 50-64yrs

Pregnantwomen

Health-careworkers

Self-Reported Influenza Vaccination Coverage Levels Among Selected Priority U.S. Adult Populations,

1989-2007*, National Health Interview Survey

Vaccine shortage: 2004-05 season

Source: CDC, NHIS. http://www.cdc.gov/flu/professionals/vaccination/pdf/vaccinetrend.pdf

*Preliminary data from 2006-07 influenza season

Estimates of Subpopulations with a Vaccine Indication, and Vaccine Coverage, 2006

Adapted from source: Immunization Services Division, CDC.

http://www.cdc.gov/flu/professionals/vaccination/pdf/targetpopchart.pdf

Options to Improve Vaccine Coverage

• Improve vaccination of existing target groups– Public awareness– Provider education and practices

Options to Improve Vaccine Coverage

• Improve vaccination of existing target groups– Public awareness– Provider education and practices

• Work toward universal vaccination recommendation incrementally – Begin with school age children– Strengthen adult vaccination efforts

Moving Towards Universal Vaccination against Influenza: Increasing Interest

• Better understanding of health and economic impact of influenza among older children and adults

• Recognition of suboptimal vaccine effectiveness among groups at highest risk for influenza complications (e.g., elderly, persons with chronic illness)– Difficult to show substantial impact on morbidity and mortality in

these groups

• Lessened concerns about vaccine supply

Moving Towards Universal Vaccination against Influenza: Other Potential Benefits

• Could current low coverage for most recommended groups be improved by a universal recommendation?– Approximately 50% of school age children already had in indication

for vaccination in the 2007-08 season

• Could vaccinating school children and healthy adults reduce illness among their contacts, including those at higher risk for influenza complications?

• Could routinely vaccinating everyone against influenza help in planning for a response to a pandemic or other large scale public health event?

ACIP Influenza Vaccine Workgroup

• Members: 3-5 ACIP voting members, ex officio members (FDA, NIH) and liaison organizations (AAP, AMA, AAFP, ACP, NAACHO, AIM, etc.)

• Teleconference (60-90 minutes) at least monthly

• Email discussions and ad hoc teleconferences

• Updates to full ACIP thrice yearly

• Vaccine supply: Adequate and improving, although local distribution issues remain problematic

• Vaccine safety: Established, but need for continued vigilance and long term studies

• Vaccine effectiveness: Established effectiveness (50-90%) in reducing influenza illness, and increasing evidence for indirect effects

• Disease burden: Highest rates of influenza but severe outcomes less common than in older or younger age groups

• Cost-effectiveness: Higher than many currently recommended vaccines but models do not fully account for potential indirect effects

• Feasibility of sustained implementation: Uncertain, but comprehensive efforts to vaccinate this large cohort are not likely to be established until a recommendation is made

*Based on workgroup teleconferences and

CDC/CSTE consultation, September 2007

Workgroup Conclusions: Vaccinating School Age Children Against Influenza*

ACIP Influenza Vaccine Workgroup: Rationale for Expanding Vaccination Recommendations to Include all

School-age Children and Adolescents*Rationale

• Evidence that influenza has substantial adverse impacts among school age children and their contacts (e.g., increased school absenteeism, antibiotic use, medical care visits, and parental work loss)

• Evidence that influenza vaccine is effective and safe for school-age children

• The expectation that a simple age-based influenza vaccine recommendation will improve current low vaccine coverage levels among the approximately 50% of school-age children who already had a risk- or contact-based indication for annual influenza vaccination

*Approved at February 27, 2008 ACIP meeting

ACIP Influenza Vaccine Workgroup: Rationale for Expanding Vaccination Recommendations to Include all

School-age Children and Adolescents*Rationale

• Evidence that influenza has substantial adverse impacts among school age children and their contacts (e.g., increased school absenteeism, antibiotic use, medical care visits, and parental work loss)

• Evidence that influenza vaccine is effective and safe for school-age children

• The expectation that a simple age-based influenza vaccine recommendation will improve current low vaccine coverage levels among the approximately 50% of school-age children who already had a risk- or contact-based indication for annual influenza vaccination

Also noted

• The potential for the indirect effect of reducing influenza among persons who have close contact with children, and reducing overall transmission within communities, if sufficient vaccination coverage among children can be achieved

*Approved at February 27, 2008 ACIP meeting

ACIP: Influenza Vaccination Recommendations for Children

All children aged 6 months through 18 years should receive annual influenza vaccination, beginning in 2008 if feasible, but beginning no later than during the 2009-2010 influenza season

*Approved at February 27, 2008 ACIP meeting

Challenges to Implementation of a School Age Children Recommendation

• Who will administer vaccine and where will vaccine be given?

• How will non-medical venues be reimbursed?

• Will healthcare practitioners and parents accept vaccination efforts outside medical home?

• Can vaccinations given outside medical settings successfully link records with the medical home, public health, and vaccine registries?

• Will efforts to vaccinate all school-age children shift focus away from children at higher risk for influenza complications (e.g., infants, older children with chronic medical conditions)?

• How can we measure impact other than vaccine coverage?

Recent Influenza Vaccine Coverage Data among Young Children, United

States

Source: L Williams, NCIRD/ISD/IISSB

G Euler, NCIRD/ISD/AB

0

20

40

60

80

100

OR MI AZ

Fully vaccinated, 2004-5

Fully vaccinated, 2005-6

Fully vaccinated, 2006-7

Fully vaccinated, 2007-8

IIS Sentinel Site

Per

cen

t (%

)

Preliminary Data

Percentage of children fully vaccinated (i.e., 1 or 2 doses as appropriate) against influenza among children 6-23 months of age, IIS

Sentinel Site Project, 2004-05 through 2007-08 influenza seasons

*Note: OR sentinel site expanded from Washington County in 2004-5 through 2006-7 seasons to include

Multnomah county in 2007-8 season. Michigan added one county to its sentinel site region in 2007-8 season.

0

20

40

60

80

100

OR MI AZ

Fully vaccinated, 2006-7

Fully vaccinated, 2007-8

IIS Sentinel Site

Per

cen

t (%

)

Preliminary Data

Percentage of children fully vaccinated (i.e., 1 or 2 doses as appropriate) against influenza among children 24-59 months of age, IIS

Sentinel Site Project, 2006-07 & 2007-08 influenza seasons

*Note: OR sentinel site expanded from Washington County in 2004-5 through 2006-7 seasons to include

Multnomah county in 2007-8 season. Michigan added one county to its sentinel site region in 2007-8 season.

*Thompson, et al. JAMA 2004

Influenza-Associated Hospitalizations By Age Group, 1979-2001*

115

22

90

472

0

100

200

300

400

500

Age Group

Ho

spit

aliz

atio

ns

per

100

,000

P

erso

n Y

ears

0-4 Yrs 5-49 Yrs 50-64 Yrs >65 Yrs

*Thompson, et al. JAMA 2004

Influenza-Associated Hospitalizations By Age Group, 1979-2001*

115

22

90

472

0

100

200

300

400

500

Age Group

Ho

spit

aliz

atio

ns

per

100

,000

P

erso

n Y

ears

0-4 Yrs 5-49 Yrs 50-64 Yrs >65 Yrs

Maintain focus!!

Monitoring Direct and Indirect Effects of Vaccinating School Age Children

• Historical comparisons, whether ecologic or individual-based, are problematic

• Season-to-season variability in influenza activity is pronounced– Patterns of circulation of specific viruses– Timing, duration, and intensity of activity

• Variable vaccine match

• Variable vaccine effectiveness, particularly during seasons of antigenic drift

Slide adapted from presentation by David Shay, Team Lead,

Prevention and Applied Modeling Team, Influenza Division, CDC,

September 2007 CDC/CSTE Consultation

Emerging Infections Program Surveillance for Laboratory-Confirmed Influenza: Cumulative Hospitalization Rates for Children Aged 0-4 and 5-17 yrs, 2007-2008 and Previous 4 Seasons (Total surveillance area:

4.7 million children aged <18, or ~7% of US population)

0

1

2

3

4

5

6

7

8

9

10

40 41 42 43 44 45 46 47 48 49 50 51 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

Week

Po

pu

lati

on

-Bas

ed R

ate

per

10,

000

Ch

ildre

n

2003-04, 0-4 years 2003-04, 5-17 years2004-05, 0-4 years 2004-05, 5-17 years2005-06, 0-4 years 2005-06, 5-17 years2006-07, 0-4 years 2006-07, 5-17 years2007-08, 0-4 years 2007-08, 5-17 years

“SchoolMist II”: 2004-2005King et al. N Engl J Med 2006; 355:2523-32

• Eleven clusters of 1 target and 1-2 control schools identified

• Healthy target school children offered LAIV in school (46% were vaccinated)

• Local surveillance identified influenza activity

• Anonymous questionnaires sent to all families at estimated peak influenza activity to ask about possible influenza illness (no lab confirmation) and effects (absenteeism, lost work days) over previous 7 days

“SchoolMist II”: 2004-2005 Impact on Target School Families

• Compared to control school families, target school families had statistically significant (P < 0.001) relative reductions during the week of peak influenza activity of:

– CDC- ILI in children (35%)

– Child physician visits (36%)

– Prescription medications (42%)

– OTC medications (56%)

– Herbal/natural medicines (36%)

King et al. N Engl J Med 2006; 355:2523-32

“SchoolMist II”: 2004-2005 Effects on families of target school pupils

• Compared to control school families, target school family surveys showed significant reductions during week of peak influenza activity of:

– CDC-ILI in adults by 36% (P < 0.01)

– Adult work days lost by 36% (P < 0.05)

– Adult physician visits by 26% (P = 0.06)

– High school days lost by 40% (P< 0.01)

King et al. N Engl J Med 2006; 355:2523-32

Could the impact of vaccination measured in the SchoolMist II study be reproduced

in real world settings?• With free vaccine and other immunization program

and study support: 46% coverage in target schools – What would it be in the real world?– With lower coverage would any substantial benefit be seen?

• Despite large study size (28 schools; 15,600 students; 8500 households) and focus on impact during the single peak week of an influenza season– No difference in overall absenteeism among students

(school data)– No difference in ED visits for children or adult contacts

King et al. N Engl J Med 2006; 355:2523-32

Measuring Impact will be an Unprecedented Challenge

• Given the variability in influenza epidemiology and vaccine effectiveness…– Illness measures might increase in some seasons even as

coverage increases– Illness measures might decrease in areas with poor

coverage

• Impact assessments will need to consider influenza epidemiology, circulating strains and vaccine effectiveness at local level

• Lesson – A long-term, large-scale effort is needed to determine impact on influenza illness rates

Thank you