SCarrots
and Sticks:
Influenza Vaccination of Healthcare
Workers
Susan E Coffin, MD, MPH
Children’s Hospital of Philadelphia
July, 2011
Overview
Rationale behind HCW
influenza vaccination
Implementing a mandatory
flu vaccination program at
CHOP
Impact of mandate
• HCW attitudes
• Nosocomial influenza rates
HCW Flu Vaccination: background
• Vaccination of health care workers (HCW) decreases…
▫ Healthcare-associated influenza infection
▫ HCW absenteeism
▫ Secondary infections among HCW’s household contacts
• Especially important in pediatric centers:
▫ Large reservoir of disease in pediatric hospitals
▫ Large proportion of hospitalized children at high risk of
severe influenza
• Growing interest in potential role of mandates
▫ Recommended by the CDC and endorsed by IDSA, SHEA,
AAP
▫ Mandates successfully implemented at several other U.S.
health systems
Nosocomial Influenza at CHOP (2000-2006)
Complications experienced by 56 patients
with nosocomial influenza*
Number (%)
Death
2 (3.6%)
Respiratory failure
3 (5.4%)
Suspected bacterial pneumonia
Bacteremia
*2000-2004; complications determined by detailed chart review
Coffin, ICHE, 2009.
12 (21.4%)
1 (1.8%)
Preventing nosocomial influenza:
why is HCW vaccination critical?
• Virus primarily transmitted by large respiratory droplets
▫ Less benefit from hand hygiene
• Virus can be shed 24 hrs before symptom onset
• Adults can have asymptomatic infections
▫ 20-50% of infected HCW were asymptomatic
• Many hospitalized pediatric patients too young to receive
vaccine or unable to mount protective immune response
Vaccination reduces the rate of
nosocomial influenza
• Observational study at
University of Virginia
hospital
• Over 13 seasons
• Increasing vaccination rate
among HCW associated with
reduced proportion of
nosocomial influenza (32%
in 1987-88 to 3% in 1998 99)
Salgado, ICHE, 2004
Direct Benefits of HCW Vaccination
Talbot, ICHE, 2005
Improving HCW Vaccination Rates:
Strategies that work
• Education
▫ Risks of disease1,2
▫ Vaccine safety and efficacy2
• Internal marketing1,3
• Improving access to vaccine
▫ Mobile carts1,2
▫ Walk-in clinics, after-hours clinics2
• Expanding responsibility
▫ Vaccine deputies1
▫ Charge nurses as educators2
1) Bryant, ICHE 2004; 2) Tapiainen ICHE 2005; 3) Spillman, 40th National Immunization Conference Atlanta, March 2006
Cognitive Dissonance 101
Flu is bad for me
and my patients.
I will get vaccinated.
Flu vaccine is unsafe.
?
?
?
?
?
I don’t get flu
vaccine.
Employer:
“Get
Vaccinated!”
I don’t get flu.
Flu vaccine doesn’t
work.
You Can’t Make Me!!!
Wake Forest Declination Form (2005)
“I realize I am eligible for the flu shot and that my
refusal of it may put patients, visitors, and
family with whom I have contact, at risk should
I contract the flu. Regardless . . .”
Adoption was associated with doubling of
immunization rates (35% to 70% over 4 yr
period)
Spillman SS presented at 40th National Immunization Conference Atlanta, March 2006
Are Declination Forms Enough?
PRO
CON
• HCW vaccination no longer a
“passive decision”
• Signals acceptance of nonvaccination
• Provides final opportunity to
frame issue
• Polarizing effect reported by
some
• Creates focus on individual
accountability
What level of HCW vaccination
is ideal?
• Likely related to proportion of vaccinated staff and
patients…
▫ Retrospective study of 301 nursing homes (2004-2005)
▫ Combined immunization rate of staff and residents
inversely associated with risk of outbreak
▫ 60% reduced risk of outbreak associated with staff
immunization rates of 55% and resident immunization
rates of 89% (OR 0.41; 95% CI 0.19, 0.89)
Shugarman, J Am Med Dir Assoc, 2006
Targeted group(s)
2004-2005
2005-2006
2006-2007
2007-2008
2008-2009
2009-2010
57%
69%
73%
90%
92%
99.6%
Direct care providers* in
high risk settings#
All direct care providers*
All^ who work in
building where
patient care is
delivered
Education and
Communication
Mandatory education module included in fall core curriculum
Linked to pandemic flu preparedness
Linked to patient safety
Remedial education$
Town hall meetings
Logistics
Declination Form
Expanded Occupational Health clinic hours
Unit- and practice-based flu captains
Flu vaccine clinics held at meetings
Roving vaccination carts
None
None
Voluntary
Mandatory
Administrative
Senior administration stresses importance of flu vaccination to clinical leaders
Biweekly compliance reports@
Weekly compliance
reports@
Use of LAIV ^
Offered to providers who
did not work in high risk
setting#
Offered to all providers except those who worked on oncology unit
Why CHOP HCW decline flu vaccine
2005-2006
2006-2007
Allergy/Reaction
39
26
Rec’d vaccine elsewhere
36
6
Concern about side effects
34
193
Never get flu
9
27
119
53
Religious
1
0
Other
32
15
Pregnancy
11
5
Fear of needles
7
0
276
392
Personal choice
TOTAL
percent vaccinated
50
40
30
20
10
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
53% MD groups >80% (19/36)
22% MD groups fully vaccinated (8/36)
0
81% of MDs vaccinated (623/777)
Urology
Interventional Radiology
Rheumatology
Infectious Diseases
Dermatology
Cardiac Anesthesiology
Endocrinology
Emergency Medicine*
Ophthalmology
Allergy and Immunology
Pulmonary Medicine
CDR and Metabolic Disease*
Nephrology*
Otolaryngology
Cardiology
General and Thoracic Surgery
General Pediatrics*
Adolescent Medicine*
Neurology*
General Anesthesiology
Oncology*
Orthopaedics
Hematology
2007-2008
Dentistry
Pathology and Clinical…
Neonatology*
Critical Care Medicine*
Plastic Surgery
Human Genetics
Gastroenterology and Nutrition*
80
Neurosurgery
90
80
Child and Adolescent Psychiatry
90
Neuroradiology
100
Nuclear Medicine
100
General Pediatric Imaging
70
Cardiothoracic Surgery
16% MD groups >80% (5/31)
% Physician Participation
Vaccination of physicians
2008-2009
60
70
60
50
40
30
20
10
2009-2010
CHOP Employee Influenza Vaccine
Program
July, 2009: “The CHOP Patient Safety
Committee recommends mandatory
annual influenza vaccine for all staff*
working in buildings where patient care
was provided or whom provide patient
care.”
*includes clinicians, support staff, volunteers, students; vendors
informed of policy and asked to ensure compliance.
Key Strategies, 2009-2010
PROGRAM ELEMENTS
• Create accurate list of targeted staff and assure
ability to provide timely, accurate reports
• Establish method for evaluating requests for
medical and religious exemptions
• Determine timeline and educate
Program Timeline, 2009-2010
PLAN:
• 6 week program (9/15-10/31/09)
• 2 week furlough for staff unvaccinated and without
exemption as of 11/1/09
• Termination if unvaccinated and without an exemption
as of 11/15/09
REALITY:
• 2 week extension due to delays in receipt of seasonal flu
vaccine
What happened: 2009-2010
• >9000 HCW vaccinated
• 50 persons established medical exemptions
• 2 persons established religious exemptions
• 145 received temporary suspension
• 9 persons terminated
Labor Relations 101
• 2 meetings to negotiate
▫ Impasse declared
Quotes from 10/26/09 negotiation:
• “You’re not making sure everyone who comes
into CHOP is vaccinated.”
• “Why can’t we just wear masks all winter?”
• “No other institutions or regulatory groups
support this.”
• “This discriminates against employees who have
less access to educational resources on the
internet.”
Labor Relations 102
• Grievance filled (November, 2010)
▫ CHOP: Termination for just cause
 “Behaviors that are detrimental to the institution
 “insubordination”
▫ Union: Breech of contract
 Not included in negotiated contract
Findings and Opinions from Arbitration:
• “There can be no doubt that the Hospital had
the right to promulgate a ‘reasonable’
rule/condition of employment that would better
ensure the health and safety of CHOP’s patient
population.”
• “It is this Arbitrator’s finding that the policy
implemented by the Hospital was reasonable in
the context of the Hospital’s young, vulnerable
patient community.”
Year 2 Experience: 2010-2011
• >9500 HCW vaccinated
• Request for medical exemptions by 7 HCW (all
granted)
• Request for religious exemptions by 3 HCW
▫ Review by retired judge
▫ 2 granted, 1 denied
• No suspensions or terminations.
Evaluating Impact of Vaccine Mandate:
METHODS:
▫ Cross-sectional study of a random sample of HCW
subjected to the mandate
25%
clinical
(n=1450)
8,093
HCW’s
50%
nonclinical
(n=1100)
▫ Anonymous 20 item questionnaire adapted from
validated previously published instrument
(electronic>>paper distribution)
Study Question:
What predicts agreement with the mandate?
• Primary outcome: attitude towards influenza
vaccine mandate
▫ “Do you agree with CHOP’s policy that requires all
health care workers to receive annual flu
vaccination (a flu shot or the nasal spray vaccine)
unless there is a medical or religious
contraindication”
Results: Survey
• Response rate (58%):
▫ 1,388 respondents (total distributed = 2,443)
 657 (47%) clinical
 731 (74%) nonclinical
• Respondent characteristics:
▫
▫
▫
▫
77% female
65% < 45 years of age
68% have worked at CHOP <10 years
90% staff previously vaccinated
• 91% felt they had received info they needed from
CHOP to make decision about flu vaccination
Results: Reasons for vaccination
• Of those who had been vaccinated in past, majority
of respondents cited:
▫ Protection of self, family and patients
▫ Job responsibility
▫ Education received at work
• Of those who declined flu vaccination in past,
majority of respondents cited:
▫ Not being at high risk
▫ Fear of side effects
▫ Belief that vaccine is not effective
Results: Agreement with mandate
• 77% respondents intended to be vaccinated before
hearing about the mandate
• 75% reported agreeing with mandate
• 23% of respondents strongly considered declining
the flu vaccine after hearing about the mandate
• 72% reported agreeing that the mandate is coercive
but almost everyone (96%) also agreed that
mandatory policies are important for protecting
patients
Results: Agreement with mandate
• ~75% of both clinicians and non-clinicians agree
that societal rights outweigh individual rights when
it comes to vaccination
• ~95% of both groups agree that parents have an
obligation to make sure their children receive
recommended vaccines
• >95% of both groups agree with policies for
requiring vaccination or screening for TB, HepB,
measles, rubella and varicella
Predictors of Agreement with Mandate
Demographic Predictors
Attitudinal Predictors
• Contact with high risk
individuals at home or at work
• Age
• Amount of time working at
CHOP
• Gender
• Previous receipt of flu vaccine
• Previous experience with flu
vaccine
• Reasons for previous flu vaccine
receipt
• Reasons for previous flu vaccine
declination
• Attitudes towards influenza
prevention
• Intention to receive the vaccine
before knowledge of the
mandate
• Attitudes towards other
mandatory vaccination
programs
• Attitudes towards vaccines in
general
Factors associated with Agreement with
Mandate: unadjusted results
Unadjusted OR
(95% C.I.)
Clinical (vs. Non-clinical)
1.49 (1.32, 1.68)
Previous vaccination
Yes (vs No)
6.3 (5.10, 7.79)
Intention to be vaccinated before
mandate, Yes (vs No)
10.6 (9.1, 12.5)
Belief in Mandate benefits
Support other employment mandates
Ethical beliefs regarding vaccines /
public health
29.0 (24.3, 34.6)
4.02 (3.36, 4.80)
6.87 (6.00, 7.86)
Factors associated with Agreement with
Mandate: multivariable model
Unadjusted OR
(95% C.I.)
Adjusted OR
(95% C.I.)
Clinical (vs. Non-clinical)
1.49 (1.32, 1.68)
1.08 (0.94, 1.26)
Previous vaccination
Yes (vs No)
6.3 (5.10, 7.79)
1.68 (1.29, 2.19)
Intention to be vaccinated before
mandate, Yes (vs No)
10.6 (9.1, 12.5)
2.64 (2.17, 3.21)
29.0 (24.3, 34.6)
14.08 (11.5, 17.2)
4.02 (3.36, 4.80)
1.40 (1.13, 1.73)
6.87 (6.00, 7.86)
3.15 (2.70, 3.70)
Belief in Mandate benefits
Support other employment mandates
Ethical beliefs regarding vaccines /
public health
Possible Implications
• Majority report that mandate is coercive
▫ Does not appear to affect agreement with mandate
• Factors associated with agreement with mandate represent
attitudes and beliefs that may be modifiable through targeted
outreach and educational activities
▫ May need to focus upon different key themes for clinical and nonclinical staff
• Reasons for previous declination of vaccination show that
misconceptions regarding risk for infection and vaccine safety
and efficacy do persist
▫ Educational modalities may not be effectively communicating key
messages
600
12
500
10
400
8
300
6
200
4
100
2
0
0
Community-onset
Healthcare-associated
% healthcare-associated
number
Do Mandates Improve Patient
Outcomes?
% healthcare-associated
Summary
• Nosocomial influenza poses a serious threat to
hospitalized children.
• HCW vaccination rates can be substantially improved
through implementation of various voluntary measures.
• Mandates may be required to achieve maximal levels of
HCW compliance but many HCW may support
mandates and believe that they are important way to
protect patients and staff
• Attitudes and beliefs associated with support of mandate
may transcend professional role
Questions?
Acknowledgements:
Occupational Health
- Mary Cooney
Infection Prevention and Control
- Keith St. John
- Eileen Sherman
Infectious Diseases Epidemiology Research Group
- Kristen Feemster
- Priya Prasad
All CHOP Healthcare Workers