Head To Head

Should influenza vaccination be mandatory for healthcare workers?

BMJ 2013; 347 doi: http://dx.doi.org/10.1136/bmj.f6705 (Published 12 November 2013) Cite this as: BMJ 2013;347:f6705
  1. Amy Behrman, medical director, occupational medicine1,
  2. Will Offley, casualty nurse 2
  1. 1University of Pennsylvania Health System, 3400 Spruce St Philadelphia, PA 19104, USA
  2. 2Emergency Department, Vancouver General Hospital, Vancouver, Canada
  1. Correspondence to: A Behrman behrman{at}mail.med.upenn.edu, W Offley willo{at}lynx.bc.ca

Amy Behrman believes that mandatory vaccination is needed to protect vulnerable patients, but Will Offley argues that evidence on effectiveness is not sufficient to over-ride healthcare workers’ right to choose

Yes—Amy Behrman

Advocacy for influenza vaccination begins with recognising the impact of the disease. Globally, seasonal influenza causes an estimated 300 000-500 000 deaths and 3-5 million cases of severe disease every year.1 Methods that distinguish between influenza and other viruses causing influenza-like illnesses estimate that influenza infections and complications cause an average 226 000 hospital admissions annually in the United States, including 3000-49 000 deaths, depending on seasonal severity.2 Influenza vaccines are estimated to prevent thousands of admissions and millions of illnesses annually with current usage.3 4 5

Complications and deaths from influenza are highest in elderly people, infants, and patients with compromised cardiopulmonary or immune systems.1 2 6 These vulnerable populations are most likely to enter healthcare settings and least likely to mount effective immune responses to vaccination.2 6 Influenza vaccines have excellent safety records6 and are most effective (59% reduction in laboratory proved influenza6 and 47-73% reduction in influenza-like illness2) in healthy non-elderly adults, precisely the demographic of most healthcare workers.

Nosocomial transmission is well documented.7 Influenza infection control should include, in addition to vaccination, hand hygiene, isolation of infected patients, targeted masking, and leave of absence for healthcare workers with influenza-like illness.8 Vaccination is a keystone intervention, differing from others by reducing risk in all encounters without repeated effort or time from busy healthcare workers.

Annual vaccination is therefore widely recommended to reduce the risk of healthcare acquired influenza.2 8 9 10 11 12 13 Advocates and opponents of mandatory vaccination share goals of enhancing patient and staff safety. Disagreements centre on evidence of efficacy, ethical concerns, and how best to achieve meaningful levels of immunisation. My institution’s evolution to a mandatory policy epitomises the issues.

Mandatory vaccination in practice

The University of Pennsylvania Health System has 18 000 staff. Immunisations are free for all vaccine preventable occupational infections. Influenza vaccine has been offered annually since 2003 but was initially voluntary. Uptake by healthcare workers averaged <40%, and many staff avoided immunisation even during years with clear evidence of hospital transmission. Despite prioritisation of influenza vaccination through enhanced availability on all units and shifts and at “flu fairs” with educational materials, over two years, vaccination rose to an unimpressive 45%.

We implemented declination forms in 2006-07 to survey the concerns of unvaccinated staff. As in other institutions, many declined because they underestimated influenza morbidity, feared vaccination would cause illness, or believed “clean living” would prevent transmission. Vaccination rates crept to 50%.

In 2008, we further increased outreach, including a music video addressing the concerns highlighted in the survey (www.youtube.com/watch?v=ruGgZbAVnko). Talented staff participated enthusiastically; the video played continuously; compliance inched to 54%.

By 2009, we perceived limits to non-mandatory immunisation despite maximal efforts. In an anonymous survey, 85% of medical staff supported mandatory immunisation, with 90% agreeing that healthcare workers have an ethical obligation to be vaccinated annually.14 Evidence for patient benefit from immunisation of staff working in long term care, although imperfect, supported our aim to improve compliance among our healthcare workers.15 16 17 18 19 Other mandatory programmes had achieved strikingly increased compliance without safety problems.20 21

In 2009, our health system approved a mandatory policy for all staff. Implementation was complicated by the H1N1 epidemic, with dual vaccine shortages necessitating a tiered approach to prioritise medically compromised staff and those caring for the most vulnerable patients. Despite this, 99% of staff were vaccinated for seasonal influenza.

Subsequently, medical and religious exemptions have remained stable at <2%, while the mandatory policy is largely accepted as integral to patient and staff safety. Exempted staff are transferred from vulnerable patient units while influenza circulates. Non-compliant staff face escalating penalties (beginning with written warnings), which have been successful without having to terminate employment. Nosocomial influenza has decreased from significant to negligible since 2010, albeit with complementary policies for hand hygiene, isolation of those with influenza-like illnesses, testing healthcare workers with influenza symptoms, and leave of absence for staff with transmissible illnesses. A preliminary safety review of 40 560 staff immunisations over three years found 29 associated clinical complaints, of which eight (0.02%) were systemic symptoms possibly related to vaccination. The remaining 21 (0.05%) were arm pain likely related to injection. All symptoms resolved fully.

Mandatory vaccination is effective and ethical

Recent studies robustly support the effectiveness of mandatory policies in improving vaccination rates.7 20 21 22 Proving that vaccination of healthcare workers decreases the risk of transmission is more difficult for reasons including suboptimal immunisation rates, variable viral severity and vaccine effectiveness, visitor exposures, evolving laboratory diagnostics, and the confounding effects of other infection control interventions. Nevertheless, existing research from long term care facilities supports increasing healthcare workers vaccination to improve patient outcomes.8 15 16 17 18 19 Although acute and ambulatory care patients are likely to be discharged before nosocomial influenza can be recognised, the principles of transmission and immunity are the same,8 23 24 and these patients also deserve vaccinated healthcare workers. More definitive studies and better vaccines are wanted,1 6 7 but existing vaccines are safe and effective for healthy adults.2 4 6 Mandatory policies make them more effective.

Finally, healthcare workers have an ethical imperative to prevent harm to patients.25 Healthcare workers can infect patients,7 8 and influenza vaccination reduces adult infections,1 3 6 24 therefore vaccination of healthcare workers should reduce risk while setting an example for patients and communities to get recommended immunisations. Maximising compliance should optimise outcomes.8 19 23 Ideally, healthcare workers will take individual responsibility for being fully immunised. When this does not occur, healthcare institutions have an ethical obligation to intervene, just as they do to optimise handwashing and minimise surgical site errors. Mandatory vaccination policies accomplish this.

Healthcare institutions should maximise the use and benefit of a vaccine that is moderately effective, extremely safe, and logically likely to reduce the risk of healthcare acquired influenza for vulnerable patients as well as decrease illness among healthcare workers. First do no harm.

No—Will Offley

The debate around compulsory influenza vaccination for healthcare workers revolves around one central question: does current scientific evidence justify over-ruling the right to informed consent to an invasive and imperfect medical procedure, with documented risks of adverse effects.

Many in this debate answer in the affirmative. For them, patient safety outweighs the right of healthcare workers to refuse influenza vaccines. They argue that compulsory vaccination is consistent with the ethic to “do no harm” and protects vulnerable people from contracting influenza from their caregivers. The only problem is that there is no persuasive scientific evidence to support this view.

Benefits are unproved

Vaccinating healthcare workers against influenza has not been shown to reduce the transmission of influenza to patients. A recent Cochrane review of five studies (four cluster randomised studies and one cohort trial of nearly 20 000 healthcare workers) concluded that “there is no evidence that only vaccinating healthcare workers prevents laboratory-proven influenza or its complications (lower respiratory tract infection, hospitalization or death due to lower respiratory tract infection) in individuals in [long term care] and thus no evidence to mandate compulsory vaccination of healthcare workers.”26

Several of Canada’s leading influenza researchers have also noted the lack of evidence that vaccination of healthcare workers reduces the incidence of influenza in patients.27 Even groups that support mandatory immunisation such as the Centers for Disease Control and Prevention (CDC)28 and Association of Medical Microbiology and Infectious Disease Canada29 have had to acknowledge the lack of data to support this assertion.

Mistaken beliefs

The argument for mandatory vaccination rests on several major fallacies that combine to inflate the perceived effect and virulence of influenza and exaggerate the effectiveness of influenza vaccines.

Firstly, the burden of disease represented by influenza is often expressed by referring to surveillance statistics. In the United States, the CDC attributes 3000 to 49 000 deaths annually to influenza.30 However, its final data for 2010 show that of the 50 097 deaths recorded for influenza and pneumonia combined, only 500 were from influenza.31 In addition, the threat from seasonal influenza is dropping, not increasing. Mortality in the US has dropped from 30-40/100 000 cases in the mid-1930s to less than 5/100 000 from the 1950s onwards, and before the advent of mass public immunisation campaigns.32

Another problem is that influenza vaccines only protect against influenza. However, 85% of influenza-like illness is not caused by influenza but by any of about 200 viral and bacterial pathogens, none of which is prevented by seasonal influenza vaccines.32

Furthermore, the ability of influenza vaccines to prevent true influenza varies considerably from year to year. It has varied from a reported high of 93%33 to a low of 4.6% in a nine year study from California.34 In 2012, vaccine effectiveness against the dominant A(H3N2) influenza was 47% in the US35 and 45% in Canada.36

Recent European studies conducted during the 2011-12 season and involving more than 9000 participants have reported that the effectiveness of influenza vaccine dropped by more than 50% within four months of being vaccinated.37 Thus vaccination imparts only partial immunity to begin with, and even that does not seem to last for the full length of an influenza season.

It is therefore unsurprising that the American College of Occupational and Environmental Medicine has adopted a position against compulsory influenza vaccination, stating that current evidence regarding its ability to protect patients “is inadequate to override the worker’s autonomy to refuse vaccination.”13 The US Occupational Safety and Health Administration, which is responsible for workplace health and safety, has stated that “there is insufficient evidence for the federal government to promote mandatory influenza vaccination programmes that may result in employment termination.”38

Risks of vaccination

Influenza vaccines are relatively safe, not absolutely safe. Adverse effects, while uncommon, are nonetheless real, particularly (but not exclusively) for children and adolescents. Studies indicate that individuals receiving trivalent inactivated vaccine have a one in a million chance of contracting Guillain-Barré syndrome39 and a 13% higher incidence of oculorespiratory syndrome.40 Happily, most of these reactions have been limited to certain manufacturers and formulations, although not all were discovered before the vaccine was administered.41 A Canadian study also found a 1.4 to 2.5 higher rate of pandemic H1N1 influenza among people who had received the 2008 seasonal influenza vaccine, which did not contain this strain.42

Ethical rights of staff

Compulsory vaccination against seasonal influenza is based on an exaggerated threat and an exaggerated cure. Despite a lack of reliable, disinterested scientific evidence to show that healthcare workers are an important source of transmission to our patients, mandatory vaccination is promoted as a panacea without due regard to risks. Compulsion strips healthcare providers of a basic right guaranteed to every other patient—the right to informed consent.

Healthcare workers can and must make a real contribution to protecting patients from influenza—by isolating patients with symptoms of respiratory infection, improving infection control, covering our coughs, washing our hands, and, above all, staying home when we are sick. But until there is more persuasive evidence, it is neither a breach of ethics nor a disservice to patients to insist that influenza vaccination remains a personal decision based on informed consent.


Cite this as: BMJ 2013;347:f6705


  • Competing interests: The authors have read and understood the BMJ Group Policy on declaration of interests and declare WO is a participant in Concerned Health Care Providers, a voluntary advocacy organisation of healthcare staff in British Columbia that formed in opposition to mandatory flu immunisation.

  • Provenance and peer review: Commissioned; externally peer reviewed.