Mandatory Influenza Vaccination for Health Care Workers as the New Standard of Care: A Matter of Patient Safety and Nonmaleficent Practice

Abstract

A growing body of literature defends the efficacy of seasonal influenza vaccination for health care workers in reducing the mortality of hospitalized patients. I review the evidence concerning influenza vaccination, concluding that universal vaccination of health care workers against influenza should be considered standard patient care and that nonvaccination represents maleficent care. I further argue that the ethical responsibility to ensure universal vaccination of staff against seasonal influenza lies not only with individual health care providers but with each individual health care institution.

Seasonal influenza remains a major cause of morbidity and mortality in the industrialized world. The World Health Organization estimates that each year seasonal influenza infects 5% to 15% of the world’s population, produces 3 to 5 million cases of severe illness, and leads to 250 000 to 500 000 deaths.¹ The United States alone averages more than 23 000 influenza-associated deaths annually, with $10.4 billion in direct medical costs and an estimated total annual economic burden of $87.1 billion.^2,3 Moreover, hospital-acquired influenza infection has a particularly high mortality rate; in a recent review of 12 hospital influenza outbreaks the median mortality was 16%, ranging as high as 60% in at-risk patient groups such as organ transplant recipients and intensive care unit patients.^4,5

As of 1981 the Centers for Disease Control and Prevention and its Advisory Council for Immunization Practices have recommended regular influenza vaccination of all health care workers, yet vaccination rates among health care workers in the United States vary and are often poor.⁶ In 5 studies the mean vaccination rate of US health care workers ranged from 41% to 57%, with lower rates among minorities, nurses, and nurse’s aides.^7–11 This is problematic because vaccination limits the spread of infectious diseases to susceptible individuals (i.e., unvaccinated patients), and modeling of a hospital influenza-outbreak ward suggests that a significant fraction of nosocomial influenza is vaccine preventable.¹²

Mandatory vaccination policies are uncommon in US hospitals.¹³ However, a growing movement among clinicians, ethicists, and legislators supports mandatory seasonal influenza vaccination for health care workers: such mandates have been advocated by the American College of Physicians, the American Academy of Pediatrics, the National Patient Safety Foundation, the Infectious Disease Society of America, the National Foundation for Infectious Diseases, the Society for Healthcare Epidemiology of America, and the Association for Professionals in Infection Control, and 19 states have enacted some form of vaccination mandate for health care workers.^14,15 Given the mounting evidence for the efficacy of influenza vaccination in infection control, I argue that the provision of health care by nonvaccinated health care workers is not merely suboptimal health care, but it is also at variance with generally accepted principles of health care ethics. Therefore, institutions are obligated to enforce universal vaccination of their health care workers against seasonal influenza. By the term health care worker I mean professionals employed by a health care institution or providing care pursuant to privileges granted by the health care institution who have regular personal interaction with patients, including physicians, nurses, physical and respiratory therapists, and similar ancillary medical staff. By health care institution I mean organizations offering inpatient medical care, including hospitals, nursing facilities, and rehabilitation centers. I would not yet include organizations offering solely outpatient care because the clinical evidence on which my arguments stand derive solely from the inpatient setting. Of note, this article aims to define ethical obligations and does not address the myriad legal issues surrounding health care worker vaccination.

ETHICAL BASIS FOR OBLIGATING INFLUENZA VACCINATION

Central to medical ethics are the principles of beneficence and nonmaleficence, which oblige medical professionals to practice in a manner that both (1) promotes the patient’s well-being, and (2) does not harm the patient’s well-being. Becoming vaccinated against influenza is certainly beneficent, by virtue of reducing the risk of transmitting influenza to one’s patients. However, I contest that vaccination against influenza should be mandatory because practicing without vaccination is maleficent because it falls below the standard of medical care.

Two elements are worth considering when evaluating whether an intervention should constitute standard care. The first is whether the intervention provides a substantial benefit to patients. The second is whether the intervention is overly burdensome (financially or in inconvenience or injury), preventing enacting the intervention from being the obvious course of action. If the evidence suggests that influenza vaccination of health care workers is safe, unburdensome, and efficacious in reducing patient morbidity, then vaccination is a component of minimally competent care and should be compulsory for health care workers and institutions.

HEALTH CARE WORKER VACCINATION AND PATIENT HEALTH

Four large, prospective trials have demonstrated that influenza vaccination of health care workers reduces death and disease in patients. In the first, 12 geriatric long-term care facilities were randomized to receive or not receive a health care worker vaccination campaign; the vaccinating institutions saw a statistically significant reduction in patient mortality (odds ratio = 0.56; 95% confidence interval [CI] = 0.40, 0.80).¹⁶ In the second, 20 geriatric long-term care facilities were randomized to receive or not receive a health care worker influenza vaccination program, which achieved a vaccination rate of 50.9%, versus 4.9% in the control facilities; patient mortality decreased by 42% in the vaccinated facilities versus controls (95% CI = 0.40, 0.84; P = .014).¹⁷ In the third, 22 pairs of matched long-term care facilities received or did not receive a health care worker vaccination program, achieving a vaccination rate of 48.2% in the intervention facilities versus 5.9% in controls; this resulted in reduced death (–5.0 deaths per 100 residents; 95% CI = −7.0, −2.0; P = .002) and influenza-like illness (–9.0 episodes per 100 residents; 95% CI = −24.0, −3.0; P = .004) among patients of vaccinated workers during the first influenza season studied, though not in a subsequent season with unusually low influenza activity.¹⁸ In the fourth, 40 matched nursing homes offered or did not offer influenza vaccinations to health care workers, achieving a vaccination rate of 69.9% in the intervention facilities versus 31.8% in the controls; the intervention was associated with a 20% decrease in patient mortality (95% CI = 0.66, 0.96; P = .02) and a 31% reduction in influenza-like illness (95% CI = 0.52, 0.91; P = .007).¹⁹

A few points deserve consideration when extrapolating these results to general practice. First, these trials took place in long-term care facilities, an idealized environment for halting influenza transmission featuring a highly susceptible patient population with considerable durations of stay and limited personal contact with individuals from outside the care setting. Even though the traditional inpatient setting also features an influenza-vulnerable population, the length of stay is lower and patient interactions with health care workers and others are more numerous. However, the greater number of patient–staff interactions may put health care workers in the traditional setting at increased risk for contracting and transmitting influenza, enhancing the effectiveness of vaccination, and at least 1 retrospective study suggests a protective effect in the acute care setting.²⁰ Second, in 2 studies showing the control group vaccination rates were markedly low, and in the study with a 31.8% vaccination rate in the control group, the relative improvement in mortality was lower. Mathematical modeling of inpatient influenza transmission suggests that universal influenza vaccination of health care workers could prevent up to 60% of nosocomial influenza infections; nonetheless, a study demonstrating the utility of an increase in vaccination rate from the typical 40% to 50% range to 100% would be welcome.²¹

SAFETY AND BURDEN OF INFLUENZA VACCINATION

To declare the vaccination of health care workers against influenza ethically obligatory we must also show that it does not pose an undue burden to workers or their institutions. I identify at least 4 potential areas of burden associated with vaccination: safety, financial cost, pain and suffering, and infringement of autonomy. The degree to which influenza vaccinations impose each of these burdens is discussed further.

Along with fear of needles and doubts of efficacy, misapprehension about the safety of influenza vaccination is the most common reason for refusal among health care workers, and knowledge about the influenza vaccine positively correlates with vaccination.^22–24 The Vaccine Adverse Event Reporting System, which tracks side effects potentially associated with vaccinations, has collected overwhelming evidence for the safety of seasonal influenza vaccines. True medical contraindication to both forms of the vaccine is rare, and many conditions perceived to be contraindications to vaccination are amenable to it, including egg allergy, HIV seropositivity, and a history of Guillain–Barré Syndrome (shown to be no more frequent among influenza vaccine recipients than nonrecipients and more likely to occur following influenza infection vs vaccination).^25,26 Influenza vaccination is safe in both children and pregnant women and is associated with improved birth outcomes during the influenza season.^27,28 The most commonly reported side effects for the inactivated vaccine are injection site pain and inflammation, red eyes, hoarseness, and cough, and for the live attenuated vaccine are runny nose, congestion, and sore throat; symptoms typically resolve within 48 hours. Other common reactions, including fever, myalgia, and malaise, are also self-limiting and occur in similar frequencies to placebo.²⁶

Serious adverse events associated with influenza vaccination do occur and can be life-threatening. However, in a 2009 review of Vaccine Adverse Event Reporting System data on influenza vaccination, the rate of serious adverse events was approximately 1 in 300 000 vaccinations, which compares favorably to the number of annual deaths directly attributable to influenza in the United States (1 death per 200 000 individuals).^29,30 This imbalance becomes more pronounced when considering the protective effect of influenza vaccination against death because of influenza and pneumonia (responsible for approximately 1 death per 5813 persons annually and the eighth leading cause of death in the United States), and more so considering that studies have shown a significant protective effect of influenza vaccination against cardiovascular death (responsible for 1 death per 522 individuals annually and the leading cause of death in the United States).^30,31 Thus, although becoming vaccinated does involve the risk of vaccine-related injury, it is difficult to argue that it increases one’s overall risk of disease or death. Because the common side effects of influenza vaccination are mild and self-limiting, and because the serious and far-reaching side effects of influenza vaccination are rare compared with the vaccination’s benefits, influenza vaccination must be considered safe.

The second consideration in influenza vaccination is cost. Even though the cost of a single dose of influenza vaccine is quite small, this might still be considered an undue burden on the individual health care worker. However, not only is it reasonable to expect health care institutions to bear the financial burden of mandatory vaccinations for their workers, evidence suggests that such an investment can reduce days lost to medical leave during the influenza season and is probably cost-neutral or cost-effective.^32–35 Universal influenza vaccination of health care workers should not represent a significant financial burden to health care institutions that stand to reap vaccination’s economic benefits; thus, if the costs of influenza vaccination are appropriately borne by the institution, said costs should be minimally burdensome.

The third consideration is the infliction of inconvenience, pain, and suffering. The inconvenience attributable to influenza vaccination is minimal; vaccines can be administered in less than a minute in the workplace. Pain and suffering associated with influenza vaccination are also minimal, particularly if the intradermal and intranasal vaccine preparations are used, and they are certainly no more than we demand of children, as evidenced by nationwide requirements for pneumococcal and meningococcal vaccination prior to attending school. Because the consequences of influenza infection for patients, potentially including pneumonia, hospitalization, and death, are serious and far-reaching, to dissuade us from an intervention protecting against influenza we would expect the pain and suffering the intervention inflicted to also be serious and far-reaching. Because they are not, the discomfort and inconvenience influenza vaccination poses is not unduly burdensome.

The final consideration is the concern that mandatory vaccination of health care workers poses an undue infringement of workers’ personal autonomy. Certainly, mandatory vaccination is particularly onerous in that it requires health care workers to take on a personal risk of harm, however small and potentially offset by the protective effects of vaccination, on their patient’s behalf. However, there are precedents for requiring health care workers to take on small personal risks because there is a significant opportunity for patient benefit. We do not allow health care workers to refuse care for HIV-positive patients despite the trivial risk of HIV transmission via accidental needlestick injuries; similarly, health care workers employed in interventional radiology and radiation therapy suites are inherently required to expose themselves to small amounts of radiation to deliver those interventions to their patients. Mandatory vaccination against influenza is another such case: in contrast to the 1 in 300 000 vaccination-associated serious adverse event rate previously discussed, in the 4 clinical trials cited earlier, the reductions in mortality reported give numbers needed to treat for a health care worker vaccination program to prevent 1 patient death ranging from 125 to a mere 11.4 patients.^16–19 Furthermore, we should consider the medically ill patient, who has little choice about being exposed to the infectious diseases of his caregivers. Because ultimately the health care worker has more choice about being in the hospital than his or her patient, the onus to be accommodating weighs more heavily on the health care worker. Finally, an effective mandatory vaccination program should offer appropriate alternatives to workers with a serious objection to vaccination: for example, they may be transferred to a position with limited patient contact or be required to wear N95 respirator masks at work, the latter having demonstrated some efficacy in reducing influenza transmission.³⁶ For these reasons, even though mandatory vaccination may infringe on health care workers’ autonomy, it does not do so to an undue degree.

ETHICAL RESPONSIBILITY FOR ENSURING VACCINATION

I have argued earlier that influenza vaccination of health care workers meets criteria for being standard care—namely, that vaccination has clear evidence of benefit to patients and is not overly burdensome in safety, cost, pain, and suffering, or infringement of personal autonomy—and thus I argue that it is ethically obligatory as nonmaleficent care. Then, who is obliged to ensure that health care workers are vaccinated? Previous calls for mandatory vaccination have focused on the health care worker; I argue that both workers and their institutions are obligated to ensure universal influenza vaccination.

Arguments for the ethical obligation of the individual health care professional to ensure she or he is vaccinated remain the most straightforward, and stem from the principles of beneficence and nonmaleficence discussed previously. Failure to vaccinate, given the minimal risk, inconvenience, and suffering involved, is unjustifiable on the part of the health care worker, especially given that health care workers are responsible for potentially exposing their medically ill patients to communicable diseases. Hand washing between patient encounters, another simple measure shown to reduce morbidity and mortality in patients, is analogous; given the trivial costs of inconvenience involved, a health care worker who refuses to wash their hands is practicing unjustifiably and unethically even if the cumulative increase in risk to patients remains small. When we accept patients into our care, we take a special professional fiduciary responsibility for their well-being, and that responsibility obligates us to follow all reasonable, evidence-based, best practices to ensure our patients’ safety.³⁷

I further contend that the health care institution has a separate ethical obligation to its patients requiring it to enforce universal vaccination of its employees and other health care professionals with staff privileges. First, there is the matter of agency. In modern medicine, hospitals and clinics have become their own entities with their own reputations: health care organizations such as the Mayo Clinic and the M. D. Anderson Cancer Center have become better known than any of their practitioners, and they advertise their services as a singular care-providing entity, making promises of excellence and quality in their standards of practice. When patients visit the Mayo Clinic for care they go with the belief that they will be receiving “Mayo care,” not simply the care standard of the admitting physician, and in that sense the modern hospital system has a responsibility to ensure it provides the quality care it claims to offer. Second, because health care institutions themselves place their patients at increased risk for harm by aggregating many ill people who potentially carry infectious diseases, are particularly susceptible to infection or death attributed to another disease state, or both, the health care institution creates for itself a responsibility to minimize the risk it has imposed on its patients. Third, health care institutions are obligated to ensure universal vaccination because infection control poses a special ethical dilemma in that it is a cooperative effort, meaning no single health care worker can effectively implement an intervention whose benefit requires the participation of all. Again, the analogy of health care worker hand washing is useful: even though a single nurse may somewhat improve his or her patient’s outcomes with consistent hand washing and has the ethical obligation to do so, the benefit will be marginal if the other nurses and physicians wash their hands inconsistently. In such cases, the ethical responsibility to adhere to the standard of care devolves to the party with the ability to enforce cooperative behavior. In the same way that the nonvaccinated health care worker is practicing without due care and unethically, the institution that fails to ensure universal vaccination of its staff has also neglected its obligations.

CONCLUSIONS

Barriers to universal health care worker influenza vaccination remain, including opposition from governmental agencies such as the Occupational Safety and Health Administration, which has argued for opt-out vaccination campaigns rather than mandatory influenza vaccination programs, health care worker noncompliance, and the threat of legal action against mandatory vaccination programs from employees or other organizations.^38,39 Although health care institutions currently have limited recourse to deal with the first concern, the latter 2 may be addressed by pursuing universal vaccination of their health care workers with an approach seeking the least restrictive means to achieve high rates of vaccination. That is, considering that the goal of mandatory influenza vaccination programs is universal vaccination of health care workers rather than the exercise of draconian action, the institution should begin by seeking health care workers’ buy-in with educational campaigns that address common misconceptions about influenza vaccine efficacy and safety and promote becoming vaccinated as part of ethical, beneficent, and professionally competent care. Where noncompliance is still an issue and particular where the health care worker has a serious religious or philosophical objection, alternatives that will protect patients from influenza, including transfer of the employee to a position where patient contact is more limited or mandatory use of a protective respiratory mask, should be considered. If all reasonable accommodations that will protect patients are rejected by the health care worker, then termination of the employee is not only ethically permissible but the responsibility of the health care institution, because its primary ethical obligation is to protect the safety and well-being of its patients. If a health care institution finds it is unable to discharge this duty for legal or other reasons, it should at minimum make patients aware of when they are interacting with unvaccinated health care workers and the risk this poses, as well as offer them the opportunity to refuse care by said individuals and request a vaccinated caregiver. Even though the potential consequences for health care institutions that do not meet these responsibilities lie outside the scope of this work, institutions should expect such repercussions as mandatory influenza vaccination of health care workers becomes increasingly accepted as a pillar of patient safety and infection control.

Fortunately, evidence suggests institutional policies ensuring universal vaccination need not be controversial or disruptive to the institution. In 2005 the Virginia Mason Medical Center began a 5 year study requiring their health care workers to receive the seasonal influenza vaccine unless granted a medical or religious excusal, with those excused required to wear a mask for patient-care purposes. The institution achieved vaccination rates greater than 98% during the study, with less than 0.7% of staff requiring medical or religious accommodation and less than 0.2% of health care workers choosing to leave the institution rather than become vaccinated.⁴⁰ Similarly, in 2009, when the MedStar Health program in Maryland enacted mandatory influenza vaccination for its approximately 25 000 health care workers, compliance was 99.9%, with 98.5% becoming vaccinated, 1.4% receiving medical or religious exemption, and only 28 terminations, representing less than 0.01% of staff.⁴¹ It appears that for almost all health care workers, opposition and concerns about influenza are not so deeply ingrained or intractable that they will choose termination over becoming vaccinated or an appropriate alternative. Thus, with a considered approach to mandatory vaccination that emphasizes education about influenza vaccination efficacy and safety and provides appropriate alternative accommodations for health care workers with persistent objections, evidence suggests universal vaccination can be accomplished with minimal distress.

Seasonal influenza carries a high burden of morbidity and mortality in the United States, particularly as a nosocomial infection. The vaccination of health care workers against seasonal influenza decreases morbidity and mortality in patients, at the same time remaining safe, inexpensive, and minimally inconvenient. Vaccination of health care workers against seasonal influenza should be considered standard medical care, and nonvaccination of health care workers unprofessional practice. The ethical obligation to become vaccinated rests on each individual health care worker as an extension of their professional duty to their patients; however, health care institutions also bear an ethical obligation to ensure and enforce universal influenza vaccination among their staff, given their ability to enact vaccination policies and their increasing individual agency and institutional fiduciary responsibilities in the modern medical era.