The trusted source for
healthcare information and
By Gary Evans, Medical Writer
In a controversial finding that could bolster legal challenges to mandatory flu vaccination of healthcare workers, researchers report1 that the four randomized controlled trials (RCTs)2-5 commonly cited to justify compulsory immunization lack scientific rigor and overstate the protective benefit to patients.
The four trials were conducted in long-term care facilities, but have generally been extrapolated to support flu immunization requirements in acute care and other settings. Among other benefits, the studies generally link reductions in resident mortality to higher staff flu vaccination rates. With healthcare worker vaccination rates lagging in most hospitals, several facilities began mandating immunization several years ago in what has since become a national trend.
As the RCTs are often cited in support of such policies, this new Canadian study attacks a core rationale for making flu shots a condition of healthcare employment. After an elaborate deconstruction and analysis using the “dilution” principal of mathematics, the authors of the study conclude that there is insufficient scientific evidence to mandate flu vaccine as a condition of healthcare employment.
“It is unsupported by science — that is clearly the answer,” says the lead author of the study, Gaston De Serres, MD, of the Institut national de santé publique du Québec in Quebec City, Canada. “When you are mandating a public health intervention, you better be sure that the burden of disease that you want to prevent is substantial. At this time, no one knows how many patients are infected annually in the United States by unvaccinated healthcare workers. Nobody knows.”
De Serres emphasized that voluntary influenza vaccination should still be encouraged and the findings should not in any way be construed as supportive of the broader anti-vaccine movement.
“I personally receive the [flu] vaccine every year,” he says. “I’ve published hundreds of papers on vaccines and I am a strong advocate of vaccines. But it’s one thing to recommend to healthcare workers that the risk of you transmitting influenza to your patients may be very small but we recommend that you take the vaccine. That’s one thing, but it is quite another to say if you don’t get the vaccine, you’re fired. It’s completely different.”
The findings are compelling enough that professional groups like the Society for Healthcare Epidemiology of America (SHEA) should rethink their endorsement of mandatory flu vaccination, says Michael Edmond, MD, clinical professor of internal medicine and infectious diseases at the University of Iowa in Iowa City. (See related story, in this issue.)
“I don’t want this to be confused with any kind of anti-vaccine rhetoric, but to me it kind of plays into it,” says Edmond, who was not involved in the study. “The anti-vaccine rhetoric focuses on false claims of adverse effects of vaccines, and that is completely wrong. But on the other side of it, we have to acknowledge the shortcomings of vaccines in an honest way as well. I would argue that we have to look at the science on both sides of it and do what the science tells us to do.”
With any intervention to improve quality or patient safety, the expectation for compliance should be directly related to the evidence, he says.
“In this case, we don’t have a high level of evidence,” he says. “Therefore, to mandate it — and in this case, to say if you are not in compliance you will be fired from your job — to me is just an unacceptable situation.”
De Serres and colleagues explain the mathematical principle of dilution by giving a real-world example of buying chicken with a 60% off coupon. As additional items are added to the shopping cart, the chicken coupon provides a lower percentage reduction of the total bill. It saves money on the chicken, but the overall savings for the shopper are “diluted” by the inclusion of non-discount items. The same principle of dilution applies in attributing percentage reductions in non-specific outcomes to influenza vaccination like all-cause mortality and influenza-like illness (ILI) caused by other viruses, the authors note.
“We limited our analysis to three outcomes — ILI, laboratory-confirmed flu, and all-cause mortality,” De Serres tells Hospital Infection Control & Prevention. “And using the principle of dilution, we looked at what could be reasonable or plausible results that could come from vaccination. We found that the four studies were reporting results for [reductions in] all-cause mortality that were not 50%, not even 100% greater. They were 10 to 12 times greater than could plausibly be achieved if the vaccine was 60% effective.”
Historically speaking, that is a spot-on efficacy estimate, as a comprehensive review of the literature estimated an overall flu shot efficacy of 59% in healthy adults ages 18 to 64 years. Evidence was inconsistent or lacking to determine vaccine efficacy in those ages 2-17 and those over 65.6 Of course, the vagaries of antigenic “drift” or more dramatic “shift” of circulating, mutable influenza viruses create an ongoing challenge to making an annual vaccine with a good match. For example, last year’s overall flu vaccine efficacy was only 41% against the predominate H1N1 influenza A strain and 55% against circulating B strains.7
“It is interesting that most of the organizations and associations that publish support for mandating immunization [in healthcare workers], use data underscoring that influenza is a terrible disease with tens of thousands of deaths annually and hundreds of thousands of hospitalizations,” De Serres says. “Well, that’s irrelevant and will not be affected by the immunization of healthcare workers. The relevant burden of disease [in making an argument for mandated immunization] are only the cases represented by unvaccinated healthcare workers.”
In contrast to the prevented-mortality estimates reported in the papers, De Serres and colleagues project that it actually may take from 6,000 to 32,000 healthcare worker immunizations to prevent one patient death from influenza.
“Influenza vaccine is only effective against influenza virus and provides no benefit against other causes of illness or death unrelated to influenza virus,” the authors report. “As more non-influenza causes contribute to the outcome considered, the lower must be the percentage reduction attributed to vaccine effects. Where that is not observed, there is a de facto error in claiming vaccine benefits.”
The paper has the devastating cumulative effect of a legal argument, and indeed concludes with a disclosure endnote that several of the authors have provided expert testimony during legal challenges to mandatory flu vaccination policies in Canada. In a recent case that surprised some, six healthcare workers fired for refusing mandatory flu shots for religious reasons recently won back pay and offers of reinstatement from Saint Vincent Hospital in Erie, PA, after their case was taken up by the U.S. Equal Employment Opportunity Commission. (For more information, see the February 2017 issue of HIC).
“Each of the four cluster RCTs used to champion compulsory HCW influenza vaccination policies reports benefits that are mathematically impossible under any reasonable hypothesis of indirect vaccine effect,” the Canadian authors conclude. “…Through this detailed critique and quantification of the evidence, we conclude that policies of enforced influenza vaccination of HCWs to reduce patient risk lack a sound empirical basis. In that context, an intuitive sense that there may be some evidence in support of some patient benefit is insufficient scientific basis to ethically override individual HCW rights. While HCWs have an ethical and professional duty not to place their patients at increased risk, so also have advocates for compulsory vaccination a duty to ensure that the evidence they cite is valid and reliable, particularly in the absence of good scientific estimates of patient impact.”
Though the study could spur legal challenges or raise questions for individual hospitals about their policies, in some sense the train has already left the station on the issue of mandated flu shots for healthcare workers. Mandatory vaccination policies are driving a trend that saw 91% of hospital workers immunized during the last flu season — a far cry from the lagging levels of the voluntary flu shot era. Still, vaccination rates for healthcare workers in long-term care settings fall well below hospitals, with 31% skipping the seasonal shots despite caring for frail and elderly people at risk of serious flu infections, the CDC reports.8
One of the groups supporting mandated flu shots for healthcare personnel is the Association for Professionals in Infection Control and Epidemiology (APIC). APIC president Linda R. Greene, RN, MPS, CIC, FAPIC, makes both ethical and epidemiological arguments in favor of mandatory flu shots.
“It is important in healthcare to take personal action for the public good,” says Greene, manager of infection prevention at UR Highland Hospital in Rochester, New York. “Those are things I don’t think we can dismiss. We need to do the right thing for our patients. APIC has a position paper on this, and certainly from our perspective this is not going to change that. We really feel that mandatory flu immunization should be considered a condition of employment. The data show us clearly that the highest vaccination rates are when there is a mandatory policy in place in an organization.”
This study raises questions, but Greene adds a compelling one of her own: Would you want the healthcare workers caring for your hospitalized family member to be vaccinated against influenza?
“The [researchers] just looked at the RCTs,” she says. “There are a number of other good studies that don’t meet the level of clinical trials, but there are some good case control studies that definitely talk about transmission of influenza from staff to patients. The other thing in the article they talked about is the dilution effect and ILI. However, flu sometimes happens without ILI. We know that many people — particularly the young and healthy — may actually have influenza without even exhibiting signs or symptoms. There are a lot of people who are not tested, so there are a lot of other variables other than this dilution methodology."
Although current data are inadequate to support enforced healthcare influenza vaccination, they do not refute voluntary vaccination or other more broadly protective practices such as staying home or masking when acutely ill, the study authors conclude.
“There is a real problem with the healthcare system and it is going to work when you are sick — not only for influenza, but the other respiratory viruses,” De Serres says. “You don’t want your professionals — your nurses caring for patients — to be at the bedside when they are sick. And that really is not often addressed. It appears to be a sort of magical solution to use the influenza vaccine and coerce vaccination to demonstrate that you really care for patients.”
Presenteeism is a widely reported problem that challenges short-staffed facilities, in part fueled by the prevailing mentality that healthcare workers don’t want to let their colleagues down by failing to report for duty.
“I found it interesting at the end of the discussion [the authors] say we ought to think about other things in addition to voluntary immunization, like staying home when sick or masking, “says William Schaffner, MD, a professor of preventive medicine at Vanderbilt University in Nashville. “I find it stunning that they, in effect, recommend those interventions, over which the data are much less [established] than for influenza immunization.”
Having long acknowledged the flu vaccine is imperfect and subject to seasonal fluctuations in efficacy, Schaffner nevertheless favors mandatory immunization of healthcare workers despite the Canadian study.
“If you take the RCTs all by themselves — off to the side — our Canadian colleagues have a point, but I don’t believe their point carries the day,” he says. “They only looked at the RCTs. These provide the most rigorous data, but we recognize that RCTs on this question are hugely difficult to do. If we are going to address the entire question of the scientific, professional, and ethical basis of mandates, you ought to look at the entire body of evidence. There are any number of observational studies of one kind or another that could be added. Obviously they have profound scientific limitations also, but at least in my mind they add to the body of evidence that makes mandatory influenza immunization more than reasonable.”
A past president of the Foundation for Infectious Diseases and a frequent host of its annual flu vaccine press conference, Schaffner has long advocated immunization for healthcare workers and others as recommended by the CDC. As mentioned, the Canadian study focused on the questionable prevented-mortality estimates linked to vaccinating healthcare workers, but there are secondary benefits that should also be noted, he says.
“That is not the only issue to consider,” Schaffner says. “We also don’t want to make patients ill with influenza, and that issue is not addressed. They put much of their emphasis on the most severe endpoint. Similarly, they didn’t look at any other potential benefits of immunization such as worker absenteeism during an influenza outbreak, or the extended [protective] benefit to the worker’s family. These may not be the primary issues underlying a mandate, but they sure have come up in virtually every discussion of influenza vaccination mandates that I have read or written about.”
Schaffner says the issue boils down to a central question: How much evidence is necessary in order to make a recommendation for a mandate?
“From my point of view, the accumulated evidence, including and beyond the RCTs, is sufficient on a scientific, professional, and ethical basis,” he says. “There must be 20 professional societies now that have agreed with this [mandated] position.”
The author of one of the recently criticized randomized controlled trials on the protective effect of immunizing healthcare workers against influenza defends his research, but emphasizes it applies only to long-term care settings.
A study1 by Canadian researchers found that the four randomized controlled trials (RCTs) commonly cited to justify compulsory immunization lack scientific rigor and overstate the protective benefit to patients.1
In an article2 published in the same journal, Andrew C. Hayward, MD, a member of the National Institute for Health and Clinical Excellence guideline development groups reviewing influenza vaccination in healthcare workers, defended his 2006 RCT.3
The study was designed to assess whether promoting influenza vaccination of staff would reduce influenza-related morbidity and mortality in elderly residents of long-term care facilities, he writes. “We hypothesized that the effect of the vaccine would be confined to periods when influenza was circulating and undertook the study over two years to minimize the risk of being unable to demonstrate an effect in a year with low levels of influenza circulation,” Hayward writes. “…During the period of more intense influenza circulation, we found highly statistically significant reductions in residents’ influenza-like illness (9 fewer reports per 100 residents in intervention vs. control homes).”
Although the findings were highly significant, the statistical confidence intervals were beset with some uncertainty about the full scope of the benefit gained through vaccination.
“We found no significant decreases in any of our outcomes during periods when influenza was not circulating in the community or in the second year when influenza rates were substantially lower than the first,” Hayward notes. “In our discussion, based on our own findings and those from other studies, we concluded that healthcare worker vaccination provides an important level of resident protection in long-term care facility settings. While we claimed that the findings may be generalizable to other settings, we did not intend to imply that the extent of the benefit would be similar in other settings. Indeed, we think the effect is likely to be substantially greater in long-term care facilities for frail elderly residents than in the acute care setting or in long-term care facilities catering for less frail patients.”
Financial Disclosure: Senior Writer Gary Evans, Editor Dana Spector, Editor Jill Drachenberg, Reviewer Patti Grant, RN, BSN, MS, CIC, Reviewer Patrick Joseph, MD, and AHC Editorial Group Manager Terrey L. Hatcher report no consultant, stockholder, speaker’s bureau, research, or other financial relationships with companies having ties to this field of study.