If there’s one thing about the COVID-19 pandemic that’s surprised pundits and my colleagues in the medical profession, it’s just how many vaccine-hesitant and outright antivaccine healthcare professionals there are out there. You would think that an antivax physician or nurse would be incredibly rare, but the pandemic has demonstrated that they are, unfortunately, far from rare. I’m not saying that the medical profession is dominated by antivaxxers, as in general vaccine acceptance is still higher among physicians and nurses than among the general public. However, there is a much more sizable minority of physicians and nurses out there than my colleagues would have previously guessed whose views range from vaccine-hesitant to outright antivax conspiracy theorist, as a new survey shows.
Before I get to the survey, though, let me just observe one thing. Contrary to what my colleagues (and, truth be told, most pundits) appear to think, a higher than totally rare prevalence of antivax beliefs among physicians is not new. While I have no doubt that antivax views among physicians and nurses have likely increased since the pandemic, given the politicization of COVID-19 vaccines, a politicization that is also not new. Vaccine mandates (and even support for routine vaccination) have become increasingly politicized. Beginning well over a decade ago, if not longer, what used to be one of the most nonpartisan and apolitical polices, school vaccine mandates, have become increasingly partisan, with resistance to school vaccine mandates of any kind becoming increasingly associated with the right and support for them with the left. Again, this predates the pandemic, although the watershed moment that turbocharged the politicization of school vaccine mandates was arguably SB 277, the law passed in California in 2015 after the Disneyland measles outbreak that eliminated nonmedical “personal belief exemptions” to school vaccine mandates. The law clearly worked in its intended purpose of decreasing the percentage of unvaccinated California schoolchildren, but it sparked a huge backlash that laid the groundwork for what we see today since the pandemic started. Pundits and my colleagues might be shocked by the increasing association of antivaxxers with the far right, but that started before the pandemic, with antivaxxers marching with far-right militias before the pandemic.
With that background in mind, earlier this month I was made aware of a new survey because Dr. Peter Hotez Tweeted about it:
And my co-author Gavin Yamey appropriately noted:
It is disturbing, but it’s not unexpected. At least, it wasn’t unexpected to me, and, I suspect, neither was it unexpected to those of us who have been paying attention to the antivaccine movement for a long time. Yet MedPage Today spun the story as hugely unexpected:
Levels of vaccine hesitancy among physicians may be higher than expected, with 1 in 10 primary care doctors not believing that vaccines are safe, according to a new survey.
Among 625 physicians, 10.1% did not agree that vaccines were safe; 9.3% did not agree that vaccines were effective; and 8.3% did not agree that they were important, Timothy Callaghan, PhD, of Texas A&M School of Public Health in College Station, and colleagues reported online in Vaccine.
The high proportion of hesitancy among primary care doctors “was certainly a surprise for us,” Callaghan told MedPage Today. “We thought it might be a very small proportion of physicians who hold hesitancy about vaccines given that we have lots of evidence of the safety and effectiveness of vaccines. However, once we dug into the data, we found that concerns about vaccines in general were far more widespread in the physician population than we might have expected.”
May I suggest, then, that given their level of surprise at the findings of their study that Dr. Callaghan and his colleagues haven’t been paying attention for at least the last decade? And he studied vaccine hesitancy before the pandemic? Dr. Callaghan’s very surprise reveals how oblivious large swaths of my profession have been to the problem of misinformation, disinformation, and the doctors who embrace the pseudoscience and even conspiracy theories fueling them. It’s not as though I, along with many others, wasn’t asking what to do about physicians who embrace and spread antivaccine conspiracy theories back in 2015.
It’s also not as though the antivaccine group Physicians for Informed Consent sprang into being two years ago. It’s been around since 2015. It’s not as though, for instance, the Association of American Physicians and Surgeons (AAPS) just popped up in 2020. It was founded in 1943 and has been spreading antivaccine pseudoscience, such as claims that vaccines cause autism and sudden infant death syndrome, promotion of Lupron and chelation therapy to treat “mercury toxicity” due thimerosal-containing vaccines, and even that shaken baby syndrome is a “misdiagnosis” of vaccine injury, dating back to a long time before I first discovered them in 2006. Indeed, just before the pandemic, AAPS published an article by Andrew Wakefield claiming that measles vaccination will produce a “sixth mass extinction“, a claim that was repurposed for COVID-19 vaccinesduring the pandemic by Geert Vanden Bossche. Unsurprisingly, having been antivax for many years before the pandemic, AAPS has added misinformation about COVID-19 vaccines to the rest of its antivaccine conspiracy mongering.
I realize that before the pandemic colleagues like Dr. Callaghan might have comforted themselves with the belief that such views were so fringe as to be negligible, but I’d argue that two physicians (one of whom, Rand Paul, is outright antivax) pandering to antivaxxers during the 2016 Republican Presidential primaries and then President-Elect Donald Trump appointing an AAPS member to head the Department of Health and Human Services might have been a clue that such views, even before the pandemic, were not as uncommon among physicians as we would like to think.
The survey says!
As annoying as I sometimes am with my penchant for saying, “I told you so,” I’m nonetheless glad that this incident prodded him to quantify the effect:
Indeed, the research project was inspired by Callaghan’s own experience with one of his doctors who was not vaccinated and tried to dissuade Callaghan from COVID vaccination.
“It wasn’t my primary care physician, but another one of my doctors realized that I studied issues related to vaccine hesitancy, and over the course of multiple visits, tried to convince me that COVID-19 vaccines weren’t safe and weren’t worth it,” Callaghan said. “It made me question whether this was a one-off, or if we have an actual issue on our hands.”
Ooh! Ooh! I could have told Dr. Callaghan before he did the survey!
(Sorry, couldn’t resist, even when my choice of meme reveals the era in which I grew up.)
Similarly, it’s not unexpected that antivaxxers would be crowing about this study:
This is one time when both sides understand something about a problem. What physicians think about medical matters is important, because physicians are considered authorities in medical matters. So when physicians express doubt about vaccines and vaccine safety, it will have an effect. So when Dr. Callaghan bemoans how such a high percentage of primary care doctors expressing vaccine hesitancy and outright antivax views, he’s right. That is a big problem. When antivaxxers gloat about how high a percentage of primary care physicians (PCPs) express antivaccine and vaccine-hesitant views, they’re correct that it’s important in that it provides them a huge propaganda weapon. They are not, however, correct that this observation somehow validates their views.
So let’s look at the study itself, published in the journal Vaccine titled “Imperfect messengers? An analysis of vaccine confidence among primary care physicians“, which reports the results of a survey about COVID-19 vaccines. I will note one issue with the vaccine that might or might not be a major problem. The survey was carried out from May 14 to May 25, 2021, nearly a year ago, when the vaccines had been approved under an emergency use authorization (EUA) for adults and the Pfizer vaccine had just been approved under another EUA for ages 12-15. On the one hand, I can see how more doctors might have still been a bit hesitant then. On the other hand, I don’t think antivaccine sentiment has gotten better among the general public since then, and physicians, like it or not, do tend to be affected by what’s going on around them. I also can’t help but note just how much the push to vaccinate children has revealed outright antivax views among physicians who argue against the routine vaccination of children against COVID-19, some of whom (I’m talking to you, Urgency of Normal and Great Barrington Declaration physicians) parrot old antivaccine tropes about the virus not being dangerous to children and how “natural herd immunity” is more desirable than vaccine-induced immunity. My guess is that, were the same survey administered today, the results would be even worse than what’s reported now, which is that:
- 10.1% of primary care physicians (PCPs) do not agree vaccines are safe.
- 9.3% of PCPs do not agree that vaccines are effective.
- 8.3% of PCPs do not agree that vaccines are important.
- Physician confidence in COVID-19 vaccines varies across vaccines.
- A troubling proportion of PCPs lack high levels of vaccine confidence.
An additional finding is that “only 5.2% (95% CI: 3.4, 6.9) of our sample of primary care physicians was unvaccinated as of May 2021, far lower than vaccine refusal rates for other types of health care workers, and the American public more generally ”. Again, I like to point out that being vaccinated does not mean that one isn’t antivaccine or vaccine-hesitant. Antivax leader Robert F. Kennedy, Jr., for instance, loves to point out how his children all received vaccines according to the CDC-recommended guidelines at the time, in response to which I like to point out that he didn’t “come out” as antivax until his youngest child had already passed through the age at which children receive the bulk of their vaccines. Another issue that any physician who had to see patients during the pandemic before there was a vaccine had to deal with was how much the fear of COVID-19 motivated us to get vaccinated as soon as the Pfizer vaccine became available to healthcare workers in December 2020. I hope that the investigators will follow up and see how many of these physicians accepted boosters since May 2021.
The authors recognize this, writing:
Of course, it is critical to recognize that the decision by physicians to vaccinate themselves may be impacted by factors beyond vaccine confidence including their high-risk work environments or the expectations of employers. For that reason, it is also important to analyze general physician confidence in vaccination and confidence in the safety of each COVID-19 vaccine as well.
Before addressing this question, let’s briefly look at the methods:
To understand vaccine confidence among primary care physicians, we developed and administered an original national survey to 737 physicians in the United States. The web-based survey was fielded from May 14 to May 25, 2021, using the survey research firm Dynata – a widely respected survey research firm regularly used in social science research , , , . Dynata invited primary care physicians (identified via responses to an initial inventory survey) to participate in our survey from their large, online, opt-in sampling frame of potential survey participants. Of the 737 respondents identified as PCPs by Dynata who began the survey, 625 of [sic] also satisfied additional screeners that we imposed as a precondition for qualification in our study; i.e., by self-identifying as a PCP working in family medicine, internal medicine, or general practice. These 625 physicians serve as the sample for our analysis.
We provide a comparison of our sample with primary care physician population benchmarks in the appendix which demonstrates that our sample closely approximates the demographic characteristics of primary care physicians nationwide. Primary care physician population benchmarks from the Association of American Medical Colleges and Medscape suggest that our sample closely resembles the proportion of primary care physicians who are Hispanic or Asian and produces similar salary estimates. Small deviations are seen between our sample and population benchmarks for gender and race, with our sample slightly under-representing women and Black physicians. Critically however, these differences are small in magnitude. Limitations aside, our study provides the only national sample to date capable of analyzing primary care physician vaccine confidence.
So the authors made an effort to make sure that their survey group matches the demographics of primary care physicians.
I also notice that, when you dig into some of the weeds, there are some even more disturbing findings. While it’s true that 67.4% strongly agreed that “vaccines are safe” (a horribly low number for physicians) compared to 47.7% of the general public, the percentage of physicians who “strongly disagreed” that vaccines are safe was indistinguishable from that of the general public (4.8% versus 4.9%).
There were some even more disturbing results in the weeds, too:
- More physicians “strongly disagreed” with the general statement “vaccines are effective” than among the general public (6.7% vs 2.1%).
- More physicians “strongly disagreed” with the general statement “vaccines are important” than among the general public (5.8% vs. 2.3%).
These are small minorities of physicians, but damn!
Also, if you look at the numbers a bit differently, there are more warning signs:
- The percentage of physicians who “strongly agree” or “somewhat agree” that “vaccines are safe” was 88.8%, compared to 72.1% of the general public. This is not as huge a difference as one might expect, and I note that the percentage of physicians who “somewhat agree” that “vaccines are safe” (21.4%) was nearly indistinguishable from the percentage of the general public who gave that answer (24.4%)—and lower.
- The percentage of physicians who “strongly agree” or “somewhat agree” that “vaccines are effective” was 89.9%, compared to 84.4% among the general public, an even smaller difference.
- The percentage of physicians who “strongly agree” or “somewhat agree” that “vaccines are important” was 89.7%, compared to 86.9% among the general public, an even smaller difference still.
The authors boiled it down thusly:
Two noteworthy patterns emerge from Table 1. First, as expected, primary care physicians are significantly more likely to agree that vaccines are safe and effective than the general public (significance denoted by the non-overlapping confidence intervals presented in the table, which indicate statistically significant differences between groups at the p < 0.05 level, two-tailed). Second and more importantly, however, there is heterogeneity in physician attitudes towards vaccination that the existing literature and physician-driven vaccination promotion narratives do not account for. 10.1% of physicians do not agree (strongly or somewhat) that vaccines are safe, 9.3% do not agree that they are effective, and 8.3% do not agree that they are important. In addition, to the extent that we might expect more vaccine advocacy amongst physicians who strongly agree that vaccines are safe, effective, and important, it is also valuable to explore the proportion of physicians most likely to believe in vaccines. Our results show that only 67.4% of primary care physicians strongly agree that vaccines are safe, and roughly 75% strongly agree that they are effective (75.5%) and important (76.3%).
What accounts for these results, though? The authors did some modeling to try to answer that question. One factor that strongly influenced vaccine hesitancy among doctors was, unsurprisingly, political orientation:
In Model 1, which focuses on confidence in safety, we find that elevated levels of political conservatism are negatively and significantly associated with agreeing that vaccines are safe. We also find that these effects are substantively large. Holding all other covariates in Model 1 at their sample means, we find that while the predicted probability that extreme liberals strongly agree that vaccines are safe is 78%, the likelihood that extreme conservative PCPs feel the same way is just 59% – a 19% difference across the full range of the ideological spectrum.
A similar pattern holds in Model 2, exploring physician beliefs that vaccines are effective. There, we found that holding all other variables at their means, extreme liberals were predicted to be 18% more likely than extreme conservatives to strongly agree that vaccines are effective. In addition, while the predicted probability that PCPs who had not had COVID-19 would strongly agree that vaccines are effective was 77%, it was only 64% for those who had.
Interestingly, political ideology did not impact agreement with the statement, “Vaccines are important”.
In any event, given the increasing ideological polarization, coupled with the observation that physicians tend to lean conservative in their politics compared to the general public, it is perhaps less surprising that so many physicians doubt the safety and efficacy of vaccines in general.
Another interesting result is that PCPs who had contracted COVID-19 were less likely to agree with all three statements (that vaccines are safe, effective, or important). Again, these are general statements, not statements aimed specifically at just COVID-19 vaccines. One has to wonder whether this result is explained by physicians who had COVID-19 and were fortunate enough not to have gotten too sick from it deciding they didn’t need to be vaccinated or whether physicians who were more of an ideological mindset to doubt vaccines before they got COVID-19 were also more likely to be less fastidious about taking precautions to avoid contracting the disease. After all, being antimask, antivax, and anti-COVID-19 interventions in general do tend to cluster.
What about COVID-19 vaccines specifically?
The authors did look at PCP attitudes towards COVID-19 vaccines specifically. At the time, there were three available in the US under EUA, manufactured by Pfizer, Moderna, or Johnson & Johnson (J&J):
Our analysis in Table 3 revealed that primary care physicians were significantly more confident in the safety of the mRNA vaccines of Moderna and Pfizer than in the Johnson & Johnson COVID-19 vaccine. Specifically, we found that while 68.7% and 72.7% of primary care physicians were very confident in the safety of the Moderna and Pfizer vaccines respectively, only 32.1% of physicians were very confident in the safety of the Johnson & Johnson vaccine as of May 2021. When we include those who were in the ‘confident’ category as well, we found that over 90% of physicians were either ‘very confident’ or ‘confident’ in the mRNA vaccines but only 68.0% of physicians fell into those two categories for the Johnson & Johnson vaccine.
This result is, of course, likely to be very much a product of the time period when the survey was administered, given that the reports of serious complications due to a rare form of blood clot associated with the Astra-Zeneca vaccine, which was based on a very similar adenovirus platform to that used for the J&J vaccine, were very much in the news a year ago. (Has it really been that long?) Given that context, I would have been shocked if the results were different.
What to do?
It should not be a surprise that there exists a not-insignificant minority of physicians who do not believe that vaccines are safe and effective, although I will admit that I was a bit surprised that it was roughly one in ten. (I would have guessed a smaller—but not that much smaller—percentage, maybeas low as 5% but certainly no lower; i.e., a smaller but still far from insignificant minority.) Extrapolating, the authors note:
Combined, our results serve to complicate media narratives and governmental initiatives encouraging primary care physicians to serve as leading vaccine promotors. While a majority of doctors are well positioned to take on this role, a troubling proportion of physicians lack high levels of confidence both in vaccines in general and COVID-19 vaccines in particular. For example, if we take a conservative approach and remove those who ‘somewhat agree’ that vaccines are safe from the proportion of physicians that lack vaccine confidence and use the total number of PCPs from the AMA Masterfile as our baseline for PCPs nationwide, our findings still suggest that almost 25,000 primary care physicians nationwide could disagree with the basic belief that in general, vaccines are safe.
These findings suggest that we cannot take it for granted that all physicians are well positioned to serve as vaccine promotors and furthermore, that interventions may be needed to increase vaccine confidence among some physicians. This is particularly true because these physicians could have an outsized impact on public health, with a single physician lacking trust in vaccine safety capable of writing large numbers of medical exemptions from non-COVID-19 vaccination, reducing the effectiveness of vaccine mandates , .
Anyone who remembers the aftermath of SB 277 likely remembers how many physicians were willing to write bogus “medical exemptions” to school vaccine mandates. Indeed, antivax leaders, such as pediatrician “Dr. Bob” Sears, jumped on the grift bandwagon as the bill was nearing passage in the California legislature, and then more joined in, both as a matter of antivax ideology and profit potential. As a result “medical” exemptions to school vaccine mandates soared after passage of SB 277, which led to the necessity of SB 276, a bill designed to close the loophole that allowed doctors to be so prolific in writing such exemption letters. History might not repeat itself precisely, but it does rhyme, as they say.
There’s also a hole in this article in that it only examines PCPs. While it is true that, on an individual patient level, what the PCP says is enormously influential, on a societal level in essence any old doctor with an MD or DO after their name will do to promote both good information about vaccines and misinformation. After all, most of the COVID-19 contrarian physicians whom we’ve been lambasting for promoting COVID-19 misinformation are not PCPs. Dr. Vinay Prasad (who famously likened public health interventions against COVID, including vaccine mandates, to incipient fascism), for instance, is an oncologist, while Dr. Tracy Beth Høeg is a sports medicine doctor. A number of them (e.g., Drs. John Ioannidis, Joseph Ladapo, and Jay Bhattacharya) do not even practice medicine but are nonetheless extremely influential in spreading false messages that COVID-19 is not dangerous and the vaccines are more dangerous than the disease for young people and children; that the pandemic is over; that, consequently, mask and vaccine mandates in schools are unnecessary (Urgency of Normal); and that “natural herd immunity” is the way out of the pandemic (Great Barrington Declaration).
But what to do? In the conclusion of the paper, Callaghan et al. suggest:
Identifying appropriate interventions is an important direction for future research and could include improving virology and vaccine biology education in medical school or echoing existing approaches for the general public by identifying trusted individuals in physicians lives to improve their confidence in vaccines.
In MedPage Today, Dr. Callaghan expands on this idea:
While political affiliation did appear to play some role in beliefs, Callaghan noted a wider problem that might be at play. “There’s not that much training on vaccines and vaccinology … in medical school,” he said. “Most medical students aren’t exposed to in-depth discussions of virology to have those strong opinions.”
“And given the clouded information environment that surrounded COVID-19 in particular, and increasingly, vaccines in general, it remains possible that [physicians] are just relying on what they’re hearing in the news and the misinformation that’s out there, as opposed to best scientific evidence,” he said.
Callaghan noted that “most physicians are well-positioned to serve as our leading vaccine promoters … With that said, 10% is a pretty big proportion to be hesitant. It suggests, for that 10%, we need to do some work in terms of education and potentially, intervention, to increase the level of confidence in that population of physicians so that we’re not at 90% in favor of promoting vaccination, but that it’s closer to 100%.”
Unfortunately, as was pointed out on Twitter:
That is indeed the problem. It’s an aphorism that’s been attributed to various writers over the years that you can’t reason people out of a position that they didn’t reason themselves into“. Or, to cite Adam Savage, Calvin and Hobbes, and even Doctor Who people seem to live by another aphorism, “I reject your reality and substitute my own.”
As much as we physicians would like to view ourselves as immune to the effects of ideology, emotion, and belief that doesn’t flow from science and reason, we are not. Indeed, it is hubris on our part to think that we are less susceptible to unreason than anyone else, but that doesn’t stop us from thinking that. Worse still, because we as a group tend to think of ourselves as more reasonable and less susceptible to conspiracy theories, misinformation, and disinformation than the average person, when we do fall for them we tend to fall hard. Worse still, because we as a group tend to be highly intelligent, we tend to be very, very good at motivated reasoning indeed. Specifically, we are very skilled at asymmetrically seeking out evidence and arguments that support our pre-existing beliefs and finding flaw in evidence and arguments that challenge those beliefs, all while reassuring ourselves that our training as physicians means that our beliefs about medicine are all strictly based on science.
It’s also difficult not to note that physicians as a group have now long been primed to be more willing to accept magical thinking in the form of “complementary and alternative medicine” (CAM), which is now more commonly referred to as “integrative medicine” or “integrative health” It’s an indoctrination process that over the last 30 years has led to a disturbingly high percentage of physicians accepting the validity of rank quackery, including acupuncture and homeopathy and the creation of expensive infrastructure to practice and “study” magic like homeopathy. This led to some embarrassment for a major academic medical center, the Cleveland Clinic, when its director of integrative medicine publicly went full-on antivax three full years before the pandemic hit.
If I sound pessimistic about the likelihood of reversing these attitudes, it’s because I am, at least in the short term. There’s another aphorism—a somewhat sarcastic one—that disproven treatments don’t really disappear from medicine until the last physicians who used them retire or die. Likely, the same will be true with vaccines and physicians doubting them. We can try to educate existing physicians with better science, and doing so likely won’t be entirely ineffective, but it can only go so far. We need to inculcate medical students with critical thinking skills and the ability to evaluate science, but even that is not enough. Medical students need to be taught to recognize misinformation, conspiracy theories, and disinformation, which is something they aren’t taught at all. (In fairness, neither are most college students, and then only if they seek out such learning.) It is true that medical school is arguably too late for this, but it’s the earliest point in physician education that we as a profession control.
Even more importantly, while I welcome physicians who were blissfully unaware of the problem of pseudoscience, quackery, and the disinformation and conspiracy theories used to support them, I also challenge them to actually try to do something about them. I can understand how, not paying attention before the pandemic, you were shocked to learn that about one in ten of your fellow physicians is vaccine-hesitant or even antivax. What I won’t be able to understand is if, now that you know, you don’t try to do anything about it.