Mortality in the elderly and influenza vaccines: response to Michael Anderson
I thank Michael Anderson for his October 28 rapid response, but am sorry he has such a negative view of my June 8 reference to his article. ( https://bmj.com/content/369/bmj.m2184/rr-4 ) Indeed, the article is a unique addition to our knowledge of influenza vaccines by taking the long view of vaccine impact on mortality and severe illness in the elderly. (Anderson, Dobkin and Gorry. Ann Intern Med 2020;172:445) It is a complex study by three economists, so in September 20 emails I solicited their opinions about a similar reference in another letter. ( https://www.bmj.com/content/370/bmj.m3563/rr-0 ) I have received no replies to my queries…..This is the first time I have heard that my June 8 letter has been shared on social media. I regret this, especially since I scrupulously avoid contact with any social media, perhaps to the point of paranoia!.....Now let me respond to Dr. Anderson’s comments:
1. I agree that the results of his article “do not preclude modest effectiveness of the influenza vaccine against severe outcomes in elderly persons.” They also do not preclude the possibility that influenza vaccines have significantly increased mortality in the elderly. For example, Supplement Table 14 indicates that during six A/H3N2-predominant seasons the vaccines increased the risk of all-cause mortality in elderly men by 16.6% (VE -16.6%, CI -32.2% to -1.1%)
Is this merely an example of the cherry picking Dr. Anderson refers to? I don’t think so, and I urge readers to study the entire article, including the Supplementary Material. As in so many other articles, the abstract cannot be taken at face value. Look carefully at all the data.
2. Some of the study’s results are of potentially large significance for public health and should not be dismissed merely because they are “statistically insignificant.” The results referred to in my June 8 letter were “insignificant” at P=0.09 and P=0.052. The 16.6% mortality increase mentioned above was actually “significant” at P=0.0245. (Since P values were not given in the article or Supplementary Material, I took the trouble to convert vaccine effectiveness and confidence intervals to odds ratios and confidence intervals, and then calculated P values, guided by Altman and Bland, BMJ 2011;343:d2090)…..Incidentally, more than 800 epidemiologists and statisticians have called for the abandonment of “statistical significance” (Amrhein et al, Nature 2019;567:305), a position I recently ratified.
( https://www.bmj.com/content/370/bmj.m3720/rr )
3. I wonder why Dr. Anderson, an economist, so staunchly defends influenza vaccines. During a medical career that began in 1962 seasonal flu shots for healthy individuals have never made sense to me. This has been reinforced by the research and the opinions of distinguished scientists and clinicians, such as Macfarlane Burnet, Kenneth McIntosh, Lone Simonsen, Tom Jefferson, Peter Doshi, Roger Bodewes, Danuta Skowronski, Margaret McCartney, Peter Collignon, and their colleagues. ( https://www.bmj.com/content/371/bmj.m4037/rr-3 ) It has also been reinforced by the knowledge that the CDC habitually exaggerates mortality figures to promote uptake of the vaccines….Over the years I have seen vaccine failures and serious vaccine adverse effects, and I perceive no decline in the frequency of influenza cases or influenza deaths—massive publicity campaigns and frequent reports of “vaccine effectiveness” notwithstanding.
4. I take a dim view of vaccinating healthy children to protect the elderly. There is no evidence that seasonal vaccines save young lives. Furthermore, they subvert the broad and lasting protection afforded by an infection, unpleasant though it may be. The vaccines have caused high fevers, seizures, Guillain-Barre syndrome, oculorespiratory syndrome, and narcolepsy in children and adolescents.
We need to take a hard look at influenza vaccine policies, and Dr. Anderson’s article is one place to start.
ALLAN S. CUNNINGHAM 29 October 2020
Competing interests: No competing interests