Intended for healthcare professionals

Rapid response to:

Editor's Choice

Covid 19: Two million deaths, so what went wrong?

BMJ 2021; 372 doi: (Published 11 February 2021) Cite this as: BMJ 2021;372:n393

Read our latest coverage of the coronavirus outbreak

Rapid Response:

Universal Covid-19 vaccination may be a chimera

Dear Editor

Fiona Godlee says “The phenomenal success of the vaccination programme” is one of the sources of hope in the otherwise grim history of the Covid-19 pandemic. (BMJ 2021;372:n393, Feb 11) Gavin Yamey says “The rapid development of safe, high efficacy covid-19 vaccines in under a year is one of the greatest scientific achievements in recent history.” (BMJ 2021;372:n373, Feb 10)….I suggest that we need to be careful about overenthusiasm.

Not many years ago US news releases from the CDC were announcing that influenza vaccines were “70%...80%...90% effective.” (Fiore et al, MMWR Recomm Rep 2010;59:1) Times changed, and in 2018 we were told that vaccine effectiveness (VE) was only 10% to 60%. ( Feb 15, 2018) We have since learned that VE rapidly declines during the flu season. During two recent seasons, for example, VE against A/H3N2-associated hospitalizations among the elderly declined progressively from 51% two weeks after vaccination to –16% at 6 months; that is, a 16% increased risk of serious illness by 6 months. ( ) This report gave no data for the first two weeks following vaccination, but at least one study has shown that flu shots substantially increased the risk of influenza-associated hospitalizations during the first week. (Emborg et al, BMJ 2011;344:d7901)….In retrospect, universal influenza vaccination may have done more harm than good. The vaccines provide modest short-term protection against seasonal influenza, but the observational studies showing 10% to 60% VE completely ignore vaccine adverse effects, including an increased risk of illness from non-influenza respiratory viruses, such as RSV, rhinoviruses, coronaviruses, etc. ( )

Maybe Covid-19 vaccines will be different, but the foregoing history should be kept in mind when we are told that a Covid-19 vaccine is 94% effective overall and 86% effective in the elderly. (Baden et al, NEJM 2021;384:403, Feb 4) This is especially important during the current frenzy driving universal vaccination. Dr. Joseph Ladapo calls universal Covid-19 vaccination a “chimera.” (Wall Street Journal, 2/5/21) He is a hospitalist and an associate professor of medicine in California, and his essay is worth quoting at length:

“Each stage of the American Covid-19 pandemic has been marked by a singular public-health message that crowded out all other perspectives. From early calls to ‘crush the curve’, with shutdowns and pleas to stay at home, then to claims that face masks would end the pandemic, these messaging strategies have sowed unrealistic expectations and delayed public acceptance of reality. The most recent message is ‘universal vaccination’, an aspiration whose unattainability may further delay the country’s return to social and economic normalcy.”

He goes on to discuss the media campaign to stoke fear, the lack of transparency of public-health officials, the uncertainties about vaccine safety and effectiveness and officialdom’s dismissal of legitimate concerns as ‘misinformation’, and the limitations in our ability to stop the spread of Covid-19 and its variants. He supports vaccination, but concludes that vaccine policy should focus on the elderly and truly vulnerable individuals, and on people who feel they can’t live or work safely without vaccination. He condemns proposals to mandate Covid-19 vaccine for schoolchildren. Dr. Ladapo now looks forward to a decline in Covid-19 mortality and more rational thinking about the pandemic. (WSJ, 2/5/21)

The pandemic in the US does appear to be on the decline. Reported Covid-19 deaths peaked in week 16 of 2020, then declined and rose to another peak in weeks 51-52 before starting the current decline. Compared with the 2017-19 average, there was an excess of 468,085 deaths from all causes in 2020, an increase of 16.6%. There were 359,781 Covid-19 deaths reported, accounting for 76.9% of all-cause deaths in 2020 ( Feb 5, 2021)

As of February 5, 2021 there had been 26.8 million US cases of Covid-19 recorded, never mind precisely how the diagnoses were made. ( This is about 8% of the entire US population. Seroprevalence surveys suggest that this figure vastly understates the number of Americans actually infected by Covid-19. From January 21 to November 15, 2020 the CDC recorded 10,846,373 cases; seroprevalence surveys estimate the number actually infected was 46,910,006, more than fourfold higher. (JAMANetworkOpen.2021;4(1)e2033706.doi:10.1001/jamanetworkopen.2020.33706) It appears, then, that something upwards of 8% to 35% of Americans have been infected so far. During influenza epidemics attack rates in unvaccinated populations are estimated to be 10% to 20%, but rates as high as 40% to 50% have been reported. (Mandell et al, Infectious Diseases, 2010, page 2268)

The current US decline in Covid-19 may indicate that natural infections have produced a level of population immunity sufficient to end the pandemic, or are approaching that level. There are still uncertainties, not the least of which are the breadth and duration of immunity provided by Covid-19 vaccines. In general, vaccine immunity to infections is not nearly as good as natural immunity, and is less likely to sustain population immunity. In the long run the safest and most efficient approach may be focused protection of people who are truly vulnerable, while allowing broad and lasting natural immunity to develop in the ~80% of the population who are not particularly vulnerable. I admit that this may be hard to do in the face of highly emotive news stories about Covid-19 deaths in young people, ICUs filled to capacity, highly contagious variants, etcetera….but we shouldn’t forget the history of seasonal flu shots. ( )

The pandemic has brought at least one surprising piece of good news: there were fewer US infant deaths in 2020. Data are still provisional, but based on current figures the projected total will be around 19,200 ( ), compared with 20,300, the figure expected from the trend in infant deaths between 2015 and 2019. (Ely et al, Infant Mortality in the United States, 2018; NVSS Volume 69, Number 7, July 16, 2020. NCHS/NVSS Data Brief 395; Mortality in the United States, 2019) The difference is highly significant, P<0.00001. What accounts for the drop? Specific data about the causes of 2020 infant deaths should help answer the question.

ALLAN S. CUNNINGHAM 14 February 2021

Competing interests: No competing interests

14 February 2021
Allan S. Cunningham
Retired pediatrician
Cooperstown NY 13326 USA <