Is the ICMR study on sudden deaths reliable? There's more to it than meets the eye..
[This article was first published on November 24, 2023 at https://empirediaries.com/]
A recent study published in the Indian Journal of Medical Research has ruled out any association between Covid-19 vaccines and sudden cardiac deaths. On first glance, the paper looks quite impressive, with over 100 authors spread over 47 tertiary care hospitals. Almost all the premier institutes in the country are involved in this research, as well as many private medical colleges. A large number of researchers from ICMR (Indian Council of Medical Research) also took part in the study. In fact, news headlines are referring to it as an ICMR study on rising cases of sudden death.
The study, with so many heavyweight researchers from premier institutions in the country and with the stamp of authority of the country’s apex medical research body, the ICMR, must be quite reassuring to Indians as the majority of us have taken the Covid-19 jabs. To criticise such a study would perhaps be deemed as blasphemy.
Einstein’s famous response
But taking a lead from Albert Einstein, I cannot restrain myself from conveying the uneasy feeling I am experiencing after reading the paper. In response to the 1931 book, ‘A Hundred Authors against Einstein’, he had famously stated that if he were wrong, then one author would have been enough.
I am no Einstein by any stretch of imagination. And instead of one author being challenged, I am raising some concerns about the work of a hundred authors. I know I may be considered highly presumptuous. But risking this, I offer some questions around the study in simple terms, so that the lay person may follow the reasoning and decide on how much reassurance about the vaccines the study actually offers.
The research in question is an observational study. It is the most primitive form of research, based on epidemiological principles. ‘Epi’ means ‘upon’, ‘demos’ stands for the ‘people’ and ‘logy’ is ‘study’, so epidemiology observes groups (of people) as opposed to clinical medicine, which actually studies individuals. As the epidemiological researcher studies large groups rather than a single patient at a time, observational epidemiological studies have more weight than a single case or a series of cases.
Again, observational studies can be divided into two categories, descriptive and analytical. Description is the occurrence of an event in terms of person, place, and time. For instance, if we consider the present concern, that is, sudden deaths among the young, we can describe these events in terms of person, place, and time.
Countries with good data – alas Bharat does not fall in this category – are showing disturbing trends of all-cause mortality and sudden deaths, particularly of young people. This is the ‘person distribution’. That these deaths are showing a rising trend following mass Covid-19 vaccine roll-out is the ‘time distribution’. The ‘place distribution’ is almost global, particularly in areas with high Covid-vaccine coverage.
The Australia example
Of particular interest is the phenomenon in Australia. The Australian data is not reassuring. Australia, due to its stringent measures, almost had no Covid-19 outbreak. It had only 1,000 cases of Covid-19 in the whole population by December 2021. After the roll-out of mass vaccination against the coronavirus disease, there were 2,172 emergency cardiac events among those in the age group of 15-44 years – up from a baseline of 1,300, an increase of 67%.
Across all age groups, Australians had a 37% increase in cardiac events. What is most concerning is the very high incidence of cardiac events in younger age groups – a 67% increase among those in the age group of 15-44 years compared to 37% increase across all age groups.
All these are descriptive observational studies. An improvement in the order of evidence base medicine is scaling these studies up to observational analytical studies by including a control group. The ideal observational analytical studies are forward-looking studies. These follow people exposed to the suspected cause and those not exposed to it, and note the rate of events in the two groups.
In the present case, the ICMR could have followed vaccinated people and taken as controls unvaccinated people, and followed them forward in time and compare the adverse events, if any in the two groups. Bharat, with its huge data dividend due to a large population of young people both vaccinated and unvaccinated, had the ideal conditions for such a study. With its much touted “transformative digital tools”, such a study would have guided the world.
Instead, for inexplicable reasons, the ICMR study chose a case-control study to resolve the question of any association of the Covid-19 vaccines with sudden deaths among young people. The case-control study provides the weakest and sometimes misleading conclusions due to inherent biases and confounders. A quick and dirty study method, it cannot provide a final verdict.
The case-control study starts with the identification of cases – in this particular study, these were sudden deaths – and it chooses controls that do not experience the event. In this investigation, the authors identified 729 cases and 2,916 controls for the analysis.
ICMR study on sudden deaths: The bottom line
To cut a long story short, the authors concluded that there was no association between Covid-19 vaccination and sudden deaths as the odds of a person dying suddenly having received the vaccine was more or less similar to the odds of a control having received the vaccine.
While there are many flaws in this study with a rather unwieldy number of authors, institutions, and some heavy statistics – enough to impress and intimidate the lay person in equal measure – this commentary will restrict to some basic limitations of observational case-control studies in general and some specific fundamental flaw in the present study in particular, which compromises the validity of the conclusions from the paper.
Observational studies are at the level of ‘police panchanamas’ recorded by constables on the spot. Police panchanamas are the most preliminary investigations, the findings of which have to be confirmed by more refined techniques, such as fingerprint reports, forensic examinations, and so on. Based on the totality of evidence, hearings, and cross-examinations in the court of law, the final verdict is given.
Similarly, observational studies in research are most primitive, and jumping to conclusions and giving a clean chit to the vaccines on the basis of one case-control study is amateurish, if not irresponsible. At stake are young human lives.
ICMR study on sudden deaths: Multiple concerns
There are more concerns around the ICMR study. The limitation of comparative analytical study is that if the exposure under question is universal, it is impossible to identify it as a cause however strong the association. For example, if smoking were universal – that is, if all inhabitants on Earth were smokers – it would have been impossible to identify smoking as a cause of lung cancer.
In such situations, historical controls can offer some solutions in taking a call. One example is the data from Australia described above – rise in sudden deaths among the young after mass vaccination in a population which had almost negligible Covid-19 infections earlier.
Against this background, the selection of controls is the biggest flaw in the ICMR study, which renders the whole study null and void. The authors of the study say the controls were selected from the neighborhood of the cases. Given the mass-vaccination drive implemented with gusto, the neighbours in all likelihood would have taken the vaccines in equal measures as the cases (sudden-death victims).
In case-control study, this oversight (usually committed by amateurs in their zeal), is termed, ‘overmatching’, that is, matching for the variable under investigation – which, in this case, is the vaccine. This will efface any cause-effect association, however strong! Elementary, my dear Watson!
A faulty panchanama may reassure but mislead the lay person as well as the scientific community (which seems to have lost common sense completely!). Such panchnamas should be discarded and the confirmation of cause-effect should depend on the totality of evidence, such as robust randomised controlled trials, cohort studies and backed with post-mortem studies, including histopathology and biochemical markers.
We expected better from our premier institutes and the ICMR. They disappointed us.
(Dr. Amitav Banerjee was a field epidemiologist for over two decades in the Indian Army. He led the mobile epidemic investigation team at the Armed Forces Medical College in Pune, India, from 2000-2004. During this period, he investigated a number of outbreaks in the country. He is presently a professor at a medical college in Pune and an academic editor at PLOS ONE journal.)
