Authors’ response

Sir,

We thank the author for seeking clarification about our paper1. Our response to each of the points is provided herewith.

We acknowledge that long-term follow up data from the randomized controlled trials on the COVID-19 vaccine carried out in India would have been an excellent data source for calculating the risk of major adverse events consequent to receipt of the vaccines. However, to determine the risk of a rare event like unexplained sudden deaths (estimated annual incidence: 0.8-6.2/100,000), it would require a large sample size or a sufficiently long duration of follow up. The sample sizes of Covaxin and Covishield trials in India were 25,800 and 1600, respectively. Needless to state that we designed this case–control study as we did not have access to the data of these trials. Although long-term large cohort studies are ideal, well-designed and meticulously implemented case–control studies with careful evaluation of its limitations can provide robust evidence towards determining cause-and-effect relationships.

The median duration of 266.3 days for controls refers to the time between vaccination and interview. The difference in the percentage of one-dose recipients between cases and controls was not statistically significant (Table II of our paper). As reported in the limitations section, we were able to get the date of vaccination for 288 of the 577 vaccinated cases. More details are given in the discussion section of our paper.

In row 11 of Table III, we have reported ‘Binge drinking 48 h before death/interview’ with an adjusted matched odds ratio of 5.96 (95% confidence interval: 2.44-14.6) according to the multivariable analysis. A subgroup analysis by type of vaccine did not indicate any differences among them. However, we wish to point out that most of the cases and controls had received Covishield™, and those reporting other types of vaccines were too small to make any meaningful comparisons.