I would be reluctant to change Covishield dose interval at a time when we still have intense transmission: Gagandeep Kang

Even as the government flatly denies that its decision to increase the Covishield dose interval to 12-16 weeks is not due to vaccine shortage and is instead based on real-world data, there is no such real-world data to support its decision while it is common knowledge that acute shortage of vaccines is leading to a drastic drop in daily vaccination across India.

After approving Covishield vaccination with a four-six-week interval between two doses despite evidence showing higher protection when the gap was more than 12 weeks, India changed the interval to six-eight weeks on March 22 when shortage of vaccines became apparent. With shortage of vaccines becoming more acute since then, on May 13, India once again changed the gap between two doses to 12-16 weeks even as highly transmissive variants were widespread in India. Dr. Gagandeep Kang, Professor of Microbiology at CMC Vellore explains the implications of the changes made in the midst of the second wave.

India approved Covishield with an interval of four-six weeks between two doses despite the MHRA approving AstraZeneca with a 4-12 weeks gap and JCVI approving a 12-week prime-boost interval. What was the Indian drug regulator’s rationale for deciding the second dose schedule four-six weeks after the first dose?

The Indian regulator’s rationale for deciding the second dose schedule at an interval of four to six weeks was based on what had been done in the clinical trials that were conducted in India.

A paper in The Lancet based on a pooled analysis of four randomised trials found vaccine efficacy reached 82.4% when the second dose was given 12 weeks after the first but only about 55% when given less than six weeks after the first dose. So why did India not increase the interval between two doses to 12 weeks in early February?

In my opinion India should have increased the interval between two doses to 12 weeks in February. However, the committees that decide on policy for India were clearly unconvinced that a 12-week gap was supported by evidence.

What evidence was available for India to change the interval between two doses to six-eight weeks on March 22?

From the clinical trials and the real-world evidence that was available by mid-March there was no data that distinguished between specifically periods up to eight weeks and those beyond eight weeks.

On May 13, the government changed the interval between doses to 12-16 weeks based on real-life evidence. But even in the UK the interval was only up to 12 weeks. So where did the government get real-life evidence to change it to 12-16 weeks?

As far as I am aware, the only data that is available beyond a period of 12 weeks is the data from the phase-3 clinical trials. Spain and Canada had proposed longer intervals, but I am not aware of real-world data from those settings.

According to a May 24 preprint, in the UK, the effectiveness of a single dose of AstraZeneca/Covishield was only 33.5% against the B.1.617.2 variant and 51.1% against the B.1.1.7 variant. After the second dose, the protection increased to 59.8% against the B.1.617.2 variant and 87.9% against the B.1.1.7 variant. With the B.1.617 and B.1.1.7 variants being the most prevalent in India should we not be reducing the gap between two doses especially since the protection against B.1.617 from the first dose is just 33.5%?

I think it is important to emphasise that the report from the UK which shows variant specific protection is only against all symptomatic infection and does not consider severe disease. Since the goal of vaccines is to prevent severe disease and deaths, either from India or from the UK we should be generating the data that tells us how well vaccines are performing with a single dose against severe disease. The UK decision on shortening the interval is not based on available evidence but does rely on the fact that they have immunised all adults with the first dose and have supplies to be able to vaccinate their population with the second dose.

Since the B.1.617.2 variant is fast-spreading in the UK, the government has reduced the gap between two doses to eight weeks in those over 50 years and the clinically vulnerable population. Under these circumstances, what is the justification for India to increase the interval between doses to 12-16 weeks?

There are two ways to make a decision. One is to have data to support the decision. Currently no country has data to inform how well vaccines work with an eight-week gap. In fact, given that India was the only country to have a six-eight-week gap between doses from March, and in above 60 years, we could potentially be able to provide that data to the world.

The second way to make a decision is to model different scenarios that can be varied for predicted vaccine performance against variants and use that to decide whether and what policy change is required. I assume the UK did this.

I am not aware that we have any data or any model that states that the decision was incorrect. 

How much of real-world data are available to support a gap of 12-16 weeks for Covishield when B.1.617 variant is widespread? 


Though the first dose of Covishield has 51.1% against the B.1.1.7 variant, the second dose increases the protection to 87.9%. Since B.1.1.7 variant too is widespread in India, is it correct to increase the gap between two doses?

Against severe disease, a single dose of Astra-Zeneca vaccine has 80% protection. If the goal is to prevent severe disease, then it makes sense to use available doses to immunize more people with a single dose. For me, in the absence of evidence of performance of a single dose against severe disease due to B1.617.2, I would be reluctant to change at a time when we still have intense transmission, and the opportunity to modify disease severity in a larger population with a single dose.

Is the logic of partially protecting more people with a single dose than increasing the level of protection in fewer people correct given that the first dose is only 33.5% protective against the B.1.617.2 variant, which is the most prevalent in India today?

Once again, these data are for all symptomatic infection. We have not reached a stage in vaccination where we are trying to prevent all disease or all infection.

Will a partially protective (33.5%) single dose lead to the generation of vaccine escape variants?

No, since 33.5% does not mean partial protection of a person. Protection is not measured at the level of the individual, but at the level of the population. Vaccine escape variants could be generated in persons in whom the virus can continue to replicate and mutate in the presence of antibodies—such individuals are usually immunocompromised. 

Considering the shortage, can people already infected receive two doses at 12-16 weeks while the rest get it at 8-12 weeks interval? 

Previously infected people have some protection, we do not know yet whether they might be equally protected with just a single dose. However, for the programme to vary schedule based on history of infection is difficult.

Published in The Hindu on May 29, 2021