My Authors
Read all threads
A recent serosurvey by NCDC at start of July of Delhi suggests ,that 23.48% of Delhi population had IgG antibodies against SARS-COV2 .(implying thia many were exposed till 2 weeks back mid of
June)
Commercial lab Antibody tests suggest similar Estimates
This implies wider spread than previously thought. However Infection fatality rate (Deaths typically taken 7-10 days back after Antibody formation or modelled as log normal distribution) has been estimated at 0.07% (3700 death/ 43-45 lakh affected out of 1.9 crore)..
It was similar in ICMR sero-survey earlier.
However these figures are very less than Estimates in Western Countries . A recent meta-analysis suggest a median IFR of 0.6%. CDC at start of June has revised its estimate multiple times [0.26% to 0.6%. ]
I must admit I was also confused about these Estimates and my knowledge has evolved with time

IFR is not a fixed figure. Fatality changes with our Non-pharmaceutical (mask, lockdown)and pharmaceutical interventions(drug,vaccine) ,ethnicity. Thus it is a moving Estimate.
A famous Scientist John Ioannidis was mocked when in his meta-analysis he suggested IFR can be low as 0.2% (and almost compared it to flu) in early part of pandemic. In its modelling of disease trajectory Imperal College London Estimated IFR to be around 1%.
In Blood Donors at multiple western countries,where typically donors are less than 70 years of age and healthy, IFR has been suggested to be 0.07-0.09%.(based on serosurvey of 10000 healthy blood donors).
The low IFR Estimates in India have led to 2 charges .
A) The disease is just like Flu ,so why bother
B) Massive number of deaths are being undercounted (Almost 6 fold) since Real IFR is 6 times higher.
I disagree with both charges and will explain why..
Traditional COVID is like Flu supporters suggest That since IFR of Flu is around 0.1%. This Disease is just like Flu. This is Wrong . The Method used to calculate IFR was based on excess death modelling as opposed to True Positive Count in COVID.
If you calculate IFR for Flu like you do for COVID the IFR would be as low as 0.01%.
medium.com/@gidmk/covid-1…
Multiple papers have been written on this false equivalence.

jamanetwork.com/journals/jamai…
B) Massive Undercounting of Deaths
While I estimate almost 50% of higher deaths than reported, I don't think that we are missing 6 fold death numbers. For following reasons
A) Our population is much younger
(Only about 3% of our population is above 70 years of age. Remember low IFR in less than 70 in serosurvey in Western countries)
B) In Western countries(New York,Geneva,Spain,Italy) where these serosurveys took place a lot of deaths occurred in assisted living facility for elderly where they live together (nursing homes where the spread and deaths was common)contrast from India where they live with family
C) the IFR of 0.6% is derived from Initial parts of Pandemic,when systems were overwhelmed . We have learned a lot since then, like Dexamethasone reduces mortality by 1/3rd.
Using methods like Proning and Low dose anticoagulation. These pharmaceutical methods acting in combination might alone have decreased IFR at least twice.
Let's see this .
I calculated age distribution of Delhi from Census 2011 figures from here. When you plug in with these Estimates from CDC, USA.
The age-adjusted IFR comes to be 0.12%.

Let's calculate Age Stratified IFR for Delhi from 2 widely acclaimed highly standardized sero-survey :Geneva and Spain

When you do calculate ,Age adjusted IFR for Delhi from Spain Serosurvey Estimates.
IFR ~0.16%.

(16 death out of 10000)

From Geneva Serosurvey Estimates.
IFR ~0.24%.

24 deaths out of 10000.

medrxiv.org/content/10.110…
Even , calculations from Lucknow with 4% raw Estimates of Antibody from Health Donors and 57 deaths , on adjusting for test characteristics gives an IFR around 0.1%.
Even in all these places , due to higher population of elderly in population and further higher population resideing in nursing homes has been a major contributor to death and thus as noted IFR can be a bit inflated.

pubmed.ncbi.nlm.nih.gov/32679085/
And these estimates were before June ,when pharmaceutical interventions with mortality benefit arrived contrast from India ,where major mortality numbers are past this .
Example In Delhi only 523 deaths were recorded till June compared to almost 3700 deaths now.
Thus it is entirely possible that our IFR has been impacted by some time, knowledge we gained by having lockdown till May.
In fact , Unfortunately our health system has been so singularly focussed on Diagnosing(now we are getting better at it) and treating COVID that we are neglecting other diseases.
Our Country is not an exception. Mortality rates have been similar in neighbouring south Asian countries with similar age profile (Bangladesh, Pakistan) and culture.
Thus while the low IFR doesn't suggest . It is Flu, we are not massively Undercounting deaths either (3-6 fold). As of now,I would peg the current Infection Fatality Rate for India to somewhere around 0.1%.
Missing some Tweet in this thread? You can try to force a refresh.

Keep Current with Anupam Singh

Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

Follow Us on Twitter!

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!