Thread by @_YogendraYadav

Breaking:
Is the Govt of India saying we have nearly 15 CRORES Corona positive cases?

That more than 1 lakh persons may have already died of Covid? 

That's 500 times more than the official figure, higher than any other country.

Don't believe me?

Follow this thread.

1/9

Govt released results of Part I of Serological survey by @ICMRDELHI

First rigorous random survey of pop.

Covers entire country outside hotspot/containment areas.

Survey done in May, data is for
disease spread on 30 APRIL
(as antibodies take a few weeks to develop) 

2/9

Big finding: 0.73% of population was Corona positive. This may look tiny, but is really big.

A simple comparison:
Positive rate in India (30 April): 7,300/m

Positive rate global (30 April): 254/m

Current global rate: 936/m  (7.29 m cases/ total population of 7790 m)

3/9

Let's translate this in real numbers

India's popln in April 2020 = 137.8 cr = 138 cr

Popln covered by Part I survey (non-hotspot) = 133 crore

Popln covered by Part II (hotspot/containment areas, rough apx.) = 5 cr

Positive cases Part I =
0.73% x 133
cr = 97 lakhs

4/9

Let's add Part II (hotspot cities) not released yet.
This news leak says15-30% of popln positive
ICMR says it's not final.
Hospot or containment zone?

Let's assume:15%

Positive cases Part II =
15% x 5 cr = 75 L

Total (I+II)=97 L+75 L=
1.72 cr cases

newindianexpress.com/nation/2020/ju…
5/9

So we had 1.72 cr cases on 30 April

Official count on 30 April: 34,863

Official count on 11 June: 2,98,283

Increase since 30 April: 8.55 times

Assuming that total count grew at same rate as official count, total cases today =
1.72 cr x 8.55 = 14.7 cr

(Should be lower)
6/9

Now fatalities:

Report says fatality rate is 0.08% of those infected.

So estimated fatality by 30 April: 1.72 crore x 0.08% =13,760 deaths

Estimated total deaths till 11 June: 14.7 cr x 0.08% = 1,17,728 deaths
(Should be lower if growth rate in hotspots is lower)

7/9

Can we be sure? NO. 
This article says method used by ICMR is prone to over-count, especially since proportions are very small.

Some scientists also believe that ICMR figures could be wrong.

Problem is ICMR has not given confidence interval for their estimate.

8/9

So, the Govt must clarify:
1. Does it still stand by the ICMR data?
2. Where are technical details about the survey?
3. What is the confidence interval for this estimate of infection and fatality?
4. When is the Govt going to release part II of the survey findings?
End
9/9