Read on Twitter
This thread explores data from the ongoing Covid-19 wave in India. What can we understand about how the wave happens ?
First, let us look at how many >45 now have at least 1 dose of vaccine, and the CFR. By end April ~45% of high risk group will have at least one shot.
1/
First, let us look at how many >45 now have at least 1 dose of vaccine, and the CFR. By end April ~45% of high risk group will have at least one shot.
1/
All data from https://www.covid19india.org/
First let us look at Maharashtra, the most publicized case in this wave. All data below are daily numbers:
2/
First let us look at Maharashtra, the most publicized case in this wave. All data below are daily numbers:
2/
Observations:
1. In early March itself, rising confirmed cases were an issue
2. but daily deaths remained low before the April spike
3. Testing and vaccinations keeping up.
4. TPR does not offer much early warning in March.
5. Active case count offered warning
3/
1. In early March itself, rising confirmed cases were an issue
2. but daily deaths remained low before the April spike
3. Testing and vaccinations keeping up.
4. TPR does not offer much early warning in March.
5. Active case count offered warning
3/
Let look at GMR data: https://www.gstatic.com/covid19/mobility/2021-04-18_IN_Maharashtra_Mobility_Report_en-GB.pdf
1. Drop in retail and transport, but shops/pharmacy crowded.
2. Stricter distancing measures = flattening of daily confirmed cases, large fall in active cases.
3. Suggests wave peaked in MH, must keep up restrictions.
4/
1. Drop in retail and transport, but shops/pharmacy crowded.
2. Stricter distancing measures = flattening of daily confirmed cases, large fall in active cases.
3. Suggests wave peaked in MH, must keep up restrictions.
4/
In MH the wave built over time. Now lets look at Chattisgarh:
5/
5/
Observations:
1. Very little build up here - from 2nd half of March it rapidly gains cases
2. Unlike MH, deaths spike much above prior wave.
3. Not much notice from active cases or TPR. Previous spikes didn't grow.
4. Picked up vaccination pace after wave began.
6/
1. Very little build up here - from 2nd half of March it rapidly gains cases
2. Unlike MH, deaths spike much above prior wave.
3. Not much notice from active cases or TPR. Previous spikes didn't grow.
4. Picked up vaccination pace after wave began.
6/
GMR data : https://www.gstatic.com/covid19/mobility/2021-04-18_IN_Chhattisgarh_Mobility_Report_en-GB.pdf
Until beginning of April, there was little drop in mobility, which probably contributed to the strong wave in this state. However it has dropped a lot, and case counts look to have peaked here.
7/
Until beginning of April, there was little drop in mobility, which probably contributed to the strong wave in this state. However it has dropped a lot, and case counts look to have peaked here.
7/
Now let's look at Delhi:
8/
8/
Observations:
1. Like CG there's little warning before April wave.
2. As in CG, DL vaccinations spiked in April (>45 eligible).
3. Case counts and deaths still on upward move.
4. Almost 50% of over 45 group vaccinated at least once now.
9/
1. Like CG there's little warning before April wave.
2. As in CG, DL vaccinations spiked in April (>45 eligible).
3. Case counts and deaths still on upward move.
4. Almost 50% of over 45 group vaccinated at least once now.
9/
GMR data: https://www.gstatic.com/covid19/mobility/2021-04-18_IN_Delhi_Mobility_Report_en-GB.pdf
Unlike MH, DL was later to start uncrowding. However this has happened now, and DL might peak in a week. Crucial to maintain social distancing but keep vaccinating fast.
10/
Unlike MH, DL was later to start uncrowding. However this has happened now, and DL might peak in a week. Crucial to maintain social distancing but keep vaccinating fast.
10/
Now to Karnataka:
11/
11/
Observations
1. As with DL and CG there's very little notice.
2. No active case/TPR build up before acceleration in daily deaths.
3. 45% of high risk group vaccinated so far.
12/
1. As with DL and CG there's very little notice.
2. No active case/TPR build up before acceleration in daily deaths.
3. 45% of high risk group vaccinated so far.
12/
GMR data: https://www.gstatic.com/covid19/mobility/2021-04-18_IN_Karnataka_Mobility_Report_en-GB.pdf
Karnataka needs urgent decrowding measures NOW. This is key to peaking quickly. Still far more crowded than DL and MH.
Continued crowds = continued transmission.
13/
Karnataka needs urgent decrowding measures NOW. This is key to peaking quickly. Still far more crowded than DL and MH.
Continued crowds = continued transmission.
13/
Now to WB:
14/
14/
Observations
1. As in all places but MH, no warning of wave intensity
2. Active cases, TPR , deaths all spike in unison
3. Weak vaccination cover (only 34% of high risk group have at least one dose, below national average)
15/
1. As in all places but MH, no warning of wave intensity
2. Active cases, TPR , deaths all spike in unison
3. Weak vaccination cover (only 34% of high risk group have at least one dose, below national average)
15/
GMR data: https://www.gstatic.com/covid19/mobility/2021-04-18_IN_West_Bengal_Mobility_Report_en-GB.pdf
As in KA the problem is apparent, social distancing measures need to increase quickly - WB might otherwise take the longest to top out. It also has a lot of distance to cover in vaccinations.
16/
As in KA the problem is apparent, social distancing measures need to increase quickly - WB might otherwise take the longest to top out. It also has a lot of distance to cover in vaccinations.
16/
Next is MP:
17/
17/
Observations:
1. Shows trends similar to WB in some ways -high death spike due to only 35% vaccination of high risk group
2. However, had TPR warning and started social distancing sooner, as mobility data shows next.
3. Active case counts falling, appears past peak.
18/
1. Shows trends similar to WB in some ways -high death spike due to only 35% vaccination of high risk group
2. However, had TPR warning and started social distancing sooner, as mobility data shows next.
3. Active case counts falling, appears past peak.
18/
https://www.gstatic.com/covid19/mobility/2021-04-18_IN_Madhya_Pradesh_Mobility_Report_en-GB.pdf
Started ensuring higher social distancing in public places by early April, and results are showing in active case counts now. This is key - multiple states show a ~2 week time delay to halt the wave when they do this.
19/
Started ensuring higher social distancing in public places by early April, and results are showing in active case counts now. This is key - multiple states show a ~2 week time delay to halt the wave when they do this.
19/
Next is Kerala, where things look a little unique:
20/
20/
Observations:
1. Has 67% vaccination cover of high risk group, which helps keep deaths in control. Vaccination has been continuous, not a spike since April as in CG.
2. KL didnt entirely come out of wave 1 before 2 began.
2. Short lull before active cases/TPR spike.
21/
1. Has 67% vaccination cover of high risk group, which helps keep deaths in control. Vaccination has been continuous, not a spike since April as in CG.
2. KL didnt entirely come out of wave 1 before 2 began.
2. Short lull before active cases/TPR spike.
21/
https://www.gstatic.com/covid19/mobility/2021-04-18_IN_Kerala_Mobility_Report_en-GB.pdf
Needs more aggressive measures for social distancing at work and shops. Has a case count spike now, but only 0.40% CFR which it can maintain due to high vaccination cover. Main risk is stress on hospital resources.
22/
Needs more aggressive measures for social distancing at work and shops. Has a case count spike now, but only 0.40% CFR which it can maintain due to high vaccination cover. Main risk is stress on hospital resources.
22/
Next Gujarat, which is also a little strange:
23/
23/
Observations:
1. Had a little advance warning, but drop in active cases end March probably affected response.
2. Unclear why deaths peak so much despite 57% high risk group now vaccinated, and sustained effort to vaccinate. More data needed.
24/
1. Had a little advance warning, but drop in active cases end March probably affected response.
2. Unclear why deaths peak so much despite 57% high risk group now vaccinated, and sustained effort to vaccinate. More data needed.
24/
https://www.gstatic.com/covid19/mobility/2021-04-18_IN_Gujarat_Mobility_Report_en-GB.pdf
Mobility data doesn't suggest much crowding - almost all parameters show decrowding through April so far.
This is a little perplexing - rising vaccinations and distancing, but still high death counts and case counts. How can this improve ?
25/
Mobility data doesn't suggest much crowding - almost all parameters show decrowding through April so far.
This is a little perplexing - rising vaccinations and distancing, but still high death counts and case counts. How can this improve ?
25/
TN: another perplexing case.
https://www.gstatic.com/covid19/mobility/2021-04-18_IN_Tamil_Nadu_Mobility_Report_en-GB.pdf
26/
https://www.gstatic.com/covid19/mobility/2021-04-18_IN_Tamil_Nadu_Mobility_Report_en-GB.pdf
26/
Observations:
1. Had some warning in active case and TPR in late March but cases grew fast
2. Despite low (23%) high risk vaccination cover, very few deaths.
3. Has not instituted strong social distancing measures. Low deaths should not lead to complacency.
27/
1. Had some warning in active case and TPR in late March but cases grew fast
2. Despite low (23%) high risk vaccination cover, very few deaths.
3. Has not instituted strong social distancing measures. Low deaths should not lead to complacency.
27/
Observations:
1. Highest case fatality rate in the country. Low vaccination cover (34%) and very little effort to social distances.
2. Somehow still managed to keep active cases from accelerating, but not out of woods yet.
3. Had early warning from TPR/active cases.
29/
1. Highest case fatality rate in the country. Low vaccination cover (34%) and very little effort to social distances.
2. Somehow still managed to keep active cases from accelerating, but not out of woods yet.
3. Had early warning from TPR/active cases.
29/
Rajasthan:
https://www.gstatic.com/covid19/mobility/2021-04-18_IN_Rajasthan_Mobility_Report_en-GB.pdf
30/
https://www.gstatic.com/covid19/mobility/2021-04-18_IN_Rajasthan_Mobility_Report_en-GB.pdf
30/
Observations:
1. Another case of very little warning, high deaths despite higher than average (54%) vaccination cover.
2. However, had inconsistent vaccination at start of wave.
3. Given attempts to social distance and vaccinate, will probably peak in a week or so.
31/
1. Another case of very little warning, high deaths despite higher than average (54%) vaccination cover.
2. However, had inconsistent vaccination at start of wave.
3. Given attempts to social distance and vaccinate, will probably peak in a week or so.
31/
Summary
1. Pick any state, go to Feb/March, how easily can anyone anticipate ~10X rise in case counts within 2 weeks from there ?
2. Yes there is an ongoing wave. Most states show peaking. Some states need more aggressive measures quickly.
32/
1. Pick any state, go to Feb/March, how easily can anyone anticipate ~10X rise in case counts within 2 weeks from there ?
2. Yes there is an ongoing wave. Most states show peaking. Some states need more aggressive measures quickly.
32/
Covid is scary this way - it's very hard to predict when its dormant and when a variant runs wild suddenly. There is no single cookie cutter approach. Some states trying hard have seen a worse crisis than others not doing as much.
33/
33/
Put aside the impulse to blame and look at any or all states - how easy is it to anticipate and frame policy here ? One can cope and respond. Most states are already near/at peak. Other countries have also encountered waves.
34/
34/
EU and US were affected horribly in the first wave last year, but India not so much. Now we're facing a major wave, but have the benefit of ongoing vaccination that has kept CFR much below EU/US numbers.
35/
35/
Their experience has been to ride it out, and push vaccinations as hard as possible. At least the US has done so. India needs to even more urgently expand vaccine production. Imports will not work out quickly - everyone is hoarding to get done before their next wave.
36/
36/
One of the scariest things of this is it impacted rich and poor - famous personalities became sick or tragically died. Access to a hospital is not a guarantee of survival, but vaccination makes a major difference. Until then, maintain distancing/hygiene measures.
37/
37/
Its tempting to find someone to blame, but the data shows extremely low predictability. US/EU were hit badly before any vaccine was ready. India is hit now, but we have vaccines.
Everyone has a role here to help us ride this out.
38/38
Everyone has a role here to help us ride this out.
38/38
