Note: I am a stats guy. The learning that I will share is based on interactions with experts far more knowledgeable than me (like @darwinbandoy).

So, let's talk vaccines and numbers.
First, vaccine efficacy = % reduction in disease incidence in a vaccinated group compared to an unvaccinated group under optimal conditions

Reference: who.int/influenza_vacc…
Also: herd immunity factor with no deaths

% of population infected to achieve herd immunity = (1-1/R0) x 100%

Where R0 = basic reproduction number, the no-interventions transmission ave. of the disease for every previously-infected case

Reference: academic.oup.com/jid/article/19…
For example w/ COVID-19 having R0 = 2 or 3, so say 2.5 (reference: imperial.ac.uk/media/imperial…),

(1 - 1/2.5) x 100% = 60% of the population should be infected without dying.
With a vaccine of 50% efficacy administered to 100% of the population, that is still not enough since

0.50 efficacy x 1.00 population = 0.50 or 50% of the population immune =/= 60% immune.
Buying a 50% efficacy vaccine isn't enough.

Especially damaging that even buying for the whole population, it is not sufficient.

Doubly damning is its expensive price.

Triply worse would be lack of published results and data on efficacy itself, especially on side effects.
So, as you can see, our math is

Efficacy x Population Vaccinated = Population Made Immune by Vaccination.
Now, let's try with Moderna, a more expensive vaccine, but much higher efficacy, at 94.5% (reference: google.com/amp/s/newsinfo…)

To achieve 60% herd immunity factor,

Vaccinated pop'n = 0.60 (immune)/ 0.945 (efficacy) = 0.6349 or 63.49%.
In this example, you have more leverage to reach herd immunity with manageable logistics especially for those who cannot be innoculated due to immunocompromised status.
Now there are cheaper vaccines still with even better efficacy than 50%, and it can be seen what makes total sense and why to surrender to 50% efficacy is NOT AN OPTION.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Peter Cayton, the Stats Guy

Peter Cayton, the Stats Guy 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!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @PJACaytonPhD

27 Dec
Scribbles in the DOH situationer.

1) possible holiday resurgence detected.
2) Wrong math again.
3) DOH estimate of Rt is touching 1.
Read 7 tweets
25 Dec
Note:
> 8 cases purged
> 6 recov's purged/changed status

Daily % Change & Crude Days-to-Double
Conf'd= 0.4030%, 112 days
Recov'd= 0.0699%, 97 days
Deaths= 0.0773%, 103 days

Net increase in active = 1,569

Data: docs.google.com/spreadsheets/d… Image
In my personal spreadsheet (drive.google.com/file/d/11R0xHB…), PH has crossed 19 new cases per 100,000 in the 14-day period from 17.200 last Dec 16, coincidentally the start of Simbang Gabi.
The personally-designed doubling time estimate has declined from 222.88 days-to-double to 210.23 days-to-double. COVID-19 infections are accelerating again.

This is the beginning of a surge, my good reader.

The end of the pandemic is far from sight.
Read 4 tweets
6 Oct
Pulse Asia's survey methodology is typical of a private survey firm's approach to sampling given limited resources but still able to make some measurable conclusions.
I am not an agent of Pulse Asia, but Pulse Asia has technical notes able to be understood by people exposed to survey sampling methods.

pulseasia.ph/ulat-ng-bayan-…
Let break this down:

1st, Pulse Asia's domain of concern in studying to have measurable accuracy with sampling is the four geographic areas: NCR, Luzon w/o NCR (termed Balance Luzon), Visayas, and Mindanao, each of the four alloted 300 samples each, thus 1,200 samples.
Read 8 tweets
5 Oct
How do you round off the following numbers, respectively:

7.5; 6.5
So, look at this:

True percent values in 4 groups:
90.5% + 4.5% + 3.5% + 1.5% = 100%

If Just Round down or Round :
90 + 4 + 3 + 1 = 98. :(

If Just Round to Even:
90 + 4 + 4 + 2 = 100. :D

If Just Round to Odd:
91 + 5 + 3 + 1 = 100 :D

If Just Round Up:
91 + 5 + 4 + 2 = 102 :(
Another:

True percent values in 4 groups:
90.5% + 4.5% + 4.5% + 0.5% = 100%

If Just Round down:
90 + 4 + 4 + 0 = 98. :(

If Just Round to Even:
90 + 4 + 4 + 0 = 98. :(

If Just Round to Odd:
91 + 5 + 5 + 1 = 102 :(

If Just Round Up:
91 + 5 + 5 + 1 = 102 :(
Read 4 tweets
15 Jul
Let me tell you this:

It is difficult to believe the curve of new confirmed cases per day, because It is highly susceptible to the limitations of @DOHgovph in terms of data management and case validation. Look at the curve of cases by onset of symptoms

1/n
Ignore the point the last portion where the line drops, because again, data delays in validation and testing.

As one can see, the cases have seemingly growing at an increasing rate, bar again the last 9 days or so of the data.

Source: drive.google.com/file/d/1M28JzQ…

2/n
Flattening of the curve entails: 1) a basis on a reference curve of no interventions or the worst scenario, and 2) knowledge of the healthcare capacity

3/n
Read 10 tweets

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/month or $30/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!

Follow Us on Twitter!