BAYES' THEOREM: The basic reason we get so many false positives to COVID19. The disease is so rare that the number of false positives greatly outnumbers the people who truly have the disease: THE MATHS:
https://t.co/oLHyxYJW9H
https://t.co/29FNwq0Qw2
"The national statistician has downgraded its estimate of coronavirus in England on October 17 to just 4.89 people per 10,000." or ~ 0.05%. That means that only 1 in 2000 people may be carrying SARS related viral RNA fragments which could be 2 months old. https://t.co/XqpNaY6BzQ
— Robin Monotti FRSA MA BSc (@robinmonotti) December 8, 2020
https://t.co/rthjPRJWeB
ITALY: CONFIRMED BY ITALIAN HEALTH SERVICE: False positives to Covid19 test as diagnosis are 95%. Legal cases started against testing under charges of fraud to procure public funding, false alarm, ideological false, and manslaughter. pic.twitter.com/C9b7BbzdKa
— Robin Monotti FRSA MA BSc (@robinmonotti) November 25, 2020
#BAYESTHEOREM MEDICAL MASS TESTING CALCULATOR:
— Robin Monotti FRSA MA BSc (@robinmonotti) December 9, 2020
Try it yourself to understand how many false positives you get by changing minor variables: \U0001f447https://t.co/7wVMvrpgAW pic.twitter.com/PHbweWK1TK
https://t.co/kFnQVoCspb
LATERAL FLOW: False positive rate of "0.4% with a sensitivity of 58% and specificity of 99.6%, would mean that 100\u2009000 people being tested would find 630 positives\u2014of which only 230 would actually have covid-19, while 400 would be false positives.
— Robin Monotti FRSA MA BSc (@robinmonotti) November 17, 2020
https://t.co/8hsZ1hNjD7
Official estimates from mass testing in England (including asymptomatic) puts Covid19 "infections" at 0.9% https://t.co/2ljzi9YfKN
— Robin Monotti FRSA MA BSc (@robinmonotti) December 11, 2020
@lucyfrazermp 20/11:
— Edmund Fordham (@EdmundFordham) November 28, 2020
2. Apparently @MattHancock tells HoC 17/09 how ONS \u201cadjusts for False Positives\u201d. Looked it up.@DesmondSwayne asks;@MattHancock doesn\u2019t answer.
Obfuscates with \u201crigorous Bayesian mathematics\u201d
\u201cOne of his academics" will \u201ctake him through it\u201d
(thread) pic.twitter.com/42YO9vaioy
#BAYESTHEOREM @ Cambridge University. 0.4% of 262 students came back as positive after the first "test". All came back as negative after the second. Government only tests once. ONS would say there is 0.4% prevalence instead it's 0%. pic.twitter.com/zeAQAAOeRN
— Robin Monotti FRSA MA BSc (@robinmonotti) December 13, 2020
https://t.co/pZcFlMBKEZ
"I am very happy for one of my academics to take him through the rigorous Bayesian mathematics, which I am sure will help to elucidate the debate on this matter still further." @MattHancock to @DesmondSwaynehttps://t.co/pZcFlMBKEZ
— Robin Monotti FRSA MA BSc (@robinmonotti) December 15, 2020
https://t.co/aidVGWOVqH
Numerical details aside, the use of Bayes's theorem and the principle described are valid. Taught in epidemiology courses.
— \u05e4\u05e8\u05d5\u05e4' \u05d0\u05d9\u05d9\u05dc \u05e9\u05d7\u05e8 (@prof_shahar) December 16, 2020
Extreme example. If disease prevalence is zero, predictive value of positive test is zero. Every positive is false positive.
A similar graph in the thread pic.twitter.com/ddvBe36OmU
\u26a0\ufe0fWHO WARNING ON BAYES THEOREM & TESTING \u26a0\ufe0f
— Robin Monotti FRSA MA BSc (@robinmonotti) December 16, 2020
"Healthcare providers are encouraged to take into consideration testing results along with clinical signs and symptoms, confirmed status of any contacts"https://t.co/GkRJzdn70b pic.twitter.com/jXPQDqqnVE
More from Robin Monotti FRSA MA BSc
I have already left both Facebook & Instagram. We need to keep agile.
They will try to ban Parler, blaming it for Capitol theatre. I think Telegram may survive as it's not based in the
Yes Telegram owner @durov received & accepted what effectively is an award, not a partnership: the Young Global Leaders membership of the World Economic Forum in 2017. Does this mean he passes users info on? I don't think so.
This is what @Snowden had to say about @durov. Since then Telegram introduced the option of end to end encrypted chats not saved in Telegram servers. These private chats cannot be forwarded, and none of the participants can capture screenshots of
Trust us not to turn over data. Trust us not to read your messages. Trust us not to close your channel. Maybe @Durov is an angel. I hope so! But angels have fallen before. Telegram should have been working to make channels decentralized\u2014meaning outside their control\u2014for years.
— Edward Snowden (@Snowden) December 30, 2017
On the 19th March 2020 the WHO released this guidance intended for healthcare workers (HCWs), healthcare managers and IPC teams at the facility level & at national and district/provincial level:https://t.co/C4aV2BnMPj pic.twitter.com/tCk1EyLskV
— Robin Monotti (@robinmonotti) December 21, 2020
It clearly does indicate both the risks of bacterial infection & to prescribe broad spectrum antibiotics as part of treatment:
"Collect blood cultures for bacteria that cause pneumonia and sepsis, ideally before antimicrobial therapy. DO NOT
delay antimicrobial therapy"
"6. Management of severe COVID-19: treatment of co-infections
Give empiric antimicrobials [broad spectrum antibiotics] to treat all likely pathogens causing SARI and sepsis as soon as possible, within 1 hour
of initial assessment for patients with sepsis."
"Empiric antibiotic treatment should be based on the clinical diagnosis (community-acquired
pneumonia, health care-associated pneumonia [if infection was acquired in health care setting] or sepsis), local epidemiology &
susceptibility data, and national treatment guidelines"
"When there is ongoing local circulation of seasonal influenza, empiric therapy with a neuraminidase inhibitor [anti-viral influenza drugs] should
be considered for the treatment for patients with influenza or at risk for severe disease."
More from Category c19
/1
https://t.co/qwQS3b2NPq pic.twitter.com/TqW1K3oC60
— Medical Nemesis (@Medical_Nemesis) February 20, 2021
Legends and myths in medicine. Debunked. Not that but few care.
https://t.co/3Hu2D9Un0J
/2
https://t.co/xnZ2EXO1f5
/3
Open access.
Where have we seen this before? 🤔