There is no question we need more rapid #COVID19 testing. The food truck model is a promising approach that advances the cause of rapid screening testing while allowing greater adaptability and providing more safeguards. Let me explain. 1/10 @KatherineJWu…
The debate around rapid screening tests has started to divide into the "at-home testing" proponents and the "impractical, doubtful it'll work" proponents. They key to effective public health innovation is the drive to try new approaches while addressing legitimate concerns. 2/10
1. Start with the "food truck model" of rapid screening test. It's almost as convenient as home. But it can be staffed by health care workers to perform tests and provide guidance. If you don't want to leave your house... you don't need to get tested. 3/10
2. Antigen tests *do not* have to have 100% accuracy provided the public health messaging is clear. Patients often make medical decisions about surgery vs. drugs based on likelihoods alone. Having a health care worker to help interpret a binary test result is very helpful. 4/10
Antigen tests help target increased social distancing to small proportion of popn with a positive antigen tests. And allows more freedom (though not total freedom b/c some false negatives) to the rest of the population. Widespread testing quickly drives case counts down. 5/10
False positives are a friction in the process of lowering transmission. They are not a reason not to roll out rapid screening tests. Everyone is wearing masks and distancing regardless of infection status - we know that's what you have to do for the system to work. 6/10
3. Our understanding from the evidence is that viral load is correlated with contagiousness. The new rapid antigen tests have high agreement with PCR. Yes we need data on antigen testing for asymptomatics. But slamming the brakes on antigen testing doesn't make sense. 7/10
4. You get benefits from rapid screening testing even if not everyone is being tested and even if there is no enforcement of self-isolation of true & false positives. Costs of individuals taking precautions are far outweighed by benefits of lowering prevalence of COVID-19. 8/10
5. Rapid screening tests are less helpful where prevalence is low b/c they'd produce a higher ratio of false to true positives. Focus on rolling these out in high-prevalence locations. A food truck model allows us to target rapid tests in real-time where needed & effective. 9/10
6. A "food truck model" program collects data that can help course correct if necessary. And allows for real-time in-person conversations with people being tested to get feedback on messaging. Plus inventory control when tests are limited. 10/10
Tagging some who've engaged in the rapid screening testing debate: @michaelmina_lab @CT_Bergstrom @apoorva_nyc @KatherineJWu @alexismadrigal @yayitsrob @ADPaltiel @RWalensky @jhuber @BenMazer

We need a broad base of support to make rapid screening test policy a reality.
Here's my piece on rapid screening tests from Aug 14. Still I think the clearest explanation for a general audience about how rapid screening test programs work.…

• • •

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

Keep Current with Zoë McLaren, PhD

Zoë McLaren, PhD 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! 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 @ZoeMcLaren

18 Sep
If you're a proponent of rapid screening tests, then you should be extremely excited about wastewater testing for #COVID19. Here's why it's so useful. Thread. 1/5
1. Everybody poops. You have universal population representation in your sample.

2. Data collection is centralized. Sewage systems bring data from every household and workplace into a single location for centralized testing. 2/5
3. Testing requires no behavior change in the population. Literally none.

4. Viral particles are detectable in sewage before people notice symptoms. Wastewater likely also would detect cases slightly sooner that antigen testing. 3/5
Read 6 tweets
2 Aug
Skipping Phase 3 of clinical trials and rushing out a COVID-19 vaccine is a *terrible* idea. It goes against medical ethics, puts people at greater risk and undermines trust in the vaccine approval system. It causes more problems than it tries to solve. Let me explain. 1/10
Today @StevenSalzberg1 claimed Phase 3 was “excessive caution” and advocated for administering millions of doses now. @DrEricDing falsely suggested the bioethics warrants giving it to volunteers before Phase 3 is complete. 2/10
Sample sizes for Phase 1 and 2 of clinical trials don’t have the statistical power to detect all side-effects or determine effectiveness in the general population. They are designed to test for safety and effectiveness with samples of several hundred people. 3/10
Read 11 tweets
31 Jul
@JenniferNuzzo I think you're underestimating the power of the test-positivity metric. Yes, it matters that the "right" people are being tested, but it's hard to game that. People are self-selecting into testing, so it's people who have reason to suspect they're positive.
@JenniferNuzzo And even if you're testing lots of unlikely-positives, that's screening. Also a valuable use of testing resources.

If TAT is long, then your test-positivity will almost surely rise because there is more transmission. It's baked into the metric.
@JenniferNuzzo Lower test positivity is always better (provided not including antibody tests etc.) -- because it means fewer cases and/or more testing (which catches more cases and means lower transmission).
Read 5 tweets
31 Jul
@nataliexdean I've been thinking about this a lot. The trick is to THINK LIKE A VIRUS. Virus is playing high-risk/high-reward strategy - narrow window of infectiousness hoping to hit jackpot of superspreading event. And hoping one of people infected in event creates their own superspreader.
@nataliexdean Waiting for people to get tested (days after superspreader event) then backtracking to find event doesn't make sense. You're looking for events, not cases. So broader screening strategy that asks about attending large-ish events makes sense.
@nataliexdean Find superspread events, screen broadly, find cases, isolate them before they create new superspread events. Timing is key. The delay before people decide to get tested sets you back right away. It's also harm reduction -- if you're doing risky things, better get tested asap.
Read 5 tweets
8 Jul
This thread racked up over 34K retweets. It's an expertly written thread that makes many important points. But its central premise – that incorrectly pooling data is the key problem – is at best misleading and at worst wrong. #epitwitter #statstwitter 1/13
It's an insidious example of armchair epidemiology: “The experts can't (fully) explain what's going on, but I can. It's simple. The real insight is from [waves hands] [insert cool paradox here].” Let me explain why. 2/13
I'm a data scientist, health economist and public health policy professor. My research uses rigorous statistics methods to draw accurate conclusions from non-random-sample data. Much of my work develops data-driven policy for another deadly respiratory disease: #tuberculosis.3/13
Read 14 tweets
27 May
I see no reason to stop ongoing RCTs for #hydroxychloroquine. The @MRMehraMD/@TheLancet comparison suffers from enough statistical bias to mask a beneficial effect of the drug. We need RCT data to know whether #hydroxychloroquine is effective for #COVID19 or not. #EBM THREAD 1/18
Short version: The study isn't an RCT. Is subject to confounding bias. Adjustments (controls/matching) are far from adequate to address bias. Tipping-pt analysis shows bias is strong enough to flip a true beneficial impact of #hydroxychloroquine to negative. #MedicalTwitter 2/18
Let's unpack this: The challenge with non-experimental studies is that you're comparing treatment and control groups that don't resemble each other because doctors determined who got the treatment instead of using a "coin flip" like in an RCT. 3/18
Read 19 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!