Twitter | Search | |
Natalie E. Dean, PhD
Assistant Professor of Biostatistics at specializing in emerging infectious diseases and vaccine study design. PhD. Tweets my own.
4,673
Tweets
561
Following
77,577
Followers
Tweets
Natalie E. Dean, PhD retweeted
Emily Dhurandhar, Ph.D. 20h
from how experts are disregarded: scientists have not done enough to earn the public’s trust. If we are to remain relevant, we must do better training in science communication, encouraging it instead of shunning it, and being more approachable
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Bertha Hidalgo, PhD, MPH 2h
Thank you for adding this, ! While 0% positive is ideal, 5% for at least 14 days (2 weeks) is a threshold recommended by CDC and WHO, assuming comprehensive testing.
Reply Retweet Like
Natalie E. Dean, PhD retweeted
The COVID Tracking Project 2h
Replying to @COVID19Tracking
The 7-day average for cases seems likely to rise. There were storm-related drops in testing and cases, which showed up in the numbers from the 3th-6th. Those numbers will be replaced with regular reporting. We might be seeing the very beginning of that today in the South.
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Natalie Exum 6h
Every parent sending their kid back to college should be asking if dormitories have been tested for . Exposure via hot water in showers will be likely route of transmission.
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Ranu Dhillon 5h
1/ Frankly the pushback hasn't only been regulatory; many 'experts' haven't agreed or understood this approach For 6 yrs, & I have written extensively about it as a go-to strategy for uncontrolled epidemics & for Covid in Jan, Feb & March
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Adam Kucharski 6h
This is an error I've seen quite a bit - if a population has pre-existing immunity, the early estimated reproduction number *already reflects that immunity*. So if you then use that number in a herd immunity calculation, you're effectively double counting.
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Madhu Pai 12h
After teaching two epidemiology courses for journalists in two continents, my summary piece : Journalists Need To Get It Right: Epidemiology Training Can Help
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Lindsay Wiley 8h
Replying to @joshmich
I’ve been railing against the term “lockdown” for months. If you mean closing indoor gathering places like bars & indoor dining & prohibiting large gatherings, say that. If you mean a shelter in place order, say that.
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Mash 8️⃣6️⃣ DAYS TILL 🗳️ 7h
This is so important when talking about models. Models aren't just data. They rely on assumptions and those assumptions often mask noisy data. Making sense of all the data coming in is hard. It's best to be humble. Reality is messy.
Reply Retweet Like
Natalie E. Dean, PhD 7h
Replying to @youyanggu @trvrb
I also see a lot of value in “on the ground” anecdotal evidence. Mobility data, while instructive, has not been validated to measure “safe activities.” Mask wearing. Bar closures or curfews. This can’t easily enter into models, but it can help us interpret results.
Reply Retweet Like
Natalie E. Dean, PhD 7h
Replying to @youyanggu @trvrb
As someone in the “too high” camp, I think the conversation is healthy and useful. Models are driven by data but require a lot of assumptions. The data are noisy! Others have used their own methodology to arrive at lower estimates. Diversity of approaches is a strength.
Reply Retweet Like
Natalie E. Dean, PhD 8h
Yes, statewide numbers are speculative since data are noisy. I think Trevor’s original point was more about acquired immunity “taking the edge off” a mitigated outbreak. For that, local dynamics seem more relevant than state numbers. Still very susceptible here in Gainesville.
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Walt Mossberg 22h
Alternative to the USPS for voting by mail. This is in Maryland.
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Larry Brilliant MD, MPH Aug 8
Testing sewage for polio virus was been an important way to confirm areas polio-free, or to point to rising risk. Some places like Boston are doing the same for virus fragments from human feces. It is an inexpensive simple monitoring tool that should be used everywhere
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Hakeem Jefferson Aug 8
For comparison, Stanford IRB wouldn't let me randomize feedback to white people about their racial attitudes.
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Christian Althaus Aug 8
New seroprevalence-based estimates of the infection fatality ratio (IFR) in Spain (overall: 0.82%-1.07%) confirm our earlier estimate at (overall: 1.0%) that was based on statistical modeling.
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Dr. Theresa Chapple Aug 7
I see an opportunity to uplift and for academics to support the work of and . Partner, not lead.
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Dr. Tom Frieden Aug 8
Plain talk, and common sense, from on testing. Don't get the Covid test back quickly, don't get paid. It's not a complicated idea. It just takes courage.
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Maria Garay Aug 7
Dropping this here ✨just in case✨ it is useful for someone
Reply Retweet Like
Natalie E. Dean, PhD retweeted
Daniela Witten Aug 8
This is the content twitter needs right now
Reply Retweet Like