Discover and read the best of Twitter Threads about #EpiTwitter

Most recents (24)

[FR, ENG below] Hello Twitter !
Je suis Louis, bientôt diplômé en #écologie et #épidémiologie des maladies transmissibles, et à ce titre, je recherche une #thèse pour la rentrée 2020 !
Mon CV est juste là 👇
Les RT sont évidemment très appréciés !
Je souhaiterais Ă©tudier les 1/9
maladies infectieuses humaines liées à l'urbanisation et aux contextes urbains, soit dues aux concentrations de populations, soit aux avancées technologiques (maladies dues aux contaminations de réseaux d'eau et leur épidémio, par exemple). Les maladies sexuellement 2/9
transmissibles m'intéressent beaucoup, et je suis également très attiré par les aspects de spatialisation/SIG, et la sociologie des maladies transmissibles (comportement, perception des risques, ...). Mes agents pathogènes de prédilection sont les parasites, les bactéries et 3/9
Read 9 tweets
A lot of people have requested us to measure the case fatality rate (CFR) and disaggegrate by age group in our #COVID19PH dashboard (…)

Let me explain in this thread why we don't #epitwitter
1/ The crude way of measuring CFR is # of deaths / # of cases. CFR indicates the severity of the disease. See ref from Gordis Epidemiology below

So it should be easy to calculate CFR, right, and specify by age, given that our dashboard gives a breakdown of cases & deaths by age?
2/ Not so fast. Let's talk about the denominator first. The denominator is limited significantly by the Philippines' testing capacity - still hovering around 1K+ samples per day (and remember, positive cases are tested more than once until they recover!)

See progress as of Apr 4
Read 12 tweets
Thanks @Virusnerdette and @hrogier. Yes, Vietnam has had a number of experiences in this area, and #epitwitter readers should remember that H5N1 outbreaks in poultry are regular occurrences, so the public health system is set up for this type of response. 1/
In 2009, when a new subtype/variant of influenza was announced to be infecting humans (Friday April 24 2009) Rogier and I were at a happy hour at O'Brien's in HCMC (early evening in HCMC, daytime/morning in Geneva/NY). The public health response got moving right away ... 2/
.. even though this was an emerging epidemic in Mexico, literally a half-a-globe away from Vietnam. I just checked my files, and the earliest quarantine date I can find is May 10 2009 (tested negative) for a passenger landing at HCMC airport, but I am pretty sure that the ... 3/
Read 8 tweets
The evidence for and against cloth mask use in the general public. 🧵
#Masks4All #epitwitter #AcademicTwitter #AcademicChatter
@mollywood @kairyssdal @Marketplace
The strongest scientific evidence comes from systematic reviews, which look at cumulative results from many studies, followed by randomized controlled trials, then observational studies, laboratory experiments, and then case studies.
Read 18 tweets
Articles like these coming out of India scare me.

First, there are only *two* full SARS-CoV-2 genomes from India, both from Kerala. Unless there's additional data that hasn't been published, I'm not sure how one can say anything about the strains circulating locally. 1/
No sources have been cited for any of the claims being made. 2/
Dr. Reddy is a gastroenterologist commenting on virus genomics, epidemiology, and public health, which he is free to do, but he is making recommendations about country-wide interventions. This is scary. 3/
Read 7 tweets
Physical distancing is getting better, but some still seem not to be following. Given that, here are some thoughts about the difference between individual risk vs group risk and why it matters for an infectious disease outbreak. #covid #epitwitter 1/6
Many people I encounter still seem to be doing the personal math on risk. They reason, not many (known) cases in my area, so the risk to me is currently low. I won’t engage in high risk behaviors like being in large groups, but small groups is fine. #covid #epitwitter 2/6
And personal risk may in fact be low in some areas. But this is the risk to one person, not the risk to all of us. The probability of winning the lottery is miniscule. But the probability that someone will win the lottery is high. That’s our risk as a group #covid #epitwitter 3/6
Read 7 tweets

Dismal prediction: We'll see increasing rhetoric out of the White House that seeks to divide the states by promoting quarantines + movement restrictions, and directing blame to current outbreak centers.

Governors are too strong when united in their frustration.
Here's a dire example. FL gov. DeSantis is setting up checkpoints at the FL border to impose quarantine on NY and LA residents.…

Meanwhile, in FL itself, this is what he's doing: Nothing.

Looks like a blame game to me.

And right on cue:
Read 5 tweets
(Thread) Attention #MedTwitter and #EpiTwitter - I am going to try crowdsourcing some advice on a publication venue for a study that my colleague @KHouseknecht and I strongly suspect could have profound impacts on #COVID19 patient outcomes. These are preclinical model data. 1/
Here is the upshot: these (in vivo) data demonstrate profound immune dysregulation occurring as a drug side effect. The drug in question is widely prescribed, notably in vulnerable populations including older adults. These are the very same patients most at risk for... 2/
...complications and death from #COVID19. Related (in vivo) work also showed direct induction or acceleration of cardiac damage from the same drug class. Given increasing reports suggesting both inflammatory sepsis and cardiac arrest as causes of death during #COVID19, and 3/
Read 8 tweets
1/11 There has been quite a bit of Twitter discussion of the broader health effects of the current crisis, much of which refers to my work on the health effects of macroeconomic conditions. This stream provides some thoughts on that. #EconTwitter #epitwitter #COVID2019
2/ First & foremost, the honest answer is that we don’t know what will happen. The current severe economic downturn is caused by a health shock, which has not been the case in over 100 years. For instance, the “Great Recession” largely occurred due to a financial shock.
3/ Consider the extreme case. Some predictions were that in the “do-nothing” scenario, there could have been over 2 million COVID-19 deaths in the U.S. this year. That’s approximately equal to the total number of deaths last year. At the other extreme, some countries appear to
Read 11 tweets
We just returned from an incredible, but poorly timed trip to Malaysia. I won't bore you with the travel delay disasters, but 2.5 days later we are home and all are healthy as of now. We are completely physically and emotionally exhausted. #flattenthecurve #epitwitter 1/12
Some stark contrasts I noticed between Malaysia and the US. In Malaysia, we went to 4 cities including Kuala Lumpur, Kota Kinabulu, Lahad Datu, and Sandakan. I don't consider Malaysia to be a 1st world country and given the levels of poverty we saw I think that’s right. 2/12
However, Malaysia's COVID-19 enforcement was evident everywhere, even in the remote rainforest where we were staying. Mobile carriers sent a message about social distancing every day. The carrier ID at the top of my phone that normally says Sprint said "Stay Home" instead. 3/12
Read 12 tweets
I'm happy to report that I am now Dr. Mitro 🤓 and I've come to Twitter to share a few things I learned while doing my #distancedefense today. My first ever thread!

#AcademicTwitter #epitwitter #phdlife #phdchat @HarvardEpi

First, some @zoom_us tips for during the defense. To prevent Zoom from being seriously distracting, I suggest:

(1) Silence the chimes when people enter and exit…

(2) Mute chat notifications during the defense…

More Zoom ideas, re: videos:

(3) When making your slides, leave space in the upper right corner for video of your face to appear without blocking content

(4) Ask non-committee audience members to turn off their videos

Read 9 tweets
As we think about COVID-19, a basic tweetorial about R0 (basic reproduction (or reproductive) number) and Re (effective reproduction number).

R0 = average # of people who will be infected by an individual in an entirely susceptible population. #EpiTwitter #FlattenTheCurve 1/n
R0 is a function of 3 key factors:
c = Contact rate: how often people are close to other people
p = Probability of transmission given contact with infectious person: how infectious the virus is
D = Duration the virus can be transmitted

R0 = c*p*D
Re (effective reproduction #) adds in the proportion of people who are susceptible (s):

Re = R0 * s = c*p*D*s

If everyone in the population is susceptible, s=1, & Re=R0
Read 7 tweets
1/12 How does #Demography impact #COVID19 deaths? In new pre-print, we illustrate how older population age structure can interact with high mortality rates at older ages to produce a large # of fatalities, as in Italy.… #poptwitter #epitwitter
2/12 #COVID19 fatalities are hitting older age groups hard. Case fatality rates for 80-90 currently 17.5% in Italy. While these numbers will hopefully be overestimates, the burden on older ages groups is frighteningly high.
3/12 With current concentration of deaths at older ages, COVID-19 deaths will hit older countries hard (Italy has 23% of population >65). Figure 1 compares Italy to South Korea (top) and Nigeria to Brazil (bottom) – two countries similar in size but different age distributions.
Read 12 tweets
I’m commonly asked “Why don’t we just test everyone for #COVID2019 ?”

I’ll explain below in this #epitwitter #medtwitter #tweetorial #thread –

Please share/RT to help spread real understanding in the midst of all the panic and guesswork!

The accuracy of medical tests is typically described using two terms:

Sensitivity (what percentage of ppl WITH the disease will test POSITIVE?)

Specificity (what percentage of ppl WITHOUT the disease will test NEGATIVE?)

… but what the patient wants to know is measured by two DIFFERENT terms:

Positive Predictive Value (PPV- if you test POSITIVE, do you actually HAVE the disease?)

Negative Predictive Value (NPV- if you test NEGATIVE, do you actually NOT HAVE the disease?)

Read 10 tweets
The death rate is NOT a biological constant. It is not fixed to a pathogen. It reflects the severity of the disease in a particular context, at a particular time, in a particular population. The probability that one dies from a disease is not only dependent on the disease itself.
It also reflects the social AND individual response to it: the level, quality, AND timing of treatment / care they receive, and the ability of the given individual to recover from it. Remember the epidemiological triangle? #epitwitter #scicomm
There are a number of rates, risk, ratios important in an #outbreak. The fatality rates can decrease or increase over time, and that it can vary by location and by the characteristics of the infected population (age, sex, pre-existing conditions). #scicomm #COVID2019 #COVID19
Read 4 tweets
I’ve spent a few hours reviewing the literature on communion and infectious disease (not too hard when there’s so little out there!). I have significant concerns about how the studies are being applied to #COVID19. Buckle up for a long thread. #epitwitter #coronavirus #InfDz 1/n
First, thank you to @pathatch for the article that let me find others through citation tracking! Second, expect some #OpenAccess access ranting. 2/n
@pathatch Article #1 “The effects of receiving Holy Communion on health.” Lisa F Wolf. Study: 681 Volunteers filled out weekly surveys for 10 weeks on illness and if they took communion. No significant difference found in the rates of illness between communion and not. 3/n
Read 20 tweets
(2) First, some #coronavirus tips for the general public:
(3) To find information on #coronavirus if you are a member of the public, check out this #COVID19 page from @CDCgov:…
Read 22 tweets
The all-cause mortality result in the NELSON #lungcancer screening trial has generated major debate.

How do we interpret this? Let's start with what we expected to observe.

@NEJM #lcsm #lungcancerscreening #EpiTwitter
NELSON showed a 24% statistically significant reduction in lung cancer mortality with low-dose CT screening in men.

If the intervention (screening) does not lead to fatal harms, then we expect the rate of death from other (non-lung-cancer) causes to be the same in both arms.
Now, a large reduction in lung cancer mortality would translate to a small reduction in all-cause mortality, since lung cancer only causes a portion of deaths (see NLST).

NELSON did not have enough power to test whether the hazard ratio for all-cause mortality differed from 1.
Read 6 tweets
Quick thread on the recent Dutch KOMPAS trial in @JAMAInternalMed… 1/
@JAMAInternalMed Thanks to @brzoskos for tagging me when this came out last week
TL;DR version: doing nothing is fine when you are doing CT scans in peeps with GFR > 30
Isn’t this similar to AMACING? Yes and No, See thread: 2/
KOMPAS was a non-inferiority RCT, funded by the Achmea Healthcare Foundation
Compared 250 mL of 1.4% sodium bicarbonate administered in a 1-hour infusion before CT vs doing *nothing* 3/
Read 32 tweets
Okay, since @KevinMKruse now seems to have given his imprimatur to this piece in the @washingtonpost today about #nCoV2019 & past plagues, it seems time for a mini-thread about #medhist & hot takes. #epitwitter: you might want to listen in on this, as it effects you, too.
@KevinMKruse @washingtonpost The @washingtonpost piece by Eisenberg et al. makes 3 main points: 1) that the #BlackDeath (the #plague pandemic usually dated to the mid-14thC) is the most commonly invoked analogy when people think of epidemics; 2) that not all "plague" epidemics/pandemics were alike; and 3) ..
@KevinMKruse @washingtonpost ... that there's an "outbreak narrative" that "we replay .. as a script with each new outbreak — whether real or fictional." First, some background on what #histmed (History of Medicine) is: it's probably pretty much as you would assume from its name. The field of history that ..
Read 17 tweets
Have you ever calculated the sample size for an #abtest and come up with a sample size that is bigger than you can ever practically get?

Does this mean you shouldn't run the test?


A paper thread for #MarketingAcad #EconTwitter #Measure #epitwitter 1/17
@marketsensi and I thought about this and came to the conclusion that the standard hypothesis test used to analyze A/B tests doesn’t fit well with the marketing problem that we are usually trying to solve.
Hypothesis tests are used by academics who want to find small effects with high confidence, but in marketing we care about the big effects. Big effects are where the profit comes from!
Read 17 tweets
Hello #epitwitter and #statstwitter,
There was a recent discussion about the Hosmer-Lemeshow goodness of fit test. I thought it would be interesting to talk to Dr. Lemeshow (who is not on twitter) about his thoughts on the test. This thread has some highlights from our chat. 1/n
In the late 70’s, Hosmer and Lemeshow were struggling with the question “How do you know that the probabilities produced by logistic regression models reflect reality?” This was the motivation for developing the Hosmer-Lemeshow goodness of fit test. 2/n
As with a linear regression, we can’t only look at the estimates from the model to know if the fit is good. For linear regression we can look at a plot (e.g., residual plot) to assess model fit. 3/n
Read 15 tweets
#Epitwitter Fun with Numbers:

It’s possible to put bounds on the case-fatality rate (CFR) for this outbreak with the available data.

A quick #tweetorial on the partial identification of absolute risk bounds!

According to the quoted tweet we have this data:

31 cases with known outcome
6 of whom died

227 with an unknown outcome

First, the CFR among known data: 6/31 = 19%

Next, we include unknown outcomes, taking them to their logical extremes (all live or all die).

The lower risk bounds is calculated under the “all live” extreme:

6/(31+227) = 2.3%

Given the available data, this is the lowest possible CFR (i.e., if no new cases arise).

Read 7 tweets
DAGWOOD abstract submitted to SER! But I also wanted to share it with y'all, because it's the single most surprisingle brain-warpingly difficult thing I think I've ever worked on, and I am super proud of what we've got so far.



DAGs are a way to organize and display causal inference models and their assumptions.

But where in the DAG are those assumptions?

They are hidden in the negative space between nodes, taking the idea of being transparent about our assumptions very literally.

For example, a causal model typically assumes a sharp (or negligible) causal null for any and all missing confounders. Those confounders are hidden somewhere in the ocean, but we can't see them in a typical DAG.

Hard to think critically about invisible things.

Read 20 tweets

Related hashtags

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/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!