Something has to be done about the memory-hole problem. Need to know that outlet X said Y at time T.

#COVID has now sent 1 out of every 150 New York City adults to the hospital since the city's first case was diagnosed nine weeks ago.

Conclusion 2: (a) we need to redesign our living/working spaces & rethink how to provide better, ventilated living/working environment for those who live in deprived & cramped areas; (b) avoid close, sustained contact indoors & in public transport, & maintain personal hygiene.

Addendum: While we have limited data, similar high risk transmission pattern could be seen in other crowded & connected indoor environments such as crowded office spaces, other workplace environment, packed restaurants/cafes, cramped apartment buildings etc.

20/ In conclusion, contact tracing data is crucial to understand real transmission dynamics. Cautionary note: This data & interpretation is based on the available evidence as of May 4th. Our understanding might change based on community testing/lifting lockdown measures. END

19/ Finally, these studies indicate that most transmission is caused by close contact with a symptomatic case, highest risk within first 5d of symptoms. To note: this preprint suggests that most infections are not asymptomatic during infection

18/ Although limited, these studies so far indicate that susceptibility to infection increases with age (highest >60y) and growing evidence suggests children are less susceptible, are infrequently responsible for household transmission, are not the main drivers of this epidemic.

17/ Increased rates of infection seen in enclosed & connected environments is in keeping with high infection rates seen in megacities, deprived areas, shelters. A recent preprint demonstrates that #COVID19 epidemic intensity is strongly shaped by crowding

16/ High infection rates seen in household, friend & family gatherings, transport suggest that closed contacts in congregation is likely the key driver of productive transmission. Casual, short interactions are not the main driver of the epidemic though keep social distancing!

15/ In summary: While the infectious inoculum required for infection is unknown, these studies indicate that close & prolonged contact is required for #COVID19 transmission. The risk is highest in enclosed environments; household, long-term care facilities and public transport.

14/ In a nursing home facility, 23d after the first resident with #COVID19, among 84 residents (76 tested), 48 were positive. This facilty had 64% prevalence of Covid-19 among residents. 50% of residents had no symptoms at the time of testing. https://nejm.org/doi/full/10.1056/NEJMoa2008457… (24/4/20)

9/ Among 31 household transmission clusters, 9.7% (3/31) were identified as having a paediatric index case (none was identified in Singapore), suggesting that children are unlikely to be driving the household transmission. https://medrxiv.org/content/10.1101/2020.03.26.20044826v1… (30/3/20) Preprint

12/ 15 confirmed/probable cases were identified after the index #COVID19 case attended a funeral (3h), shared a meal (2h), birthday party while having mild symptoms, suggesting family gatherings likely plays an important role in transmission https://tinyurl.com/yaswk4hh (17/04/20)

11/ In Singapore, 3 clusters of 28 cases were identified (2 churches, 1 family gathering). In all clusters, transmission accounted for 1 close contact w a symptomatic case, suggesting transmission largely occurs in close contact in congregation. https://thelancet.com/action/showPdf?pii=S1473-3099%2820%2930273-5… (21/4/20)

10/ In a population-based study in Iceland in which 9199 were tested, of the 564 children <10y, 38 (6.7%) tested positive, vs 1183/8635 (13.7%) adolescents and adults tested positive, suggesting lower incidence in children. #COVID19 https://nejm.org/doi/full/10.1056/NEJMoa2006100… (14/4/20)

8/ In a French chalet cluster, 11/15 contacts tested positive (all adult), 75% attack rate. One child (9y) was negative, attended 3 schools & ski class while symptomatic, among 172 (73 tested) contacts, 1 had #COVID19, while 33% had influenza! https://academic.oup.com/cid/advance-article/doi/10.1093/cid/ciaa424/5819060… (11/4/20)

7/ Among 2761 close contacts of 100 confirmed #COVID19 pts in Taiwan, 22 secondary cases were identified, household attack rate was 4.7%, rates were higher in close family, >40y, if exposed within 5d after symptom onset (0 infection after 5d) https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2765641?utm_campaign=articlePDF%26utm_medium%3darticlePDFlink%26utm_source%3darticlePDF%26utm_content%3djamainternmed.2020.2020… (1/5/20)

6/ A symptomatic index #COVID19 case, her husband who subsequently acquired infection and their 350 close contacts were followed up, 43 developed symptoms, none tested positive, suggesting close & prolonged exposure is required for transmission. https://thelancet.com/journals/lancet/article/PIIS0140-6736(20)30607-3/fulltext… (13/3/20)

5/ Among 392 household contacts of 105 index #COVID19 cases, overall household attack rate was 16%, the secondary attack rate was highest in spouse (28%), all adults (17%) and was lower in <18 age group (4%). https://academic.oup.com/cid/advance-article/doi/10.1093/cid/ciaa450/5821281… (17/4/20)

4/ Among 349 #COVID19 cases in 195 clusters, household attack rate was very high (17%), non-household attack rate 7%. Secondary attack rate was lower in <20y (5%) and highest in >60y (18%), suggesting susceptibility increases with age. https://medrxiv.org/content/10.1101/2020.04.11.20056010v1… (15/04/20) Preprint

3/ Based on 1286 close contacts of 319 #COVID19 cases; household and transport contacts had higher risk of transmission (80% of infections caused by 9% of cases), household attack rate of 11.5%, children were as likely to be infected https://thelancet.com/journals/laninf/article/PIIS1473-3099(20)30287-5/fulltext… (27/3/20)

2/ 445 close contacts of 10 #COVID19 cases were followed up, of those 54 (12%) developed symptoms, suggesting secondary attack rate of 0.45%, household attack rate of 10.5%. No other close contacts incl community members, HCWs were positive. https://cdc.gov/mmwr/volumes/69/wr/mm6909e1.htm?s_cid=mm6909e1_w… (6/3/20)

1/ 2147 close contacts of 157 #COVID19 cases were followed up: Overall infection rate was 6%, higher infection rate among friends (22%) and household (18%), and main risk factors include contact in household (13%), transport (11%), dining (7%). http://html.rhhz.net/zhlxbx/028.htm (4/3/20)

A lot of discussion recently about transmission dynamics, most of which are extrapolated from viral loads & estimates. What does contact tracing/community testing data tell us about actual probability of #COVID19 transmission(infection rate), high risk environments/age? [thread]

A tale in two parts:

Did you start a research, publication, review or data tracker at some point? If yes, what has happened to your efforts? Has some other initiative stepped in to fill the gap you saw? Or are there still unmet demands there? @SciBeh would love to hear from you!

RT @Psychonomic_Soc: Starting next week... @siminevazire @STWorg @ceptional @psforscher @farid_anvari @stefanmherzog

RT @DrHannahWhite: The @instituteforgov * has distilled 48 pages into one graphic for you...

As we start to recover from #coronavirus, we all need to play our part by following the rules. This is how to #StayAlert

RT @ce_pickles: @reformthinktank has called for JRS to allow part-time working– i.e. employer pays worked hours, the scheme non-worked hour…

Eurozone heading for its worst GDP contraction on record at 7.75% this year according to @EU_Commission forecasts.

There is now a flood of COVID-19 preprints from the behavioural sciences - these need critical evaluation and discussion! Are you a behavioural scientist? Then please join our community-based discussion platform now, see:

New accessibility feature in @ChromeDevTools: simulate vision deficiencies, including blurred vision & various types of color blindness. Find out how people with vision deficiencies experience your web app, and resolve contrast issues you didn’t even know you had!

I have been corresponding with the authors of the well-known Santa Clara County COVID-19 preprint, and I am alarmed at their sloppy behavior. The confidence interval calculation in their preprint made demonstrable math errors - *not* just questionable methodological choices.

Everyone makes mistakes, but the record must be corrected ASAP. I emailed them on Saturday morning asking them to do so. In the last three days they haven't corrected anything yet, but a subset of them have released a new study without saying how they did the analysis this time.

Given the critically important and time-sensitive policy decisions being made now, if the authors are still pressing their case in the media using possibly incorrect calculations, then I feel I should make my criticism public too.

The errors are not debatable and can be seen in these two screenshots of the supplement: 0.0034, the standard error meant to measure uncertainty about prevalence pi, is not the square root of 0.039, and the variance of a binomial estimate of proportion depends on the sample size.

I can't redo the whole calculation myself because parts were not described anywhere, but I have low confidence that those parts were done correctly; if not, the corrected confidence interval for prevalence in Santa Clara County might well stretch all the way to include zero.

The authors said by email that they used a built-in Stata function and aren't sure themselves how the software used the input weights. I suspect they misapplied that function (too complicated to tweet why) but I don't know Stata well enough to be sure; it seems neither do they.

We are extended our usual monthly #CyberSectionChat to a weekly event for most of April.This will run as a "Meet & Mingle" and an opportunity to socially connect during lockdown and physical distancing. If you want to connect with the community, be here at 7pm (BST) on Thursdays!

But remember, there are 10,000 infections overall. So if we simulate transmission randomly from each infected person accounting for the above variation, then add up the new infections, we'd expect the following range of possibilities: 6/

For coronaviruses, there's evidence that some infectious cases may generate a lot transmission (superspreading events), some very little, and some none. E.g. we looked at this for MERS-CoV (https://eurosurveillance.org/content/10.2807/1560-7917.ES2015.20.25.21167…), building on this study of SARS etc: https://nature.com/articles/nature04153… 2/

But what about a larger epidemic? Say there are currently 10,000 infections in a population (as there may be in many countries now). How many more would we expect these people to infect in the next few days? 4/

In other words, we might have high variation at the *individual level*, but once we have a large number of infections, the *population level* dynamics are relatively much less variable. This is the logic behind most population-based epidemic models (e.g. the SIR model). 7/

Let's assume high variation in transmission at the individual-level – some cases generate lots of infection, but most generate none. If we assume SARS-like potential for superspreading and early COVID-19 transmission (R=2.5), we'd get following pattern at individual level: 5/

Now you might say, "Surely these superspreading events are predictable? Shouldn't we therefore include them in all models, then target these events to bring outbreak under control?" Unfortunately, like many 'obvious' solutions, it's rarely that simple: https://twitter.com/AdamJKucharski/status/1240774378834534400?s=20… 9/

Often SIR-type models will incorporate some randomness in the transmission process to account for this population-level variability during each generation of infection, e.g. https://thelancet.com/journals/laninf/article/PIIS1473-3099(20)30144-4/fulltext… 8/

In other words, it's important to think about age groups and context of interactions for respiratory infections, but if we focus too much on individual-level social behaviour, we may risk adding complexity without necessarily adding more accuracy. 11/

Intuitively, it may seem like individual social interactions might a good predictor of spread. But our analysis of 2009 flu pandemic found that average social behaviour of an age group could capture patterns better than individual-level contacts... https://journals.plos.org/plospathogens/article?id=10.1371/journal.ppat.1004206… 10/

It's tempting to add as much detail as possible to a model, and criticise anything simpler. But the appropriateness of a model will depend on situation you're facing, the evidence you have that a specific process is predictive of risk, and question you're trying to answer. 13/13

Indeed, as this (aptly titled) piece suggests, complex models may be no more reliable than simple ones if they miss key aspects of the biology. Complex models can create the illusion of realism, and make it harder to spot crucial omissions https://pnas.org/content/103/33/12221… 12/

When there are small numbers of cases, this variation is obviously very important. It can influence the risk of a new outbreak (Fig 3: https://thelancet.com/journals/laninf/article/PIIS1473-3099(20)30144-4/fulltext…) and the potential effectiveness of measures like contact tracing/ring vaccination (https://wwwnc.cdc.gov/eid/article/22/1/15-1410_article…) 3/

A common criticism of population-based epidemic models is that they don't account for individual-level variation in transmission (i.e. superspreading events). But how much of a problem is this? 1/

a google doc tracking all app development:Shared research on privacy preserving contact tracing apps [MOVED]This document has moved and been merged with another shared research effort: Unified research on privacy-preserving contact tracing and exposure notification for COVID-19docs.google.com

Our analysis of the social distancing measures implemented in Santiago (up to March 18th) in terms of mobility patterns. In line with observations in other countries, in the initial phase, variations are visible but not drastic. Traffic and patterns are similar to the weekends.

As countries deploy data-hungry contact tracing, we worry about what will happen with this data. Together with colleagues from 7 institutions, we designed a system that hides all personal information from the server. Please read and give comments!

Governments cannot be trusted w/ social network data from Bluetooth. So w/ colleagues from 7 unis, 5 countries, we've built & legally analysed a bluetooth COVID proximity tracing system that works at scale, where the server learns nothing about individuals

The comic explaining #DP3T Bluetooth contact tracing (also explains Apple-Google protocol) from @ncasenmare now available in 1) simplified one-pane v (img), 2) versions in French, German, Spanish, Thai & Finnish on the repo! Wanna translate it too?

The intention to share data after being diagnosed with COVID-19 is crucial for contact tracing apps to work! (Note for armchair epidemiologists: I always said "coronavirus" in the survey to avoid technical jargon https://twitter.com/markhumphries/status/1248214342618628096…) 7/16

This is not surprising, given that samples like this one tend to be more digitally literate than the general population (check @kmmunger's https://osf.io/3ncmk/). If we didn't get a high percentage, Bluetooth usability would be a big problem for contact tracing apps. 3/16

The last question was "Do you agree that healthcare authorities and phone companies must start sharing health, contact and location data to trace infections of Coronavirus?" The most frequent answer is "Neither agree nor disagree", with less than 20% disagreeing. 8/16

Are people willing to use contact tracing mobile phone apps for #COVID19 ? As a personal weekend project, I spent some money I got back from a canceled AirBnB booking in a quick and dirty survey. Results in my first Twitter thread ever. 1/16

For about 100 Euro I got 108 US respondents in SurveyMonkey. Not representative but at least roughly balanced in terms of gender and age. The first question was: "Are you able to turn on and off the Bluetooth connection of your smartphone?". 95% of the people say yes. 2/16

The second question was "Would you install an app on your smartphone to share your Bluetooth connection data to trace Coronavirus infections?" The majority of respondents said "definitely would" or "probably would", but still far from an ample majority. 4/16

The third question was the same but for GPS location data. Here the result is very similar to the case of Bluetooth connections asked before. This is somewhat surprising given that location data is more revealing than anonymized contacts through Bluetooth. 5/16

How bad would be a 50% penetration rate of contact tracing apps? From 50%, the app starts being effective to control an outbreak, but effective penetration should be above 60% to rely on contact tracing apps. See https://arxiv.org/abs/2003.10222 10/16

The fourth question was "If you were diagnosed with Coronavirus, would you allow your information to be shared with phone companies to search for people you might have infected?" For this, the intention is higher, with more than 60% saying that they would or probably would. 6/16

The second is the mandatory use of the app. If anonymity is guaranteed for app users (e.g. what https://pepp-pt.org aims at), this could be an option that does not threaten individual privacy rights. But it's important to have a long public discussion about this before! 13/16

This high frequency in the middle is surprising, given how polarizing this topic can be among experts. It also shows that a small minority of respondents are against the use of the data already owned by tech companies, something that @rcuevasrumin and @acrumin have proposed. 9/16

In any case, please take my armchair sociologist's survey with a big grain of salt. I hope this thread motivates sociologists to run a better-designed representative survey not just in the US (and also using another platform like MT or Qualtrics). 15/16

Note that the penetration in that plot is the effective rate. Not all phones would work, people forget to use the app, Bluetooth has errors, etc. If just the intention in a survey is about 50%, the effective rate would be insufficient given the simulations above. 11/16

I see two solutions to this problem. The first is a massive information and mobilization campaign to increase the adoption of the app. Scientific communication is critical for this. A good early example is this comic: https://ncase.me/contact-tracing/…. 12/16

For more content on the mandatory use idea, check @ccansu's article: https://medium.com/@cansucanca/why-mandatory-privacy-preserving-digital-contact-tracing-is-the-ethical-measure-against-covid-19-a0d143b7c3b6…. 14/16

To sum up, contact tracing is a promising technological approach to control outbreaks of #COVID19, but we need to keep its social component in mind if we want it to be effective! 16/16

Update on the thread! You should totally check this survey by @STWorg, it's exactly what I was talking about: https://stephanlewandowsky.github.io/UKsocialLicence/index.html… Take home message: Ask your colleagues before spending 100 bucks in a toy survey 17/16

Ever wondered about the "impact" of research on programming languages and software engineering? Political decisions affecting hundreds of millions are being taken based on thousands of lines of 13+ years old C code that allegedly nobody understands anymore. #COVID19 #cs

"back to the future", not "back to normal"I think that we should also see the current situation as a (tragic) opportunity to not go "back to normal" in all respects, rather to go "back to the future", that is: Trying to sustain certain changes that are beneficial to humanity on a large scale (e.g., combating climate change). Importantly, by that I don't mean indefinitely prolonging government interventions, but rather simply building on a changed mindset among many people.E.g., the pandemic has shown that many (most?) meetings can be meaningfully replaced by videoconferencing (and thus a lot of travel and travel costs can be avoided), that home office can work (in companies and institutions that previously claimed or acted otherwise) and many more things (online scientific conferences etc.).The pandemic has, for a better or worse, created a unique situation were the usual objections and obstacles to change on a large scale (e.g., coordination) do not have as much force (e.g., massive, worldwide changes with respect to many things).What can behavioral science contribute to this discussion?

“It’s quite generational, with younger people finding it more difficult than older people.” ⁦@BobbyDuffyKings⁩ spoke to ⁦@DAaronovitch⁩ on BBC Radio 4’s The Briefing Room about who is struggling most under the UK government’s lockdown measures

That's basically the argument of my piece, though put much better and more succinctly.

Well in this case (a) lots of psychologists have said absurd things; (b) some have been pronouncing on the UK government's overall pandemic strategy; AND (c) I'm sceptical our discipline has much to contribute to this in the first place. The latter isn't true for immunology.

Thing is, whether fear or complacency is a "bias" hinges on whether there actually is a medical threat, which is not in the realm of psychology. Research on heuristic effects is very robust. But to call heuristics a "bias" you need proof that they lead to an untrue conclusion.

I was thinking something similar while reading: at least some of the cited psychologists made statements about immunology/virology/etc. and, well, no surprise they don't know about it

New study: fake news only makes up a tiny bit of our media consumption. Great! But that totally misses the point: micro-targeted fake news only needs to convince a tiny minority of the population to disrupt an election. Misinformation also kills people, literally. Just sayingCitar TweetScience Advances@ScienceAdvances · 5 abr.Although “fake news” has often been cited as a pervasive threat to democracy since the 2016 election, a new study finds that deliberately false or misleading info disguised as legitimate news makes up only a tiny fraction of Americans’ information diets. https://fcld.ly/v9nas8g

Just wanted to note that the paper states this in the intro: “.. even if its prevalence is low relative to other types of content, fake news could be important either because it is disproportionately impactful or because it is concentrated on small subpopulations”

Yes, but we don't know if micro-targeting is effective at all. Very little evidence do far.

I would love to see far more research on disinformation and COVID-19 exploitation on platforms other than Twitter. We know YouTube, Reddit, the alt-tech platforms are hotspots. To my knowledge, we know very little about emerging platforms Tik Tok at all.

Most journals in the biomedical and behavioral sciences require that all co-authors contribute to the writing "or critical revision" of the paper, which is unwieldy when there are dozens or hundreds of co-authors.

All should be recognized for this in a way that our norms will allow recognition for when applying for grants and jobs. Unfortunately, nobody seems comfortable with putting papers on their CV for which they are mentioned only in the acknowledgments.

Large numbers of researchers are coming together to work on new projects, because of COVID-19, for example with @PsySciAcc. But journals' authorship "guidelines" (rules) deter these large-scale collaborations.

Moreover, ICMJE guidelines require that all authors are "accountable for all aspects of the work", which is sometimes unreasonable. To me, the overarching principle is that papers should say who did what, including everyone who contributed substantially to the paper.

Also, the names in acknowledgments don't get encoded in databases like Web of Science. Therefore we have to use the author list. But the word "author" is closely associated with "writer", which is one reason that authorship guidelines say all should contribute to the writing.

Yet the authorship guidelines of other fields, like chemistry, don't require writing contributions. McNutt et al. in PNAS, without explaining fully what they were doing, recommended eliminating the writing requirement for biomedicine, as I explain here https://mdpi.com/2304-6775/7/3/48/htm… .

In addition they recommend the adoption of CRediT, which is a machine-readable taxonomy for indicating which area of the project each author contributed to. Hundreds of journals are now adopting CRediT (and some, like @PLOS, adopted it a few years ago).

Most, however (including @PLOS), have kept the ICMJE writing contribution requirement. I think we should rewrite authorship guidelines from scratch, as contributorship guidelines. But no one has written such a thing, to my knowledge.

In the meantime, it may be a good idea for journals to adopt the McNutt guidelines (https://pnas.org/content/115/11/2557.short…), who also sneakily relaxed the accountability guideline, or go further afield, like to the American Chemical Society, which requires only that

an author must make “significant scientific contributions to the work”. Thanks to Ulrike Hahn @SciBeh for raising this issue in the COVID-19 context, prompting this thread! BTW, here's my glam-mag piece on contributorship

For those interested in using CRediT, @mrtn_kvcs & @BalazsAczel have a tool we will be announcing soon designed to help collaborators (esp. large groups) to agree on contributions early in the project rather than at the end, which breeds disagreement. For now, DM me for details.

You need to know this background because it helps with search terms. For example, once you know that SARS, MERS and COVID-19 are respiratory syndromes caused by coronaviruses, you can use this to search for relevant psychology / mental health research.

An example: here's the PubMed search I posted the other day referencing relevant risk perception research https://tinyurl.com/tqdgcr3 Check the search box

OK, back to searching: here's a PubMed search that pulls up psychology and mental health research for SARS, MERS and COVID-19 https://tinyurl.com/tcjlsa7 Note, I use http://tinyurl.com to link to complex PubMed searches because Twitter mangles the URL.

A brief guide for psychologists wanting to find research on the role of psychology relevant to COVID-19. You need to search for studies in the same way you search for studies normally. However, some pointers to sites and key words might be useful...

http://PubMed.gov is your co-pilot. If you're not familiar with it, it's the database of medical research. Most (but not all) mental health/psychology research is on there. Google Scholar complements PubMed well (full text search, broader scope), but we'll focus on PubMed.

Firstly, you need to familiarise yourself with the area. Difference between pandemic, epidemic and outbreak? Epidemic diseases? Flu variants by virus name? Difference between pathogen / disease name? Lots are 'neglected tropical diseases' and so might be unfamiliar. Look them up.

The more you read, the more you'll know. e.g. Médecins Sans Frontières have *years* of experience deploying psychologists to support patients and staff in medical isolation for dangerous infectious disease who interact with staff in full PPE https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=medecins+sans+frontieres+ebola+marburg+psychologist&btnG=…

Note: this is a specialist area like any other. Don't expect to 'get it in 30 minutes'. Be aware of how your limited knowledge might affect your conclusions. Assume you are missing something. Unless you work in this area (I don't) you probably are.

Note that the search terms 'epidemic' and 'epidemics' often bring up a lot of unrelated studies because they're used to describe the 'obesity epidemic' and so on. They may work better when paired with 'infection' or 'infectious', depending on the context of your search.

Including lots of specific disease / pathogen names can be useful because a lot of research is disease specific (e.g. 'psychological responses to the MERS outbreak'). Take home lesson: effective searching is a research method in itself and requires refinement over time.

Here, I used the search term "(psycholog* [ti] OR mental [ti])" as good base terms (they mean: show me every article that has a variant on the word stem 'psycholog' in the title or has 'mental' in the title) and then I specify a bunch of disease related-terms.

Also, be aware of related but non-specific research. Lots of intensive care psychologists have been working for years on psychological management of ICU patients and post-ICU trauma (hats off @dwadepsych - currently flat out reshaping their service for COVID-19)

The search term "next pandemic" is good for bringing up papers on preparations for unknown future pandemics.

Here, I've included 'epidemic' and 'epidemics' - because few studies discuss risk perception with regard to the obesity epidemic - it's quite infectious disease specific, which makes them useful in this context.

But I also includes search terms relating to SARS, Ebola, malaria, H1N1 to because lots of studies just include disease-specific terms. You could these a lot. Also, learn how to refine your search with operators (AND, OR, NOT, quoting, title only etc).

Lots of psychologists have been working for years on health messaging and behaviour change around infectious diseases.

We have in-country experience. Lots of us at the Maudsley were involved in psych support for Ebola staff. This was a major project (co-ordinated by Idit Albert, Al Beck, Elaine Hunter if I remember correctly). @charlewcole did his MSc on it! So junior folk with experience too.

Using behavioural science to help fight the coronavirus https://esri.ie/pubs/WP656.pdf And so on. There is lots of work. There are people who specialise in this. It is not an 'entirely unprecedented situation'.

Lots of relevant reviews and systematic reviews already. How Should Clinicians Integrate Mental Health Into Epidemic Responses? https://ncbi.nlm.nih.gov/pubmed/31958385 A systematic review of mental health programs among populations affected by the Ebola virus disease https://ncbi.nlm.nih.gov/pubmed/32087433

Finally, is the journal article locked and you have no access? Use @Sci_Hub - yes, I know but priorities.

New research is coming out all the time on the pre-print servers http://medrxiv.org and http://psyarxiv.org Google Scholar indexes these sites, PubMed doesn't. However, these are not peer-reviewed papers, so be aware of the quality.

“proper science without the drag” – Move to the medical model of journal review: “Yes/No” decision. We suggest the temporary adoption of this model for crisis-relevant material by journals. [happening already, but potentially even better models: @Meta_psy and @F1000Research?]

New avenues for post-publication critique: Waiting for subsequent articles too slow, social media helpful but diffuse. New formats of immediate publication critique required, so field can learn from, and analyse, research from 1st submission. Journals should add comments.

The behavioural science community must adapt to produce necessary new science, inform policy decisions, and support wider society. What changes should we make? discussion proposals here: https://psyarxiv.com/hsxdk. Main messages are:Crisis knowledge management: Reconfiguring the behavioural science community for rapid responding...The present crisis demands an all-out response if it is to be mastered with minimal damage. This means we, as the behavioural science community, need to think about how we can adapt to best support...psyarxiv.com

We need to shift our sense of what theoretical disagreements matter: focus on commonality! Many “rival theories” will say basically the same thing on key matters, focus on that for the benefit of other researchers and policy makers.

Ensure clear system of flags to distinguish different levels of review, particularly for outsiders to a field and policy-makers. Such a system must be able to follow individual publications through the time line of pre-review deposit, to final publication. [is OSF enough?]

Let’s create “open think tanks” where policy proposals could be posted for wider scrutiny in an open, parallel, and public version of the decision processes that actually take place inside government would.

To enable critique, and allow others to build on our research, we need data and code sharing; but preparing data and models for sharing is a resource intensive process. The right balance will need to be struck.

We need to continuously aggregate to try and form accessible narratives to establish policy-relevant conclusions. Wikipedia-style system should be set up now.

We need to rapidly establish dynamic databases of expertise to allow speedy, dynamic access and responding. Extant systems are fragmented, incomplete, and focussed on past research. Info needed not just on what people did, but doing now.

Managing disagreement: even when all parties act in good faith, setting aside all theoretical differences they can, disagreement will remain. Here, suppressing disagreement is undesirable. Policy-makers need to know when scientific community has a range of legitimate opinions.

Can we develop new forums for making a case and leaving third party arbiters to decide? Experiences from the development of adversarial collaborations in the last decade are likely to have valuable lessons here (Kahneman & Klein, 2009).

Funding: We need resilience (given the nature of the crisis), and we need epistemic diversity: We need small, accessible, rapid response, proposals! Also, potential diversion of already granted funds should be allowed.

Scientists live in silos: Different labels across areas are barriers to entry. Our information indexing systems (Google, WoS) are insufficient as they don’t summarise or integrate. Wikis and query posting systems needed! We are working on this.

Our team of journalists has been tracking a growing list of sites publishing misinformation about COVID-19 in our Coronavirus Misinformation Tracking Center. https://newsguardtech.com/coronavirus-misinformation-tracking-center/… Our browser extension flags those and other unreliable sources of medical information.

ANNOUNCEMENT: To help combat dangerous misinformation during the COVID-19 crisis, NewsGuard is temporarily removing the paywall from our browser extension to make it free to all users. https://newsguardtech.com/free

We're pleased to launch this effort with support libraries, NGOs, security companies, and internet service providers from a range of countries and regions, including @bt_uk, @MediaLiteracyEd @TrendMicro @FOSI @AtlanticCouncil @mcclatchy @FNOMCeO & more. https://newsguardtech.com/press/a-statem

In the midst of a crisis, truth matters more than ever. We hope this initiative can help more people stay informed and healthy during this time.

Today @GordPennycook & I wrote a @nytimes op ed "The Right Way to Fix Fake News" https://nytimes.com/2020/03/24/opinion/fake-news-social-media.html… tl;dr: Platforms must rigorously TEST interventions, b/c intuitions about what will work are often wrong In this thread I unpack the many studies behind our op ed 1/

Platforms are under pressure to do something about misinformation. Would be simple to rapidly implement interventions that sound like they would be effective. But just because an intervention sounds reasonable doesn’t mean that it will actually work: Psychology is complex! 2/

These are cases where intuitively compelling interventions may actually be problematic. Its essential for platforms to test if the results from these experiments generalize to actual behavior on-platform But also, intuitively UNappealing interventions may actually work well! 6/

Crowdsourcing also robust against "gaming": 1) Poll random/selected users rather than allowing anyone to contribute their opinion-Prevents coordinated attacks 2) Knowing ratings will influence ranking≠gamed responses-Most ppl dont care about politics https://psyarxiv.com/z3s5k/ 8/

For example, its intuitive that emphasizing headline's publisher (ie source) should help people tell true vs false Low quality publisher? Question the headline. But in a series of experiments, we found publisher info to be ineffective! Details: https://twitter.com/niccdias/status/1217473772166381573?s=20… 3/

What about warnings on articles factcheckers mark as false? Seems like that should reduce belief- and it does! The problem: Most false headlines never get checked (fact-checking doesnt scale) & users may see lack of warning as implying verification! https://twitter.com/DG_Rand/status/1236102072795308033?s=20… 4/

Another example: General warnings to "Watch out for fake news!" Should help keep users on their toes, right? But this can lead to people not just disbelieving false headlines, but also rejecting TRUE headlines (ie being generally suspicious) https://link.springer.com/article/10.1007%2Fs11109-019-09533-0… 5/

Take crowdsourcing: When Facebook announced they would promote content from news outlets that users said they trusted, everyone thought it was a terrible idea! But turns out layperson source ratings actually agree quite well with fact-checkers: https://twitter.com/DG_Rand/status/1089999404898095105… 7/

Similarly, nudging people to think about the concept of accuracy makes them less likely to share misinformation This is the case in survey experiments (eg looking at sharing intentions for false and true headlines about COVID-19) https://twitter.com/DG_Rand/status/1240010913270370305?s=20… 10/

And of course, sometimes experiments find that interventions DO work the way intuition suggests For example, when people think more carefully, they are less likely to believe false headlines (but not less likely to believe true headlines) https://twitter.com/BenceBago/status/1220099034465144838?s=20… 9/

Finally, if you want to learn more, below is an always-updating doc with links to ALL of the papers @GordPennycook and I have written about misinformation / fake news (most of which also have Twitter thread summaries) https://docs.google.com/document/d/1k2D4zVqkSHB1M9wpXtAe3UzbeE0RPpD_E2UpaPf6Lds/edit?usp=sharing… end/

...and also in an actual field experiment on Twitter where we sent an accuracy nudge message (asking them to rate the accuracy of a random headline) to over 5k users and found an increase in the quality of the news they subsequently shared https://twitter.com/DG_Rand/status/1196171145227251712?s=20… 11/

My group, w @j_a_tucker and Paul Resnick's groups, are having a great experience in such a collaboration with Facebook around crowdsourcing https://axios.com/facebook-fact-checking-contractors-e1eaeb8b-54cd-4519-8671-d81121ef1740.html… I hope FB, and other platforms, will do more of these! 13/

TAKE-HOME Platforms need to do rigorous tests- and if they can show they are doing so, the public needs to be patient The key: Platform transparency about evaluations they conduct internally, and collaboration with outside independent researchers who publish 12/

COVID-19 Social Science Research Tracker

A comment on Everett et al. (2020): No evidence for the effectiveness of moral messages on public health behavioural intentions during the COVID-19 pandemic

https://cmu.edu/epp/people/faculty/research/Fischhoff-Analyzing-Disaster-Risks-Avian-Flu-JRU.pdf… and Fischhoff, B., Wong-Parodi, G., Garfin, D., Holman, E.A., & Silver, R. (2018). Public understanding of Ebola risks: Mastering an unfamiliar threat. Risk Analysis, 38(1), 71-83. doi: 10.1111/risa.12794 2)

We need to build on what's already there, eg., on risk analysis and risk perception: https://cmu.edu/epp/people/faculty/research/Fischhoff-Expert-Judgments-of-Pandemic-Influenza-Risks.pdf…

It was pointed out to me, by @VeroRabeloPhD, that the graphic I'm referring to was authored by Mimi Zhu, a queer Chinese Australian writer/organizer, and then circulated by Spears

I'm deeply angry at psychology at the moment, and I can't think of an outlet for that anger. So, Twitter it is. #Phdchat #AcademicTwitter #AcademicChatter

Global hackathon aimed at combatting Covid-19MEDIA_kit ENGCONTACT Elis Tootsman (+372) 506 6145 elis@accelerateestonia.ee Linkedin Fighting a Global Crisis: “The Global Hack” (9-12 April) webpage The Global Hack, the biggest online hackathon ever The Global...docs.google.com

UCLA Researchers Use Big Data Expertise to Create a News Media Resource on the COVID-19 Crisis

Anyone can get it

Overconfidence is a pernicious bias, even in experts. It's astounding how few experts' confidence intervals included the correct estimate of #COVID19 infections in the US by 3/29 when forecasting for just two weeks in the future. (of course, non-expert estimates are even worse)

.@ThreeLittleMigs and I are polishing a health behavior for @PsySciAcc's COVID project. We expect to reach thousands of participants all over the world. We want our survey to be useful to epidemiologists. Would any epidemiologists be willing to give us feedback? DM me if so!

an interesting source of statistics, both on COVID-19 and other issues that help provide some context to numbersWorldometer - real time world statisticsLive world statistics on population, government and economics, society and media, environment, food, water, energy and health.worldometers.info

Finally journal editors are losing their lock on scholarship- COVID19 is speeding this up. @elife will "make the posting of preprints to bioRxiv or medRxiv the default for all eLife submissions" @eLife ahead of the pack; the leaders are @Meta_Psy and other small journals. @SciBeh

We're working hard at the moment to scale up our attempts to detect COVID-19 disinformation, online manipulation, attacks on public health officials and miracle cures. If you've come across this kind of stuff, I'd really appreciate hearing about it.

Rapid Registered Reports initiative aims to stop coronavirus researchers following false leads

RT @paimadhu: I suspect most folks rushing to pre-prints with correlations between #Covid_19 & a host of variables (e.g. BCG, malaria endem…

It was a privilege to chair the @BPSOfficial #COVID19 Coordinating Group today. The challenge is significant but the breadth of psychological skills & knowledge within BPS gives us a real capacity, & duty, to play a major role in tackling the pandemic & its impact on society(1/2)

Within the framework below we're prioritizing areas where there's time urgency & BPS can make a sig. impact. ATM these are disease prevention, staff wellbeing, effect of confinement esp on vulnerable groups, psychol care of patients & relatives & adapting psychol services (2/2)

“social distancing was unnecessary” is the new “why should we vaccinate against diseases nobody gets any longer?”Citar TweetPhilip Bump@pbump · 9 abr.A notable shift downward in projected deaths from coronavirus is already being spun as "experts were wrong!!" instead of "hey, the thing experts said would drive down deaths might be driving down deaths." https://washingtonpost.com/politics/2020/04/08/leading-model-now-estimates-tens-thousands-fewer-covid-19-deaths-by-summer/

It was amazing to lecture at #NET_COVID today. We need this! Network epidemiology needs fresh ideas, we all need more model literacy, and things to do while in quarantine. Catch up on our first meeting w my introduction to network epidemiology models:

I'm trying to compile #COVID19 reports that use cellphone data around the world. I've done a few (see link below) https://leoferres.info/blog/2020/04/10/covid19-mobility-reports/… Can you help me by leaving a message, or replying to this tweet or just email with a relevant document? Thanks!

Testing for COVID-19 is generally done by nasopharyngeal swab, which is not easy to self-administer. But let's assume an easier nasal self-swab is possible at scale (e.g. https://medrxiv.org/content/10.1101/2020.04.09.20057901v1…) and 95% of people attempt the test each week. 4/

What happens if someone is infected? Once the sample arrives at the lab, the PCR test then needs to detect the virus. At best, it might have around a 95% chance of doing this (i.e. sensitivity = 95%) https://medrxiv.org/content/10.1101/2020.04.05.20053355v2… 6/

I'm seeing more and more suggestions that contact tracing and/or physical distancing isn't needed and we could solve COVID-19 with widespread testing alone. E.g. just test everyone once a week/fortnight to get R<1. Sounds straightforward? Unfortunately not... 1/

On average, each person infected with COVID-19 can spread infection to around 3 others in the absence of control measures. So if we want to get fully back to normal, testing will need to stop at least 2/3 of transmission (so each case infects less than one other person). 2/

Of course, people also have to collect the sample properly, pack it and ship to a lab to be tested. Let's suppose 95% manage to do this. (For context, this study found less than 80% of participants correctly collected and shipped a nasal swab): https://ncbi.nlm.nih.gov/pmc/articles/PMC4653956/…) 5/

Even if we ignore the (enormous) logistical challenges of pulling together resources and expertise to run the required millions of daily PCR tests, we still have to remove 2/3 of the opportunities for transmission. So let's suppose everyone receives a weekly test. What next? 3/

Optimistically, let's assume successful detection prevents 75% of onward transmission on average. We've estimated we can detect 86% of infections, so multiplying together that means preventing just under 65% of transmission.... 9/

We don't just need to detect infections, we need to reduce transmission. If testing is weekly, and people become infectious after 3-4 days, then (assuming same-day turnaround of results) we'd only detect half of cases before they became infectious https://nature.com/articles/s41591-020-0869-5… 8/

So to recap, our optimistic assumptions have 95% of people doing the test, 95% doing it properly, and 95% of infections being picked up. Multiplying together, that means we'll detect around 86% of infections. But there's another problem... 7/

We really do need better testing/tracing strategies and other innovations to reduce COVID-19 transmission. But this figure shows the fundamental challenge we face with COVID control - this is the problem any strategy has to solve: https://twitter.com/trvrb/status/1250855521327788032?s=20… 12/12

But remember, we need to stop at least 2/3 of transmission to control COVID (and even then it won't disappear immediately). Unfortunately, even our very optimistic scenario falls short of this - if mass testing were feasible, it would prob need to be biweekly, not weekly. 10/

Inefficient population testing could also crowd out better targeted approaches, such as contact tracing, which can informed by we know about transmission chains and - as we get more data - which contacts/environments are particularly risky 11/

Right now the comments on this very thread are full of this sort of thing from (non-trollbot) accounts. This one asserted that medical experts are lying to the public "explicitly for short term political gain at the expense of human lives." and then doubled down when questioned.

I'm astonished that people think doctors, biomedical researchers, and public health professionals—who have devoted their lives to helping people—are lying about #hydrochloroquine treatment for #COVID19 just to hurt Trump, rather than expressing justified skepticism based on data.

Until the past month, medical professionals have been among the most trusted people in America (below). We're seeing a dramatic turn in this, right at the time they are literally risking their lives to help the rest of us.