🦠 COVID-19 | On honesty, openness and transparency
I am a scientist by education, banker at JPMorgan for a few years, then mature PhD student in Chemical Biology at Oxford under the supervision of Christofer Schofield (FRS) and Peter Ratcliffe (Nobel laureate in medicine in 2019). Founder and tech investor focusing on media and education. I care about science, learning and Democracy which are good bedfellows.
📊 Daily Data Brief:
Cumulative case: 4,982,762 (+96,794) cumulative cases
Active cases: 2,701,898 (+39,009) (this is the number of currently infected patients)
Total Deaths: 324,523 (+4,645)
Serious/Critical Cases: 45,440 (+586)
Recovered: 1,956,341 (+53,140)
Source: Worldometers
1) Seven-day rolling average of new deaths (updated daily as ECDC releases). Major update with per country graphs now available (Link) (US, UK, France, Germany, Sweden, Nordic Comparison) (👈NEW❗️)
Showing a chart from the FT today highlighting the worrying situation in Brazil and India
2) U.S. states reopening risk map: this analysis includes current estimated R (reproduction number) for each state (Link)
3)Rt estimate per country (NEW❗️). This is a new resource link in the data section from a team which has led accuracy in modelling fatalities in the US for the past few weeks.
Another rich (albeit late) Corona Daily today.
A great article on the fallacious “follow the science” by the President of the Royal Society. Tom Frieden does a very good thread supporting the article whilst answering an important question: Why do some young healthy people get asymptomatic infection and other severe illness from COVID19?
Brazil is quickly turning out to be a human catastrophe as reported in the Wall Street Journal.
The “red block” of worst performing countries has Russia and Eastern Europe as well as the U.K. and Sweden. That is the picture of the day.
Kai Kupferschmidt talks to us about k an important metric to determine for SARS-CoV-2 and understand to inform public policy going forward.
Helen Branswell takes a closer look at Moderna’s data and she is more skeptical than enthusiastic about it.
Cambridge is physically re-opening in Summer 2021 at the earliest.
A fascinating study on the importance of data representation for the public to understand the exponential nature of the pandemic.
A great explainer on contact tracing.
A new and great data resource upfront and the data section: the modelling by this team has led in accuracy for several weeks now.
🛑 Article of the day: Venki Ramakrishnan (President of the Royal Society) writes “Following the science”.
It is no surprise and clearly welcome that the president of the Royal Society expressed himself about the politicised use of the expression “follow the science” (and as I argued in a previous edition of derivatives of it such as “publish the science”), when political leaders publicly communicate their policy decisions.
Ramkrishnan makes a useful distinction between “scientific advice” and “science”. It is scientific advice which interfaces with policymakers and constitute “only one of the things they need to consider” in their decision making.
Early on in the article he also notes the uncertainty and evolving nature of science which contrasts with the urgent need to decide for political leaders:
“At the frontiers of science, there is always uncertainty, and to pretend otherwise would be foolish. What science does is to try to gather evidence to reduce the uncertainty, but this happens only gradually as data are gathered and hypotheses tested and discarded until some idea of the truth emerges. But even those “truths” can fall by the wayside in the face of new and contradictory evidence. The entire process is based on honesty, openness and transparency, in which the evidence is published for all to see and argue about. It is no coincidence that scientists are highly trusted.”
Ramakrishnan also rightly points out, that despite the sensible need to give a centre role to scientific advice during a pandemic, it has by no way a monopoly on the matter as Jana Bacevic argued superbly in her LSE blog. In gathering all the advice from these different domains, politicians could choose to use the scientific method. The “honesty, openness and transparency” which underpins it and is which bestows the trust of the public in science, would equally serve them well.
Unfortunately, it is not the method which political leaders have chosen, instead preferring to use ‘science’ as a convenient potential scapegoat for their failures. That dishonest politicians are not trusted and consequently not re-elected would be welcome. Unfortunately lives will be lost in the near term and science and its well-earned trust will be damaged in the long term, particularly if people like Ramkrishnan do not speak out as he so urgently did with this article.
It is actually the opposite which should happen, and the convenient, short and inappropriate appropriation of science by politicians, should instead be replaced by a long term support and healthy interface with it. It will help us to tackle some of the upcoming complex problems in which science and scientific advice have a key role to play. Investing in it will increase preparedness and resilience. As Ramkrishnan points out:
“Not spending on resilience to predictable crises is a false economy.”
A number of advocates of mask wearing will also be delighted of Ramakrishnan view on the issue. A fantastic and urgently needed read. (Link)
🦠 Thread of the day: If the President of the Royal Society provided the theoretical piece on Science, Tom Frieden provided the practical piece. And how lucky we are to have the former Director of the US Center for Disease Control and Prevention guide us on the scientific evidence and hypothesis formation around the question: Why do some young healthy people get asymptomatic infection and other severe illness from COVID19: (Twitter thread)
Apart from being informative on an important question, it also highlights how there is no such thing as following “the” science.
A treat in-person teaching group which strongly supports Ramakrishnan article.
🌍Tweet of the day: The new red block now including Sweden and the UK…
🇧🇷 Luciana Magalhaes, Ryan Dube and Jeffrey T. Lewis write “Brazil’s Nurses Are Dying as Covid-19 Overwhelms Hospitals” in the Wall Street Journal. Brazil is fast turning from a tragedy to a human catastrophe:
“The coronavirus is spreading rapidly throughout Brazil, overwhelming a health care system that is ill prepared to handle a pandemic of this magnitude and proving especially deadly for the medical workers on the front lines.”
President Bolsonaro is a staunch opponent of lockdowns. While there is no such thing as “following the science”, there is murderous incompetence:
“The people want to work to put food on the table for their families,” Mr. Bolsonaro said Sunday via his Twitter account. He has said the lockdowns are hurting ordinary Brazilians. “Millions (in Brazil) already know what it’s like to live in Venezuela,” he said in a recent tweet, referring to the hunger and poverty in that neighboring country.
The president’s stance on the pandemic has brought him into direct conflict with infectious disease experts, including people in his own health ministry, and several governors. The president fired Health Minister Luiz Henrique Mandetta in April, and on Friday Mr. Mandetta’s replacement, Nelson Teich, resigned.
Some models are predicting that Brazil will end up with 350,000 deaths above any other country. A catastrophe. (Link)
💉 Helen Branswell writes “Vaccine experts say Moderna didn’t produce data critical to assessing Covid-19 vaccine” in STAT news. Branswell had urged caution in the Moderna announcement on Wednesday. She was not the only one:
Since then Branswell talked to a few experts and highlighted a few areas of concerns in the announcement: “The silence of the NIAID”, “The n = 8 thing” (the significant piece of good news was on 8 trial participants when the phase 1 enrolled 45), “There’s no way to know how durable the response will be”, “There’s no real way to contextualize the findings” and “Moderna’s approach to disclosure”.
There is a particularly salient detail at the end of the article:
“Moderna has been more forthcoming with data on at least one of its other vaccine candidates. In a statement issued in January about a Phase 1 trial for its cytomegalovirus (CMV) vaccine, it quantified how far over baseline measures antibody levels rose in vaccines.”
Why did Moderna not produce more data on this trial? Maybe it just did not have the data yet. I do not think that Moderna could be blamed for the irrational exuberance of the markets. Was it wrong or simply opportunistic to leverage it with a $1.3 billion stock sale? To be followed. (Link)
🎓 “Cambridge University: All lectures to be online-only until summer of 2021” for the BBC. This is a major announcement, and probably one which will disappoint a number of students which will not have the student life they expected next year. It also raises questions as to whether this would become the new normal and whether increasingly the Cambridge educational model, and that of other universities will increasingly move online and closer to the Open University model.
The formative university years are much more than pushing content down a stream to students and learning also involves and requires much more than an online interaction. In taking such a decision, Cambridge publicly stated its belief that COVID-19 will become endemic and require a new normal for the foreseeable academic year at the very least. (Link)
🦠 Kai Kupferschmidt writes “Why do some COVID-19 patients infect many others, whereas most don’t spread the virus at all?” for Science.
Just when you thought (and Kupferschmidt himself) you had heard about all the variables you needed to know about a virus there is a new one:
The article and the ensuing two threads are about the value of k for SARS-CoV-2. The lower the k the more transmission of a pathogen starts from a small number of people. Studying and ultimately understanding dispersion can inform and guide better policy decisions on how to contain an epidemic like COVID-19.
A recent preprint, from Adam Kucharski (from the London School of Hygiene and Tropical Medicine, and frequently featured in the Corona Daily) estimated that k for COVID-19 could be as low as 0.1.
It is again one area where evidence needs to be gathered. At the moment, science is formulating hypothesis which could be strengthened through more data. It is work in progress, time consuming and difficult to measure and understand why k for a pathogen is low. As Kupferschmidt writes:
But studying large COVID-19 clusters is harder than it seems. Many countries have not collected the kind of detailed contact tracing data needed. And the shutdowns have been so effective that they also robbed researchers of a chance to study superspreading events. (Before the shutdowns, “there was probably a 2-week window of opportunity when a lot of these data could have been collected,” Fraser says.)
We then fall into the thorny question of privacy which would enable this data to be gathered as countries start to re-open. The research around k is also prone to bias.
It is a fantastic article demonstrating again science in action. It is also hard to see this article not being politicised by digital contact tracing proponents or one who advocate a hasty re-opening (a subset of which believe that lock-down should have never happened). (Link and Twitter thread #1 and thread #2)
🧮 Alessandro Romano et al. write “The public do not understand logarithmic graphs used to portray COVID-19” for the LSE blog.
Absolutely fascinating study. We all know that “words matter” and that exponential growth is not intuitive, but Romano and his colleagues show that data representations matters a lot as well:
“Mass media routinely portray information about COVID-19 deaths on logarithmic graphs. But do their readers understand them? Alessandro Romano, Chiara Sotis, Goran Dominioni, and Sebastián Guidi carried out an experiment which suggests that they don’t. What is perhaps more relevant: respondents looking at a linear scale graph have different attitudes and policy preferences towards the pandemic than those shown the same data on a logarithmic graph. Consequently, merely changing the scale on which the data is presented can alter public policy preferences and the level of worry, even at a time when people are routinely exposed to a lot of COVID-19 related information. Based on these findings, they call for the use of linear scale graphs by media and government agencies.”
Must read. (Link)
🚔 Caroline Chen writes “You Don’t Need Invasive Tech for Successful Contact Tracing. Here’s How It Works.” for ProPublica. The title is somewhat mis-leadind as the article is mainly a great explainer about contact tracing, how it works, where does the U.S. stand in its capacity, how is it useful beyond containment and to help about the disease. The digital skepticism is a small piece of the article. I was hoping it would be more prominent as it would be somewhat at odds with Kupferschmidt article on finding k and the evidence supporting its value.
As contact tracing is coming and you might get a call, read Chen’s excellent primer. Natalie E. Dean (Assistant Professor of Biostatistics at @UF) also did a great thread citing the article. (Link and Twitter thread)
📊 A picture is worth a thousand words: Global (🌎) and local (with relevant flag) visualisation and forecasting tool
🦠 “Science Forum: SARS-CoV-2 (COVID-19) by the numbers” (Link)
“The COVID-19 pandemic is a harsh reminder of the fact that, whether in a single human host or a wave of infection across continents, viral dynamics is often a story about the numbers. In this article we provide a one-stop, curated graphical source for the key numbers (based mostly on the peer-reviewed literature) about the SARS-CoV-2 virus that is responsible for the pandemic. The discussion is framed around two broad themes: i) the biology of the virus itself; ii) the characteristics of the infection of a single human host.”
🇺🇸🌎
NEW❗️This model has led accuracy for several weeks in the US. It also does projection for Europe and Rest of the World. (Link)🇺🇸 “Is your community ready to reopen?”: A map of the US (50 states and 2,100+ counties) looking at reopening risks with metrics around 3 criteria: 1. Is COVID in retreat? 2. Are we testing enough? 3. Are our hospitals ready? (Link)
🌎The Financial Times (visualisation) has a data tracking page which is in front of the paywall, looking at cases and fatality curves for selective countries and metropolitan areas/region. It is not as extensive as the Madlag link below, where you can see static as well as animated images for a greater number of individual countries. (Link)🇺🇸 The Johns Hopkins University resource center was the first one I used back in January they have now made available in their latest iteration a county by county dashboard in the US including information about health capacity, insurance coverage, ethnicity and age breakdown of the populatio (Link)
💊 The "Map of Hope" provides a geographical overview of planned, ongoing and completed clinical trials. It is put together with data from WHO Clinical Trials Search Portal by the Heidelberg Institute for Geoinformation technology. (Link)
🌍 MRC Centre for Global Infectious Disease Analysis started to publish weekly death estimates for countries (Link)
🇺🇸 The US Center for Disease Control and Surveillance (CDC) publishes “A Weekly Surveillance Summary of U.S. COVID-19 Activity” (Link)
Google has published a new website to “See how your community is moving around differently due to COVID-19”. They have a lot of data to do so… (Link)
🌎 Country by Country Curves: This is a GitHub made by my friend Francois Lagunas. He has written a script to scrape deaths and number of cases in order to visualise the rate of growth on a logarithmic scale. Great resource (Link)