[BostonCard]

There has also been a slew of incorrect information.  Masks are a great example. Early of the Health leaders were saying not necessary.  Now that’s changed. Mortality rate.  Heard very high numbers early on. About 4.3 percent.  Now looking like 0.2 to 0.4. Still higher than flu.  But (fortunately) much less deadly.  Transmission rate.  Early on said 4 to 1.  Now believed closer to 1 to 1.  Transmission from objects - wipe down everything they said.  Now, risk from surfaces very low.

Couple quick points:
Transmission rate (R0) has been estimated in the 2-3 range the whole time.  This is one of the parameters about the virus that was estimated early and more or less correctly, whether you look at the data in China, Italy, or New York.  Since that time infectiousness (Rt) has fallen to about 1, but it has fallen due to our actions (social distancing, lockdowns, etc).  The CDC's "best estimate" right now is that R0 is 2.5, in line with earlier estimates.

Early in the epidemic, there were warnings not to take the "naive" case fatality rate seriously.  Since not all outcomes had been fully adjudicated (not everyone who was going to die had died yet) the numerator couldn't be determined correctly.  Moreover, since we expected that very sick and dying patients were more likely to be tested, while less sick patients were not, the denominator couldn't be determined correctly either.  Furthermore, there is a distinction between the case fatality rate (proportion of all people who get sick with COVID-19 who die) and the infection fatality rate (proportion of all people who test positive, regardless of symptoms, who die).  The CFR is not as high as the WHO's initial estimate of 3.4%, but it is close to 1%.  The IFR is lower because of the number of asymptomatic people.  Based on deaths and serology prevalence in New York, the IFR is 0.86%; it is higher (1.4%) if you use total excess deaths rather than confirmed covid-19 deaths.  This is about what I had estimated earlier based on the Spanish seroprevalence data (the study, which was a randomized sample of the Spanish population showed 5% of Spaniards with antibodies to COVID-19; in a population of 47 million, that translates to 2.35 million infected; given 27,100 deaths, the IFR is 1.1%; if you go with excess deaths, the number of COVID fatalities is more along the lines of 44,000 and the IFR is closer to 1.9%).  I think in the end, the IFR will probably settle out to about 1%, lower than the original estimates of CFR, but higher than some of the other studies (like the one from Bhattarcharya and Ioannidis) have found.

I think with regards to the two other items you mentioned (masks and surfaces), I agree that messages have changed over time.  I think mostly this has to do with a better understanding of the disease and new information.  I will admit that I was one of those skeptical about the efficacy of masks, though there have been staunch supporters of it from the beginning here (I'm thinking of OutsiderFan).  For masks, a bit of it was influenced by a desire to preserve PPE for healthcare workers.  That shouldn't affect the interpretation of the data, but when data was still being developed, it was reasonable for recommendations to take that into account.  For transmission via fomites (surfaces), in the early periods of the pandemic, guidelines were based on limited data; basically analogies to other viruses such as influenza, and the persistence of COVID-19 on surfaces.  It was a reasonable extrapolation, and the messaging was more like "it could be transferred via surfaces" but that gets understood as "it is transferred via surfaces".  And, of course, while such transfers are rare, it is still reasonable to say that it "can" be transmitted by fomites, even as data has shown that the the dominant form of transmission is person to person.  More importantly, this is exactly what science is supposed to do.  As new data comes in,  underlying assumptions have to be questioned and our understanding of "the truth" has to change.  So, I get that the evolution of messages over time can be confusing, but I honestly don't know what can be done to prevent that, given the hunger for information early in the pandemic, before a full understanding of the data is available.

BC

[BostonCard]

There has also been a slew of incorrect information.  Masks are a great example. Early of the Health leaders were saying not necessary.  Now that’s changed. Mortality rate.  Heard very high numbers early on. About 4.3 percent.  Now looking like 0.2 to 0.4. Still higher than flu.  But (fortunately) much less deadly.  Transmission rate.  Early on said 4 to 1.  Now believed closer to 1 to 1.  Transmission from objects - wipe down everything they said.  Now, risk from surfaces very low.

Couple quick points:
Transmission rate (R0) has been estimated in the 2-3 range the whole time.  This is one of the parameters about the virus that was estimated early and more or less correctly, whether you look at the data in China, Italy, or New York.  Since that time infectiousness (Rt) has fallen to about 1, but it has fallen due to our actions (social distancing, lockdowns, etc).  The CDC's "best estimate" right now is that R0 is 2.5, in line with earlier estimates.

Early in the epidemic, there were warnings not to take the "naive" case fatality rate seriously.  Since not all outcomes had been fully adjudicated (not everyone who was going to die had died yet) the numerator couldn't be determined correctly.  Moreover, since we expected that very sick and dying patients were more likely to be tested, while less sick patients were not, the denominator couldn't be determined correctly either.  Furthermore, there is a distinction between the case fatality rate (proportion of all people who get sick with COVID-19 who die) and the infection fatality rate (proportion of all people who test positive, regardless of symptoms, who die).  The CFR is not as high as the WHO's initial estimate of 3.4%, but it is close to 1%.  The IFR is lower because of the number of asymptomatic people.  Based on deaths and serology prevalence in New York, the IFR is 0.86%; it is higher (1.4%) if you use total excess deaths rather than confirmed covid-19 deaths.  This is about what I had estimated earlier based on the Spanish seroprevalence data (the study, which was a randomized sample of the Spanish population showed 5% of Spaniards with antibodies to COVID-19; in a population of 47 million, that translates to 2.35 million infected; given 27,100 deaths, the IFR is 1.1%; if you go with excess deaths, the number of COVID fatalities is more along the lines of 44,000 and the IFR is closer to 1.9%).  I think in the end, the IFR will probably settle out to about 1%, lower than the original estimates of CFR, but higher than some of the other studies (like the one from Bhattarcharya and Ioannidis) have found.

I think with regards to the two other items you mentioned (masks and surfaces), I agree that messages have changed over time.  I think mostly this has to do with a better understanding of the disease and new information.  I will admit that I was one of those skeptical about the efficacy of masks, though there have been staunch supporters of it from the beginning here (I'm thinking of OutsiderFan).  For masks, a bit of it was influenced by a desire to preserve PPE for healthcare workers.  That shouldn't affect the interpretation of the data, but when data was still being developed, it was reasonable for recommendations to take that into account.  For transmission via fomites (surfaces), in the early periods of the pandemic, guidelines were based on limited data; basically analogies to other viruses such as influenza, and the persistence of COVID-19 on surfaces.  It was a reasonable extrapolation, and the messaging was more like "it could be transferred via surfaces" but that gets understood as "it is transferred via surfaces".  And, of course, while such transfers are rare, it is still reasonable to say that it "can" be transmitted by fomites, even as data has shown that the the dominant form of transmission is person to person.  More importantly, this is exactly what science is supposed to do.  As new data comes in,  underlying assumptions have to be questioned and our understanding of "the truth" has to change.  So, I get that the evolution of messages over time can be confusing, but I honestly don't know what can be done to prevent that, given the hunger for information early in the pandemic, before a full understanding of the data is available.

BC