Thread by @danielgoyal

How lethal is Covid now?

A few scientists/statisticians have been comparing the ONS Covid survey with Flu modelling studies to compare infection fatality rates.

It is so important to understand how dangerous Covid is, but IMV, such comparisons are simply too inaccurate.

It is important to understand Covid risk:

1. We, the public, need to understand the risk so we can make a choice regarding masks, socialising and of course vaccination.

2. Policy-makers need to understand what impact Covid is likely to have on society and service provision.

The ONS survey has been an incredible resource. For those who don’t know, the ONS randomly samples thousands of U.K. citizens each week with PCR swabs. From this they can estimate the total number of Covid cases in the U.K.

It is effectively a weekly point prevalence study!

Of course, this method picks up those who never would have come forward for testing, very mild cases and asymptomatic cases.

So, this tells us the total number of Covid infections/cases.

Flu rates are not calculated this way, unfortunately.

Europe and the US use modelling to calculate how many cases of flu there are. For example the CDC use confirmed cases and then use behaviour modelling to estimate how many didn’t come forward for testing (X2).

They estimate 10% of the population get flu in the US yearly.

The other major factor making the comparison between survey data and modelling data unusable, is the way deaths are calculated.

In the U.K., excess mortality is used in the flu model and in the US “Flu-associated” deaths are used…

“We look at death certificates that have pneumonia or influenza causes (P&I), other respiratory and circulatory causes (R&C), or other non-respiratory, non-circulatory causes of death, because deaths related to influenza may not have influenza listed as a cause of death”
CDC

The argument is probably right…Flu can compromise other conditions and therefore contribute to decline. But for Covid deaths we use confirmed deaths based on death certificates. The doctor looking after the person must consider Covid either caused or was factor leading to death.

If flu circulates at higher levels than estimated, if there are more than double the number of people staying at home with flu and not contacting the doctor, then the infection fatality rate of flu will be much less than the estimates.

And if we remove the associated deaths and stick to just death certifications, the IFR of Flu drops significantly.

Perhaps all IFRs for all infectious diseases should undergo similar surveys as the ONS so we can compare. But they don’t. So we can’t.

Lau et al hold sums it up:

“The argument is that the RT-PCR CFR in locations under certain criteria (e.g., with large-scale community transmission, without medical breakdown, with extensive testing) should be considered as a reliable reference value for policymaking”

doi.org/10.1016/j.ijid…

There is no doubt Covid is far more dangerous to the public at this present time, as it has been over the last two years. Very few would argue otherwise.

Yearly mortality rate in the U.K. is currently ~60/100k versus the CDC FLU estimates of around ~2/100k.

It still looks like the individual risk of severe outcome once contracted Covid (for an average joe with triple vaccine cover) is currently low. Even most CEVs, the risk is likely relatively low. But that is with vaccines. And now. Without vaccine cover the risk is high.

And even with good vaccine cover, the risk of Covid overwhelming hospitals repeatedly and causing repeated disruption to other urgent and routine care is very high just now.

While it may make us feel better to think of Covid like the flu, it is currently inappropriate.

IMO:
If vaccinated, get on with your life and take sensible precautions.

New vaccines and anti-virals are on the way.

But don’t think of it like the flu…it just isn’t.

For policymakers, expect Covid to at least double the pneumonia burden [better comparison].

Interested to hear others view on it, inc:
@Kit_Yates_Maths @JamesWard73 @chrischirp @devisridhar @SGriffin_Lab @jburnmurdoch @EricTopol @DrEricDing @Bob_Wachter @dgurdasani1 @JeremyFarrar @CMO_England @DrGregorSmith @zorinaq @PaulMainwood & others…