It all hinges on the estimates of VITT. Yesterday @COVIDSciOntario posted a VITT brief covid19-sciencetable.ca/sciencebrief/v… that gave updated estimates of risk: 1:26 000 to 1:127 000 (as opposed to govt doc using 1:100-250K), and I believe ours is the most accurate estimate of risk of VITT.
If we use the revised numbers, the incidence of an ICU admission from AZ vaccine becomes anywhere from 1.27-3.85/100 000. Using existing risk-benefit analysis, it means that it never makes "statistical sense" to use an AZ vaccine where the COVID incidence is moderate (7.5/100K/d)
It also means that AZ makes sense for ages 40-60 if the incidence is high (i.e. 30/100K/d). Where in Canada does that occur? North & South AB, Calgary, Edmonton, Peel, Toronto, and maybe Winnipeg. We should be getting +++ mRNA vaccines to these places ASAP.
Everywhere else, it no longer makes sense based on what we know, to give AZ vaccine.
If we cannot get mRNA vaccine to those other places then, yes, AZ makes sense. AZ made sense in other places a few weeks ago, when the wait for mRNA was longer and the incidence was higher.
This is my opinion based on the numbers: that we should stop administering AZ vaccine b/c of safety not efficacy. We are lucky, there is tons of mRNA now.
I am certain others will disagree, and I'm ok with that. I'm just trying to inform. But this is what we know today.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Reluctantly, I feel I need to clarify some issues around why AZ doesn't make sense for most of Canada right now.
When NACI evaluated—using the hard endpoint of deaths—the risk-benefit of AZ vs. no AZ, it used a lower incidence 1 per 100 000.
With this modeling, it makes clear sense to give AZ vs. waiting for age 50-69 in a moderate incidence setting, and for all ages in high incidence setting.
But what happens if the VITT rate is 1:26 000 or 3.85/100 000? You get this ...
This means that it is only a slam dunk (vs. no vaccine) for age 50+ in high incidence (30 cases/100K/day) settings, and 40+ in very high incidence (60 cases/100K/day) settings. Even if VITT incidence is 1:40 000 (or 2.5/100K/day), your expected VITT deaths/100K are 0.63-1.0.
Highlights for remdesivir 1. Not recommended for patients not on oxygen 2. Not recommended for patients mechanically ventilated. 3. Recommended for patients on low-flow supplemental O2 4. Consider for patients in between low-flow O2 and MV
Much of our advice comes from insights around this table of 28d mortality and the studies involved. We recognize that this isn't straightforward guidance.
I will start off by saying what we know about systemic corticosteroids in patients with COVID:
They likely work and save lives if patients need supplemental O2 or supported ventilation. The strongest evidence comes from the massive RECOVERY trial (N=6435, nejm.org/doi/full/10.10…). Importantly, dexamethasone appeared potentially harmful for patients not requiring O2.
It's possible that the best decision for the population and the best decision for individuals are different.
In a young otherwise healthy person where other vaccines are available soon and they can mitigate their risk, the risk of illness/death from AZ > risk of death from COVID
Would a parent give their 12yo kid AZ if it were shown to be safe and efficacious in studies but still carried the VIPIT risk? I seriously doubt it. Then how about a 15yo? 20 yo? 21?
At some age benefit > risk. NACI decided that that inflection point is 55y. I agree with them.
However, to the general public, the value of getting as many people as possible vaccinated is huge.
The argument of being in a 3rd wave isn't being lost on me, but most of these vaccines during the 3rd wave will/should not go in young people.
1. Let us first be very clear—we are in a 3rd wave—and epidemiologically, we are somewhere comparable to early-mid December 2020. From @COVIDSciOntario@Billius27
2. A huge difference is ICU capacity. In mid-December, we had around 276 patients with COVID in ICUs, and around 181 on ventilators.
Right now, that number is 355 and 205, the ICU workers are tired, and those numbers are just starting to rise.
3. How quickly cases rise is unpredictable. If we're lucky the rise will be more like 🇫🇷 than 🇮🇹 or 🇦🇹 but it is very clear that truly steep growth often occurs and (if you follow the international press) it always feels like it is taking you by surprise. It ain't.
As all of Ontario waits with baited breath to learn how @fordnation has decided to act on what the mostly leaked Science Table shows, I thought it important to explore what "all options on the able" look like.
A 🧵:
First, there are no MUST DOs.
There are places around the world who have kept schools open and seen cases drop, and there are places who have managed to drop without closing borders, restricting travel, or curfews.
But COVID is transmitted by people interacting, usually indoors in prolonged close contact.
The data (and my experience) seems to show that some of this post-holiday surge was related to people behaving as people do. globalnews.ca/news/7555586/p… It appears that we will hear more of this tomorrow.
Although we deserve a fair amount of blame, govt should shoulder much, too...