We’ve heard about vitamin D possibly being beneficial in reducing the symptoms of COVID-19. I’ve been taking a small (1,000 IU) vitamin D supplement for years because I live in Rainland. I have no idea how much vitamin C I consume, but I haven’t gotten scurvy. I usually drink two 24-ounce (3 servings) cans of Monster Zero Ultra per week. (Not this week though; I drink them when I go to the office, and I’m on vacation this week.) One serving of Monster Zero Ultra has RDA 100% of vitamins B3, B5, B6, and B12; so I get 300% of the RDA when I drink one. FWIW, one serving of Vegemite has 50% of vitamin B1, 25% of B2, 25% of B3, 50% of B9.
Make the link for me between “more testing” and “lower death rate.”
Is it because with more testing you identify people who have it so you can quarantine them? Or treat them? But there really isn’t any specific treatment, is there? And where does the death rate factor in? Don’t mean to be dense, but I’ve had trouble with this concept from the beginning.
I should have said case fatality rate instead of a true infection fatality rate. That’s simply the deaths per people who test positive. If you’re only testing very sick people, you get a false impression of a high death rate because your denominator is too low. If you have more widespread testing, that denominator increases and your rate increases.
I’m not Tfletch1, but I can hopefully answer your question.
In this case, he is referring to case fatality rate (CFR), not to be confused with infection fatality rate (IFR). CFR is simply the number of people that died from COVID-19 divided by the number of people who have tested positive. So, if you have 100 deaths and 1000 total positives, you have a CFR of 10%. If you suddenly test a lot more people, without having more deaths, that number goes down. 100 deaths with 2000 positive tests and the CFR drops to 5%. Increasing testing drops the ratio. The statement of his/hers that you quoted means that the death rate is going up rather quickly as increasing testing isn’t dropping the CFR as fast as you would expect. It does make it a somewhat crappy number to use when estimating infectiousness, but is all we have for now. For trending purposes, death rate or hospitalization rate is probably a more accurate number to use at this time, although it is not perfect as well (we could be getting better at handling the tough cases, for example).
IFR on the other hand is when you take the number of deaths and divide by the number of people who are actually infected. This either requires a truly random sample of testing or a comprehensive testing framework where everyone is tested. As far as I know, neither of those have happened yet except on an extremely small scale (the CDC is attempting the former for a few select areas in Atlanta, but hasn’t finished). This number can tell how infectious it really is.
Edit: I see Tfletch1 snuck in there while I was typing. I’ll leave mine in case the IFR portion helps your understanding.
If they’re both otherwise behaving in the same fashion and the only difference is that one of them is coughing and sneezing, sure.
In practice, the asymptomatic person is more likely to be out and about and coming close to other people; plus which, visible coughing and sneezing is more likely to make other people back away rapidly.
Plus which, the question isn’t really ‘do obviously sick people cause more infections than asymptomatic and mostly asymptomatic people?’ The question is, ‘do asymptomatic and mostly asymptomatic people cause a significant number of infections?’
Yep. Usually IFR is taken from a model because, just like with Covid-19, other diseases, like the flu, will have a higher CFR than the true IFR. People simply don’t come in for testing unless they’re pretty sick. I’m not an epidemiologist but I think they get data from a random sample AND wait until the outbreak is over to get the best estimate of IFR.
Something I seldom see mentioned is the potential of this virus for deaths over the long term. New York and New Jersey are approaching 2,000 deaths per million and they are supposedly less than 50% infected. So this would mean that we have a potential for about 1 death in every 250 people or about 1.5 million deaths before it runs it’s course. I think you have to ignore transmission rates when talking about potential because all the transmission rates do is change the time frame of infections.
Amazing to me that he presented with those symptoms the second time around, and was just given an x-ray and sent home. I guess they presumed that he could not catch it again.
In case people don’t want to click the link:
On 5/31/20, the patient sought care with self-reported fevers, headache, dizziness, cough, nausea, and diarrhea. A chest x-ray was performed and he was discharged home.
Exactly. I got into an argument with my brother about looking at the big picture. He was saying that the IFR is going to end up being a lot lower than the CFR because there are probably so many more people infected than recorded. I told him that at this point it doesn’t matter because that’s simply saying that it’s more contagious. In the long run, you’ll have the same number of people who end up in the hospital or dead. Of course, this was when scientists were first speculating that aerosol spread is an important mode of transmission.
So as this has been progressing, the CFR has improved significantly but they’re finding that it’s more contagious than originally thought.
My cancer-patient mom was in a car unmasked for 3 hours round trip with a friend who began having symptoms the next day and got pretty sick (but was not hospitalized). Friend had a sneezing fit on the day she was with my mom (but not while she was in the car with my mom- she was in a store, waiting for my mom’s oncologist appointment to finish). Friend tested positive. And then my mom tested positive, twice. But she never had any symptoms. And she’s not exactly the picture of health. Bizarro as to who gets it and who doesn’t.
Both were unmasked? I was going to say that both being masked decreases the viral innoculum. But if they’re both unmasked…bizarro but definitely a good bizarro!
I wasn’t thinking they could use it to actually track down the specific infected individuals. But it would give a nice measure of whether the prevalence was increasing or decreasing. You could use the data to compare towns that opened schools in one mode with towns that opened school in another mode, and see if either one was working better to contain the disease. You could check to see if the amount of virus in sewage spiked hairdressers, or gyms, or restaurants were allowed to open. It would be an early warning of increased case loads in a region and could help the authorities divert resources to where they are needed. It could be used as a benchmark for when to open or close various types of establishments, too – maybe there are critical thresholds, and if the prevalence is less than X it’s reasonably safe to open stuff up, but above X the R shoots up.
It’s just something that could be measured in a reproducible way to produce actionable data.
There was another one posted earlier. These are all ‘antigen’ tests.
I’m forming the opinion that PCR testing is really laboratory testing, and scales very badly. To get cheap rapid testing, we have to replace amplification technology with sensitive direct-sample testing, typically with a laser-excited florescent dye that reacts with a COVID antigen.
Also, spit tests (as 2 of these 3 are), are less sensitive, but with an easy cheap fast test, you can just test everybody all the time.
The problem seems to be that, limited as PCR test machine availability is, antigen test equipment seems to be even rarer.
I haven’t seen any indication that my government (Aus), or NZ, or USA, have any real interest in cheap mass testing, but I won’t be surprised if private companies take it up for workplace testing.
News from the Victorian outbreak: for the first time in about 2 months, 4 weeks into our 6-week Stage 4 Restriction period, daily case numbers are back down to double digits (from a high of close to 700). Long may it last.