The fatality rate depends heavily on the demographics of the infected population. IFR of a entirely young demographic would be below 0.01%. IFR of a nursing home would be 3 orders of magnitude higher.

If someone shows me a low IFR I can safely bet that they surely aren’t finding 50% of their cases in their nursing homes.

Conversely, if 50%+ of your cases are in nursing homes, your population IFR is going to look a lot worse than if you had kept that relatively small population safe.

Singapore is militant on testing everyone. This presumably gives a high fidelity on the "true" spread of the virus.

What is also true is that Singapore has been extremely militant on isolating and controlling the virus. Their staff have ample PPE and their hospitals have very successfully isolated covid positive patients. For instance, anyone positive here is quarantined in a govt run facility and cannot return home until testing negative two days in a row. This may have recently relaxed but was true throughout circuit breaker/lockdown. I believe they are also trying to partition covid patients from non covid patients at the hospital level. (eg NCID)

Both anecdotally from frontline doctors, and from reading various news reports, a big problem about the first wave of covid was how the problems can snowball. Hospitals were very easily filled up with sick patients, but worse any existing patients could very easily catch covid. For example cancer patients who are immunocompromised.

So the lethality of the disease is actually a function of its local penetration as more people die if hospitals effectively experience cascading failure.

My understanding is that tipping point is quite fine, this is especially true back in March when everyone was still hoping it wouldn't reach their shores, and hospitals had very limited space for isolating covid patients and for treating patients that required ICU.

While hospitals have been empty I also know that hospitals have become extremely aggressive at not accepting patients. For example, my friend's mother was bedridden for a month but was never seen by a doctor. Meanwhile you have people dying at home and in care homes, so I think "hospitals are empty" doesn't tell the whole picture.

This is false. Even if it were true, you could look to NYC's excess deaths and see a huge surge in them that would correspond to a similar IFR.

This is a common "fact" asserted in these discussions, and it's never backed up with any evidence.

> The soaring death toll has been fueled by the adding of 3,778 people who were not tested for Covid-19 but are presumed to have died from it.

Singapore’s organised approach is clearly going to produce much more reliable statistics than NYs presumption based approach.

Doctor friends tell me it's usually pretty clear when they have a COVID patient.

https://www.nbcnews.com/news/us-news/new-york-start-reportin...

> Asked about the numbers Wednesday, Dr. Oxiris Barbot, commissioner of the city Health Department, said the "unfortunate reality is there have been people who have died either directly because of COVID or indirectly because of COVID."

> Barbot said only time would tell what that number really meant. Some deaths, for example, could have been registered as having been caused by heart attacks because people had not yet developed coronavirus symptoms, when they should have been counted as probable COVID-19 victims, she said.

So NY officials state that their recorded death toll should reflect both direct and indirect fatalities, whatever an indirect fatality is supposed to mean, and that a person who dies without Covid symptoms, and who has not tested positive, should be counted as a fatality. So unless they have proof Covid wasn’t a contributing factor in any way, it’s counted, which does actually sound a lot like “ anybody who could have potentially been infected”...