Hacker News Reader

Comment by vmchale | Hacker News Reader

> hospital admissions) be the almost singular criterion to influence public policy / restrictive measures?

From a control theory perspective, it's one of the worst measures because it's delayed. Much more efficient to measure cases - then you don't need restricted social life as long.

View on HN

It's an important metric because earlier indicators aren't as useful. The rate of new cases is tricky because we don't really care if someone gets COVID-19 anymore than we care if they get the common cold, assuming they have only mild symptoms and fully recover. Hospitalizations are the first unambiguous metric we have to know there is a serious problem that may require a dramatic response.

> It's an important metric because earlier indicators aren't as useful.

Sure they are. They're less precise, but they're more useful, because they're more likely to let you combat the spike before it gets out of control.

It's like a smoke alarm. Maybe it's going off because it's over-sensitive and someone just took a shower... but it's a better early warning system than waiting for active flames to appear.

I agree, but once vaccinations are widespread and we are confident that severe illness will be very rare, it would be hard to take drastic measures based only on the rate of new cases. There could be a huge surge of cases but we would expect a low hospitalization rate, so we would likely want to wait.

Once vaccinations are widespread, yes.

Until then, getting out in front of a spike remains necessary.

Immunity through vaccination + natural immunity is getting closer to 'widespread' now (in the US). We are likely near 30%, and of that 30%, a growing proportion is the most vulnerable sections of the population. A spike in cases now would be increasing less connected to a future rise in hospitalizations. I think its fair to say, that if we do not see a spike within the next 3-4 weeks, we are in the clear for future spikes. By April 1st, we should be nearing 40%, with the difference primarily being now vaccinated old and immune compromised people.

You use cases to determine if you need to react to stop spread since you get results faster. You use hospitalization to determine the time-lagged efficacy of your interventions since it is more precise. What is more useful depends on what you are trying to get out of it. In this particular case, we are discussing the efficacy of vaccination and how that might influence restriction lifting, so you want to err on the side of caution and accuracy and use hospitalization.

In the wild phase of the pandemic, yes.

But once enough people are vaccinated, the pattern may change. For example, you may still catch quite a lot of cases through PCR testing which is very sensitive, but the share of asymptomatic cases will be much higher and the share of people who are going to suffer a severe case much smaller.

The entire societal signature of the disease will change depending on vaccination levels and maybe even particular vaccines used.

The point isn't that we use hospital admissions as a metric to infer cases or some other metric, it IS the metric that matters. The problem over the last year wasn't that there exists a person who is at risk, it's that there didn't exist any healthcare system capacity to treat them.

From very early on in the pandemic, it seemed to me like our goal for optimal balance between caution/risk was to try and maintain the highest level of hospital occupancy that is sustainable. If every single person quarantines perfectly, then our hospitals are empty, but so is every business. If no one quarantines, our businesses are full, but so are hospitals and everyone's viral load; i.e. maximum mortality.

With the vaccines now available, it seems like our goal should still be be to maintain the highest level of sustainable medical system occupancy.

Call me a renegade, but I think the only goal should be to minimize the total number of deaths from Covid. Trying to maximize sustainable medical system occupancy is almost certainly not aligned with the former goal.

No perfect solutions there. If there was perfect quarantine, we would still lose additional lives from suicide, domestic violence, etc. from the increased stress. There is definitely a balance to be struck, and making sure we have capacity to treat all of the sick seems to be as good as any.

Yeah, making sure we have capacity to treat anyone sick is certainly one of the highest priorities (and the reason there was such a scramble trying to contain the virus from wreaking havoc if we went past that). And I get that there are other "costs" to the virus and lockdown and a holistic, balanced view is useful. Heck, I have two young children who have been remote schooling for almost a year now while both parents work full-time.

I just object with the parent's following statement:

> With the vaccines now available, it seems like our goal should still be be to maintain the highest level of sustainable medical system occupancy.

If we are still trying to maintain the highest level of sustainable medical system occupancy while the percentage of population that has been vaccinated is slowly increasing, it would actually mean we are doing a horrible job in trying to limit preventable deaths when a fully vaccinated populace + herd immunity is not too far away. Maybe when a much larger percentage of the population is vaccinated, the spread and rate of deaths will be low enough that this may change, but we are still nowhere near that stage.

Reply on news.ycombinator.com