[throwaway5466]

For people that didn't read: it's "disappointing" because both the control and study groups largely tested negative (93% and 87% respectively). That 87% figure sounds worse than the control but it actually is better than in the French study.

What happened in essence is that almost all cases were mild so the patients almost all got better after a while, with or without medication. This doesn't mean HCQ is ineffective, only that its efficacy couldn't be measured in this instance. And I wouldn't call a study where almost everyone gets better at the end "disappointing".

[halflings]

"That 87% figure sounds worse than the control but it actually is better than in the French study."

This is a moot point, because the French study was not a randomized control trial, on 6 people, at different stages of the disease from this study.

[cycrutchfield]

It's disappointing because it's no better than placebo.

[arcticbull]

In this case (as with many others) placebo is pretty effective, but also, these folks didn't get sick enough to need it, as with the overwhelming majority of young and healthy people. Another, broader study is required. I'm sure we'll find out more soon -- this reads like a mulligan though.

[hannob]

> In this case (as with many others) placebo is pretty effective

Where do you get that idea from? In the study it looks like everyone got better, but that may just be regression to the mean (i.e. the normal progression of covid-19 in those patients).

If you'd want to know if placebo is effective on covid-19 you'd have to do a completely different study, comparing placebo to doing nothing. Though the likelyhood that placebo therapy is "pretty effective" here is very low. Placebos usually work best in things that have a strong subjective factor, e.g. pain treatment.

[dajohnson89]

any data on what percentage of “young healthy” people die/get hospitalized? also, what’s the definition of young and healthy?

[arcticbull]

I mean, I’ve been saying this for a while and I don’t want to tire y’all out but, young being defined as under 40 and healthy being defined as without comorbid conditions.

The median age of the dead in Italy is 80.5 and 99.2% of them were sick, averaging 3 comorbid conditions [3]. So basically the opposite of that.

[1] https://twitter.com/AndyBiotech/status/1241741127205572609?s...

[2] https://www.cdc.gov/mmwr/volumes/69/wr/mm6912e2.htm

[3] https://www.google.com/amp/s/www.bloomberg.com/amp/news/arti...

[blacksqr]

I keep reading this statistic about deaths in Italy, but how much of it is shaped by doctors desperately deciding to withhold care from the sickest in order to save others?

i.e., would the age curve be different in an ideal situation where everyone was getting optimal care, and is the statistic thus misleadingly suggesting that younger people are less at risk?

[dev_tty01]

Would the curve be different in an ideal situation where the hospitals weren't completely overrun and out of resources? Yes, of course. However, I would guess that if care was withheld, it was based on disease state and patient robustness. Age is part of that equation so it is hard to separate it out. In other words, a fit healthy, no comorbidity senior whose immune system is fighting hard is probably not having treatment withheld just because of age. I'm not there of course, but based on accounts I think this is right. At least I hope so.

As an aside, you were certainly not being judgmental about the clinicians in Italy and I applaud you for that. Those professionals are doing incredible work in unbelievably difficult and personally dangerous situations, so I hope we can all agree not to second guess their decisions from the cheap seats.

[brandmeyer]

Data: https://covid19.colorado.gov/sites/covid19/files/COVID%20-%2...

This is a summary of all cases in Colorado. All of our state's (small handful of) fatalities have been among residents aged 60 and over. The hospitalization rate is ~5% or so for adults in the 30-50 year old range, and it rises to about 20% in the elderly age groups.

Caveats: with only 600 cases to draw inferences from, there's a little less than one digit of precision in those estimates. Testing availability is also still somewhat limited in the state with a little more than a 10% positive rate.

[asveikau]

The Imperial College paper that was being passed around said about 1% of 20somethings, 3% of 30somethings, and 5% of 40somethings wind up hospitalized.

I think if your population trends younger this can mean pretty high absolute numbers even if low percentages. A number of news sites in various places said the under 50 crowd is using significant portions of hospital resources.

[Terretta]

Across the United States, 38 percent of those hospitalized were between the ages of 20 and 54.

Nationally, 12 percent of the intensive care patients were between the ages of 20 and 44.

https://www.nytimes.com/2020/03/23/nyregion/nyc-coronavirus-...

[bmm6o]

I keep hearing statistics like this and I yell at my TV every time. Counting up the people in the hospital is pretty easy, but it doesn't answer the more important question of how many infected 20-54 require hospitalization. Even comparing those numbers to the larger demographics would be helpful.

[Terretta]

There are a variety of these sources.

https://statmodeling.stat.columbia.edu/2020/03/07/coronaviru...

https://www.statista.com/chart/21173/hospitalization-icu-adm...

[Fjolsvith]

> Across the United States, 38 percent of those hospitalized were between the ages of 20 and 54.

> Nationally, 12 percent of the intensive care patients were between the ages of 20 and 44.

Versus 26 percent of the ICU patients being aged 45 to 54, and 62 percent of the ICU patients being 55 or older.

[throwaway5466]

You mean "not worse". If almost everyone got cured there's hardly any leeway to be even better.

[cycrutchfield]

The difference is not statistically significant is what I mean.

[Reelin]

There's a really important distinction to be made here though. "No better than placebo, and half the control group died" would mean it probably doesn't work. "No better than placebo, but no one needed to be hospitalized" presumably means that no medical care was required in the first place. If no medical care was required, then it will be impossible to demonstrate a positive drug effect by definition.

[StreamBright]

I think what you are trying to say is that we cannot measure the difference.

[tigershark]

It’s actually worse than placebo...