ITAKLinus
Posts: 630
Joined: 2/22/2018
From: Italy
Status: offline
|
quote:
ORIGINAL: Chickenboy
quote:
ORIGINAL: ITAKLinus
quote:
ORIGINAL: Chickenboy
quote:
ORIGINAL: Canoerebel
That's another drop in daily mortality for Italy. Good news, if trending (but last week the numbers flattered only to deceive).
Mortality is a lagging indicator of the leading edge of the infection 'wave'. Those that die from this virus and its complications were infected some weeks ago. I too want to see the mortality drop off, but I'm more interested in the waning of 'new cases' to track the progression of the virus through a given population.
Also, I've been following Japan more closely. They've got one of the oldest populations on the planet and they've had comparatively few cases to date, in spite of widespread testing. As far as I'm concerned, they're the champs at social distancing and 'flattening the curve'-a model for others. I'd like to hear more about how they've successfully stemmed the tide so far.
Of course, they're an island. Relatively xenophobic culture. Social distancing is their normative behavior anyways. No PDAs or touchy-feely in public. Perhaps that's the key to all this.
There is a major flaw in this reasoning. It relies on a number, the new infections, that is not really reliable.
Deaths are much easier to count and are "certain" by definition (unless the population starts to massively hide dead bodies, something I find quite... unlikely).
Now, it's probable that the number of infected people is grossly underestimated and that it reflects more the capability of a system (region/city/country) to test its citizens. Basically, if I have 10.000 positive people per-day and I can do 3.000 tests, I will never ever be able to track/test all of the 10.000.
On the other side, deaths are very easy to track even if the information we receive is delayed.
I suppose another important indicator is the number of people who got accepted in intensive care units. Still, also there, there can be under-tracking due to the fact that some people don't reach the hospital etcetc. Also, since the number of ICUs is finite, we can have that the number represents more the capacity of the hospitals, rather than information about the spreading of the virus.
I'm a strong proponent of deaths as the most reliable information. Secondarily, people admitted to ICUs and hospitalized.
You raise some good points. But if you want to use deaths (post awareness) as the most reliable indicator of successful implementation of an interventional strategy then you again need to look at Japan as a best-case scenario.
Their first case was identified on January 21. Their total deaths attributed to COVID-19 stands, as of today, at 41, with 235 recovered. Granted, their histogram of new confirmed cases has increased over the last 30 days, but its still less than 100/day. I understand that their testing is not as widespread as, say, South Korea. So maybe the 'other shoe has yet to drop', but it's worth looking at their (intrinsic?) social distancing as a means of preventing BOTH initial cases as well as the inherent mortality associated with those cases. Could the much ballyhooed South Korean response have benefited similarly? It bears more discovery.
I also assume from your comments that you mean people admitted to ICUs and hospitalized that *also happened to have COVID-19* (read: confirmed cases of COVID-19)? Tracking raw seasonal ICU visits / hospitalizations without the rationale for why they were hospitalized assumes too much causality to COVID-19. You will also be left with data confounders such as other respiratory diseases, age of the population at large and accessibility of hospital beds / ICU visits.
It is tempting to follow mortality from COVID-19 cases as a tracking measure. It's always been harder to 'hide the bodies' than it is to hide/ignore/understate/not find/not look for confirmed disease cases. But I'm old enough to remember the stigma associated with the AIDS epidemic in the 1980s. Death certificates ('cause of death') were routinely restated to avoid the social stigma of having a loved one associated with dying from AIDS. As a result, 'official' causes of death of Pneumocyctis carinii or Kaposi's Sarcoma or Cryptosporidiosis or other atypical diseases common in terminally immunosuppressed AIDS patients dramatically increased.
My father died from idiopathic chronic interstitial pulmonary fibrosis in 2017. He had a gradual decrease over about 5 years in his ability to breathe and oxygenate. His lungs effectively were scar tissue with insufficient lung function. "Idiopathic" means 'uncertain etiology/cause'. We worried a lot about him during a typical seasonal flu for the years he was sick, as any little respiratory ailment would have thrown him over the edge. He died from a massive heart attack. Had he instead picked up a fatal pneumococcus bacterial pneumonia or seasonal influenza or COVID-19, what should his death certificate had read? I believe that, in spite of his history of his disease (that disease is incurable and only goes 'one way') had he died in hospital today and be tested positive for COVID-19, that that's what would be on his death certificate. Unless there was a social stigma associated with it, in which cause of death verity may have come through.
I was reading about a man in Thailand that died in hospital with COVID-19. He was only in his 50s. Of course, he was originally in the hospital for Dengue. But I guess that got subsumed in the mortality reporting. We'd have never heard about his death unless he tested positive for COVID-19. And, by its mere presence, it became the presumptive cause of death.
So I believe that too many cases of co-morbidity are being wrapped into "COVID-19 mortality". Parsing out the details and sticking to some global system for ascribing official causality is far too cumbersome to already overworked hospital staff. At least with an RT-PCR test result, you have >95% chance of a "real" test result (positives are positives and negatives are negative). I don't think the same verity in the details is in the reported mortality for COVID-19.
I was making precisely the example of Kaposi's sarcoma and HIV/AIDS a couple of evenings ago. I agree with your point. Indeed, I'm quite into the way of counting Italy has adopted: counting deaths WITH coronavirus. Compared with the average death ratios for a given period, it tells roughly the delta due to cov-19, net of statistical variance.
In a situation where you cannot credibly track the infected people because of a lack of capacity in testing, I suspect that the indicator of "people died WITH coronavirus" compared with the historical death ratios is a good estimate of the additional deaths coming from the virus.
In some countries there are weird ways of assessing the death cause (your Kaposi's sarcoma example and mine regarding lung cancer in Ukraine) and Germany is definitely counting the people died OF coronavirus, instead of died WITH it as we do in Italy.
As I was mentioning in a post earlier, even if we take into consideration the differences in the lifestyle between Italy/Spain and Germany, it's still quite a huge difference.
Counting the deaths OF coronavirus, I think Italy still has to reach the number of 20. It's quite a big difference with the hundreds of daily casualties we have.
Given that, as you say, "hiding the bodies" isn't practical (or logical...), I see the deaths WITH cov-19 as the only reliable data together with infected people in ICUs. The total amount of cases, albeit potentially a much better indicator, is simply not even remotely accurate. For example: who tells you that the increase we experienced yesterday in Italy is the true increase and not just reflecting some kind of bottlenecks in the testing capacity?
When I started seeing the guys managing the emergency in Italy saying that the infected people can be maybe 20.000 or 30.000 more, I decided it wasn't reliable at all: even if we take the lower estimate of 20.000 untraced cases, it's quite a big difference over a confirmed amount of infected totalling at roughly 50.0000.
Dengue example touched my heart, though. I'm still recovering from the little friend called dengue I experienced 10 days ago.
PS. I'm fairly sure that Spanish healthcare system is doing the bodycount precisely as we are doing in Italy, while Germany (and for what it matters Czech Republic), have gone in the opposite direction. Spain is looking very, very bad currently.
_____________________________
Francesco
|
Printable Version
RE: OT: Corona virus - 
New Messages
No New Messages
Hot Topic w/ New Messages
Hot Topic w/o New Messages
Locked w/ New Messages
Locked w/o New Messages
Post New Thread