Chickenboy -> RE: OT: Corona virus (2/26/2020 5:31:15 PM)
I believe that this virus and disease will have a similar trajectory to that of MERS/SARS some 17 years ago. Coronaviruses are not Orthomyxoviruses (influenza) and act differently in transmissiblity and mutation rates, so again the MERS/SARS analog is probably more useful than H1N1 (1918; 2009) or other seasonal influenzas.
In the case of MERS/SARS, the virus circulated amongst mainland China for some time before diagnosed, a bit longer before there was recognition of it by central authorities and then longer yet before there was even a haphazard and incomplete response. International spread was a feature of the disease. Here's a very interesting and concise timeline from the CDC's website:
To whit: The disease ran rampant in China for a while. Then it spread internationally. Cases that were not laboratory-confirmed were later re-stated (see entry for July 2003) and halved. In between, the overreaction to cases was addressed (de-stigmatization of SARS) and travel advisories lifted after a few months.
A curious thing about the SARS virus was how it disappeared after a year or so. No isolates were recovered post-2004. Poof. Gone. NIH states that the disappearance was due to quarantine and isolation of affected cases and-of course-due to the research money poured into NIH on interventional strategies.
COVID-19 does behave somewhat differently in the human host (replicates higher in the respiratory tract versus deeper in the lungs like SARS) and therefore may spread more readily. By nature of its replication patterns, it will also probably have a lower case fatality rate. That's exempting the exacerbatory effects of other co-pathogens (COVID-19 PLUS influenza, streptococcal pneumonia, TB, asthma, emphysema, other respiratory disease in the same patient) of course. I have not seen *any* efforts to parse the effects of COVID-19 from other co-pathogen effects in infected, but this is likely one of the rationale for making sure to get one's annual flu vaccine ASAP if you haven't already done so.
Viruses spread more readily when they make one sick, but aren't fatal. Paradoxically, the sicker people are with a particular virus, the easier it is to diagnose and ultimately contain. The Los Angeles Times had a nice article on this a couple weeks ago.
What do I think will happen? Something more akin to SARS, albeit with a more significant global spread in the interim. I expect the peak of this will probably not be too far off and that, with the onset of warmer weather in most of the Northern Hemisphere it will abate. I would be surprised if we heard much about this 12 months from now, excepting in cooler climates (you listening, Oz land?) as they slip into their annual influenza cycle.
The economic and social impacts on our increasingly globally connected supply chains are a totally 'nuther question. I'm limiting my observations to the disease cycle and virus longevity in our public per se.
Good stuff, CB. Curious what your thoughts might be on the "mortality rate" aspect of COVID-19.
I read somewhere yesterday that the mortality rate in Wuhan was around 2.4% and everywhere else in China and the rest of the world it was around .7%. I can't site the reference for those numbers, so let's just use them as an exercise. Is it a safe assumption that it's calculated from all the reported deaths and all the total reported infection cases? In other words, isn't it very reasonable that there are many cases that were never officially reported, at least in China/Wuhan, as the individual didn't go to the doctor or a hospital? It seems to me those unreported cases would really cut the official, reported mortality rate. My guess would be cutting the reported rate in half.
Too many assumptions on my part? Am I just way out in left field?
Not at all, USS Mike.
I'm a diagnostician by training and trade. So I need a solid working case definition. "Suspect" versus "Probable" versus "Confirmed" cases mean something very different to me and so do the means at achieving the diagnosis. So the first question I always ask is, "How are you defining a case?" It's the building block of disease interventional strategy and is usually glossed over in the lay press. Unfortunately, it makes all the difference in the world. You would hope to have a diagnosis which is informative, selective and efficient without being too broadly inclusive.
With COVID-19 (or SARS or influenza), do you want a 'case' to be an instance of virus recovery (or PCR positive) from someone? Is that sufficient? A general canvassing of the population regardless of associated illness? Well, that tends to lead to exhorbitantly high numbers of samples being processed and laboratory sample throughput issues. You can't build laboratory capacity sufficient to meet that spike in cases.
Which is why Wuhan changed their reporting measures ('case definition') dramatically a couple weeks ago. A backlog in laboratory confirmation (from 'sick' patients) led to their easing of the case definition to symptomatology. If you have respiratory disease, a fever and/ or CT/MR changes consistent with lower respiratory disease you're a 'case' by the new definition. Hence an overnight 'surge' in cases recogized.
But case numbers based upon self-reported symptomotology are really 'squishy'. They are prone to conflation with myriad other diseases with similar symptoms. Or there may be co-infections. Or they may not be an exact match for what you're trying to measure. Is a person from Wuhan with a severe respiratory disease and no fever a case? How about a person with a fever from Hong Kong with no respiratory component?
If a 'case' is someone that's exposed to the virus, then how many cases are there? Dunno. Virus (PCR) testing has been abandoned as the gold standard at this point. We don't know how many people have been exposed to the virus that haven't gone in for 'confirmation'. I read somewhere that some 15 million Chinese left Wuhan before the 'quarantine' barriers were erected. How many of those had been exposed to the virus. Nobody knows. I think probably a fair number.
I think it's also fair to say that a case fatality rate for an identical pathogen exposure would vary depending on the nature of the medical care sought that may lead to patient survival. Biased comment follows: I doubt the quality of medical supportive care afforded someone in those newly erected Wuhan tent hospitals would approach that of the Mayo Clinic. If your measure is outcomes (death or no death from viral exposure), quality of care variables make a big difference.
Without having a solid grasp on what a 'case' actually constitutes and the numbers exposed that may be aclinical / subclinical, you can't formulate a meaningful 'case fatality rate'. And you can't readily cross-compare case fatality rates with different standards of medicine or baseline diseases of co-morbidity that confound symptomatic diagnoses.
What does my gut say? If your 'case fatality' case definition is: 'people exposed to the virus (and only the virus) that die from the virus (and only the virus)', then that number is very very low. Not zero. But very very low. Other risk factors (additional infectious disease exposure, smoking habits, air pollution, environments that engender person-to-person spread, age of patient, immunosuppressive background diseases, access to antivirals, access to ventilators, etc.) are IMO every bit as important or more important than this virus' unique abilities.