• Josh Schiffer

March 23, 2020

Lower the death rate. I am happy to report that this concept has legs. A number of investigators in Seattle, myself included, are putting together proposals to get trials off the ground and funders are interested. I will feel great relief if multiple trials can be done across the globe testing all promising agents including at minimum remdesivir, favipiravir and hydroxychloroquine. I know that in the right hands, these challenging trials can be done properly so that the data can be trusted. Fingers crossed.

In normal times, I see patients and write equations about herpes viruses and HIV for a living. It is valuable and fun science but plodding and usually under the radar. Our results are of interest to niche communities. We have developed a couple of ideas (most recently T cell anti-proliferative therapy for HIV cure) that were worth selling to the scientific community. However, there has never been any real urgency. The methods to convince the field were standard and familiar: publish solid papers and give convincing talks.

I am currently in unfamiliar territory. I am convinced that the strategy of early home testing and treatment is a vital and is a currently overlooked piece of the “dance” portion of the epidemic curve eloquently described by Tomas Pueyo in my last blog. I have next to no experience performing advocacy work. It is awkward for me to stand on the hill and scream, but I am doing it because I feel like lives are in the balance. As a favor, if you find the idea compelling, I beg you to share this essay widely as I try to do the same.


An American nightmare. This is a recap article about the failures of the system to generate enough test kits at the early stages of the epidemic. My friends and colleagues are again quoted about the early days of the epidemic in Seattle. All of this has been described in earlier posts but worth a read. Like the Wuhan slide stack I included several days ago, I think this a story of historical importance:


Disaster in Italy. Italian physicians are pleading with providers in other countries to figure out a way to provide home care to keep the hospitals manageable. They fear that the unimaginable scenario where hospitals contribute to the epidemic by facilitating infection of uninfected patients and health care workers, who then bring it home, happened in Italy. From the sky news video below, it is not hard to envision this being true. I was almost in tears reading this.

Yet, home care is not so easy. For people who are sick enough to require oxygen, it is next to impossible. Available clinical reports suggest that the transition from only a little sick to very, very sick occurs rapidly with COVID. Therefore, COVID patients with some degree of respiratory compromise cannot realistically be observed outside of the hospital. The best method of homecare would clearly be early identification and treatment of the infection. I will repeat my mantra: lower the death rate.


Tons of potential drug options here which is great. It is sort of cool that Haldol (Vitamin H) and metformin are on here. I still remember the doses of these winners from residency. Again early treatment is critical:

------------------------- Faviparivir. This drug is on our minds because of reports in the Guardian that it performed well in Wuhan. The first publication came out today. The investigators treated early!! (during the first 7 days) head to head versus kaletra (an HIV drug which does not appear to be the ticket for COVID). They showed more rapidly improving CT scans and elimination of viral shedding in the faviparivir arm. This seems promising. BUT, the study is not randomized, it is in an engineering journal, and there are other key missing details. Plus, the drug is licensed by a Japanese company (Fuji films interestingly). However, in this study, the manufacturer is listed as a Chinese company. I definitely do not know what to do with this information, but we are desperate. I hope this agent makes the list of drugs we can test properly in an RCT. The flow of information is distinctly not normal. ------------------------------

Atul Gawande makes the point that with thoughtful and broadly implemented policies, health care workers should be at low risk of infection in the hospital. Costco may be another matter.


An alternative to flatten the curve? My vote is a strong NO. Here, David Katz advocates for allowing herd immunity among the least vulnerable (the young) while isolating those at risk. With this approach those spring break partiers in Florida would be at the vanguard of public health policy. The author is coming from a good place and any rational person shares his concern for the economy under physical distancing. However, the concept that the elderly can be exclusively placed in a bubble while the rest of society carries on and gets infected to develop herd immunity, seems pretty naive to me.

On a somewhat related note, this is a very cool article that shows the flow of people underlying pandemic spread on a global level starting in Wuhan. Just imagine scaling this graphic down to your own city: moving dots would capture employees entering and leaving nursing facilities, people over the age of 65 visiting restaurants and spending time with their families etc… We would need to severely limit all of these flows to protect the most vulnerable. Intuitively, it seems impossible to pull off

Preventing the second wave is going to be ever so difficult. A rapid re-emergence of cases occurred in Hong Kong after relaxation of social distancing. This is depressing but expected. Managing the dance, as described by Tomas Pueyo will require enormous foresight and diligence (not in abundance in the oval office presently):

This is a nice description of what the process could look like:


Tracking the epidemic: A friend sent me this graph which makes it seem like New York is on the highway to hell. Obviously, the situation is not good there. To hear about a venerated institution like Sloan Kettering running out of PPE is simply beyond comprehension. However, be careful interpreting this type of data. Several reminders:

1) I am following the data closely in WA and still see no evidence of flattening the curve. In this graph and others that I have seen, the scaling of the x & y axis relative to one another can be used to make things seem as they are not.As of today, the epidemic in Washington still appears to still be scaling up and if the scale were 0-5000 on the y-axis on the above plot, the Washington line would look steeper and more frightening. In fact, here it is:

2) New York is doing South Korea level testing which is one explanation for the higher number of cases. The other is that a lot of people live in New York (see below). Washington is testing a bit less so state to state comparisons are hard in that respect. New Jersey and Ohio which probably have terrible outbreaks are testing very few people. Do not interpret graphs like this as the true number of cases in a state. Extra modeling is needed to estimate these numbers which are always higher than the number of confirmed cases. The degree of undertesting determines by how much we are underestimating state level infection.

3) Also consider what these curves might look like per capita.I suspect that if the true number of cases (unknown but can be estimated) per state were graphed per capita, that we would see 50 lines with similar slopes but different starting points. This would be the fairest way to project the data in a comparative fashion and to see if the rate of spread differed according to certain key variables (rural vs urban, widespread testing, policies, etc…). I am assuming that this is being done in real time my smart people somewhere.

4) Flattening the curve will take at least 1 and probably 2-3 weeks to see after physical distancing policy is implementedso it is not worth looking at the NY data for improvement until at least next week: expect ~50k cases in a week or so if testing remains consistent and do not interpret this as a failure of social distancing.

5) Sadly, deaths will rise for weeks even after the epidemic slows down because it takes several weeks from infection to death and hospital care may tragically be less effective than usual. Nevertheless, the death rate is not reflective of the pace of the epidemic. Do not interpret low death rates now, as a success of social distancing.Do not interpret high death rates later, as a failure of social distancing.Keep your eye on the ball. Follow incident cases per day at the state level to assess our current policies.

6) See the Stats article about hospitals in Italy possible adding fuel to the fire and keeping R0>1. This concept is just so incredibly grim and terrifying. The very valuable “canary in the coalmine” perspective of the brave Italian health force most certainly needs to be validated with hard data and hard modeling. Yet, my personal perspective is that the conclusions of soldiers on the ground are often right on point.

I remember as a resident when we started seeing CA-MRSA pneumonia and severe soft tissue infections in the MICU. These were unprecedented in their severity, and targeting of young, previously healthy people. We all were thinking, huh? Two months later, great articles started emerging about the CA-MRSA epidemic.


Rand Paul does not meet the ethical standards of a doctor. I have no words for this. He either lied about not being symptomatic or had his doctor order a COVID test because he was not practicing social distancing. If he indeed had symptoms and was working out in the senate gym, lunching with fellow senators and attending meetings, then this is criminal negligence. Many of his colleagues are in high risk groups. Just an unfathomable level of incompetence and disregard for the well-being of his friends and colleagues. If he was asymptomatic, then why on earth was he not serving as an example for his constituents and practicing even basic social distancing, particularly while his test results were pending? First do no harm you jackass.


Home testing. Seattle people, please consider enrolling in this study! This is amazing and is the leading edge of the sword for long term containment strategies:


Bless Tony Fauci. Great quote: “ I know, but what do you want me to do? I mean, seriously Jon, let’s get real, what do you want me to do?


And finally some local news that feels like an early, cruel April fool’s joke:

It appears that coronavirus has infected our local bridges and that West Seattle will resemble Gilligan’s island for the next 6 months. Even though there are zero people on the roads, the traffic is going to be terrible. Quintessential Seattle. It sounds like the bridge has been collapsing under its own weight for some time. I really do not know what else to say.

1,073 views1 comment

© 2020 by Joshua T. Schiffer | Proudly created with

  • Facebook Clean Grey
  • Twitter Clean Grey
  • LinkedIn Clean Grey