This is an interview I did on KCBS San Francisco Sunday afternoon. The producer asked me if I could comment on what people who are short of cash should do to survive. So that’s what I prepared for. But then the interviewer, Susan Kennedy, led with a bigger-picture question and got, I think, a surprising response. The whole thing goes less than 4 minutes.
It has antiwar.com on it because my friends there did the work of converting the mp3 into a soundcloud.
READER COMMENTS
Rob Rawlings
Mar 23 2020 at 9:27pm
I’ve read various articles on the Diamond Princess and (with no deep understanding on my part) the evidence seems quite strong. I find that hard to reconcile with the mainly anecdotal evidence from China, Italy and Spain that their health systems are (or were) overrun by Covid-19 cases and that people suffered and needlessly died as a result. The anecdotal stuff leads us to think that had better isolation processes been in place earlier then at least some of this suffering and death could have been avoided.
I live in a city in Southern California with a population of about 50,000 people and which has 2 reported cases (and, for what its worth, no-one I know here is either sick or has reported to me that they know someone who is sick). I got a call today from my medical practitioner today to cancel my annual checkup on the basis that they needed to keep their resources free to focus on the pandemic, which sort of caused my BS detector to register a bleep.
Can anyone with a better understanding of the science and data comment as to whether the Diamond Princess scenario would lead us to expect our healthcare system to get overwhelmed by Covid-19 (and should that be happening already?) and if so, what measures the data from that cruise ship would indicate would be high priority for us to minimize suffering ?
JFA
Mar 24 2020 at 6:49am
I posted this on the last piece about the Diamond Princess: https://academic.oup.com/jtm/advance-article/doi/10.1093/jtm/taaa030/5766334. “The outbreak was traced to a Hong Kong passenger who embarked on January 21st and disembarked on January 25th. After docking near New Taipei City, on January 31, the ship arrived in Yokohoma, Japan. By the following day, the Japanese health ministry ordered a 14-day quarantine for everyone on board and rushed to close its ports to all other cruise ships… Passengers were quarantined in their cruise ship cabins, and only allowed out of the cabin for one hour per day.” Positive cases were immediately evacuated and sent to various hospitals.
Yes, cruise ships are closer quarters than a city, but the cruise’s patient zero was on there for 4.5 days, and at the end of 10 days, 17 percent of everyone on the ship (because of how quickly this was published the authors underreport the number of passengers who eventually tested positive by around 80 people). The authors of the paper estimate an up to an 80 percent infection rate without quarantine. I’m not sure that’s entirely comforting. I don’t know why the Diamond Princess would be your go to example, when you have Iran, Italy, Hubei, and Spain.
I think the Diamond Princess and Hubei (see this: ) demonstrate how effective lockdowns can be.
The question is whether they are effective *in the long term* and whether they are worth the cost. And can a short-term lockdown buy enough time to increase capacity (in what will have to be very imaginative ways) to make the lockdown worth the cost? Reading Marginal Revolution and finding this paper (https://www.nber.org/papers/w26867#fromrss), I am increasing skeptical what the US is doing will be effective in the short or long term. But we shouldn’t be looking at the Diamond Princess for hope.
JFA
Mar 24 2020 at 6:51am
This link showing the effectiveness of the Hubei lockdown didn’t show up: https://promarket.org/wp-content/uploads/2020/03/Zingales2.png
Charley Hooper
Mar 24 2020 at 1:38pm
Here’s the timeline as I understand it.
The infected passenger from Hong Kong was on the ship from January 20 through 25. Passengers on cruise ships are in very close and regular contact with other passengers and the crew. The quarantine was announced on February 4 and may not have started until the following day. That gives 16 or 17 days for the passengers and crew to spread the virus. And then, even after the quarantine was instituted, many said it was done poorly. For instance, a crew member reported that the crew had been expected to interact with passengers even under the quarantine. Even Dr. Norio Ohmagari, top government adviser and director of Japan’s Disease Control and Prevention Center, said that the quarantine process might not have been perfect. On March 1, all passengers and crew disembarked.
If we assume that the quarantine was imperfect, but got better over time, there was probably a three-week period where the virus had relatively unfettered access to the passengers and crew. My conclusion: the great majority, if not all, of the passengers and crew on the Diamond Princess cruise ship were exposed to the SARS-CoV-2 virus.
Phil H
Mar 24 2020 at 6:51am
Rob – I have no deeper knowledge than you, but just some rough figures:
If 20% of people on the ship got the virus and that carries over to your town, that’s 10,000 people infected. 15% of people get serious symptoms, and 0.5-1% die. That would be 1000 people needing to go to hospital within a fairly short time, and several hundred probably needing ventilators. Your local hospital definitely doesn’t have 100s of ventilators. The local health service would crash.
JFA
Mar 24 2020 at 10:47am
It doesn’t quite carry over to a town. A cruise ship is a very closed space where people have lots of contact with others. A town might eventually see 20% infection but over a longer time horizon (how long is anyone’s guess).
Michael Sandifer
Mar 24 2020 at 12:10am
I’m very uncomfortable using the Diamond Princess example upon which to base state or national policy responses. There’re potentially important selection biases, and potentially many important details that are unknown. I think it’s irresponsible to generalize at this point.
Alan Goldhammer
Mar 24 2020 at 9:53am
+1 to Michael’s point. This was a small case study and extrapolation of numbers beyond it is foolish at best and dangerous at the extreme given the infectivity and potential adverse outcome of SARS-CoV-2 infection. There is an article in today’s Washington Post regarding infections on another cruise ship, the Grand Princess. Hundreds of people refused to be tested as they realized the quarantine time would be extended if they tuned up positive for the virus. In this case we don’t have a baseline of infection.
I’ve been tracking all the clinical research and drug development work for a couple of weeks now (I’m retired from the biopharma industry) s well as some of the models. There is a decent paper from Greek and Italian researchers that looked at the Lombardy region data. The identified Day Zero as January 21 and based on modeling from using the lock down date of March 8, they predicted that cases of new infections would begin to drop now. I don’t have Lombardy specific data, but Italy has seen a decrease in new cases for the past three days. If the model trend hold up, Italy should see a complete fade out by the end of may. Of course this is an uncomfortably lengthy period of time to engage in social distancing but it’s the best approach that we have.
Maybe on the of the drug trials shows up some positive information. Personally, we are being done a huge disservice by constantly touting a pharmaceutical treatment based on one or a handful of patients. We are about to see a very large open label trial in NYC of anti-malaria drugs coupled with azithromycin. If this is done properly with good data acquisition, we might find something out.
Most of the trial data that the Hubei researchers have published so far has not been encouraging.
Mark Z
Mar 24 2020 at 7:56pm
It’s likely the best study we have though, so there’s not much else to generalize from. The selection biases in country-level data are almost certainly much worse. Ideally many countries would already have been doing random testing and follow ups. But there are probably some serious impediments to doing that, e.g., low risk people may not want to participate for fear of having to quarantine. At least as far as fatality rate goes though, it seems US data seems consistent with the ‘optimistic’ fatality rate from the cruise ship.
john hare
Mar 24 2020 at 5:55am
@David
I liked the idea of interest loans instead of helicopter dropping cash. Make people think about whether they really need the money and where will they be in a year when the note comes due. Obviously not time in a short interview to discuss the terminally irresponsible that will borrow everything possible with default in mind.
@Rob
Part of the overwhelm problem as I see it is something often mentioned on this blog. Heavy restrictions on medical personnel and their toolkit. One thing often mentioned is the lack of ventilators causing deaths. I design and build many tools in my company and I looked them up the other day. The approved machines are incredibly expensive with computers and monitors galore. However, older units from back when were fairly simple as at heart, a ventilator is an intermittent low pressure/volume air compressor. One of the early ones was powered by a windshield wiper motor. With proper instructions, a competent shop could build one of them in an afternoon. It would require 24/7 attention by a family member or friend as there wouldn’t be all the computerized automatic bells and whistles. Try getting something like that approved.
@Michael
If decisions must be made, they must be made on the basis of available information. The Diamond Princess represents one controlled experiment as opposed to the uncontrolled ones globally. Honest decision makers can calculate allowing for selection biases. That would likely have better outcomes that decisions responding to mob panic reactions.
Shutting down the economy will likely cause as many or more deaths than the original problem. Stress, panic, financial insecurity, and skewed medical availability will have costs. Think of the 1929 market crash suicides by people that still had more resources remaining than most of the population at that time.
Comments are closed.