Phase 2, ready or not, here it comes

Yesterday the Italian prime minister announced the timeline of the loosening of the lockdown. Some manufacturing restarted yesterday, and some customer-facing businesses will gradually reopen between next Monday and the beginning of June.

Since the start of the lockdown 51 days ago, it was clear that the condition to “reopen” was to have in place a “test-trace-isolate” plan to find infected people as soon as possible after their contagion, trace their contacts, and isolate them. For this, one needs an infrastructure for large-scale testing with quick turnaround, a well-staffed agency to do manual tracing or an app to do it automatically, and facilities to isolate people who are infected and not in need of hospitalization.

None of this has been done. There hasn’t been a sufficient ramp-up of testing capacity; as far as I know no additional people have been hired and trained for manual contact-tracing; there is an app for digital contact-tracing, but the plan to adopt it appears to have been shelved; if people test positive and are well they are asked to stay home, potentially with their family members, who are free to leave as they please.

All our eggs are in the basket of social distancing. The humor site Kotiomkin posted “Basically, phase 2 will rely on everybody’s common sense. We are fucked”.

The problem is that social distancing requires a common-sense avoidance of close contacts with other people, and the government can give guidelines on how to achieve it, but eventually it has to rely on everybody’s sense of responsibility. Unfortunately, the national mood around regulations is to immediately look for loopholes.

For example the initial lockdown measures stipulated that one could leave home to go buy groceries. Then when the police would stop people tens of miles away from their home, people would say “I drove here to buy groceries”, “but we are thirty miles away from where you live”, “yes but here the groceries are better”. So a subsequent amendment stipulated that one could buy groceries only in the town of residence, making it hard for people living next to a town border, for whom the closest grocery store was across the town line. In fact, every time I have encountered a crazy Italian law or regulation, and asked around for the likely reason it was instituted, it was usually to close a loophole in a previous regulation, which in turn had been put in place to close a loophole in a third regulation, and basically it’s loophole-closing all the way down to Roman law.

I think that, from this point on, the story of the Italian covid-19 epidemic will not show the future of the rest of the Western world, but will evolve in its own timeline. Meanwhile, there are a couple of lessons that are still relevant, particularly in the comparison between Lombardy and NYC, which continue to track each other remarkably well.


One is that, at one point, it was decided to move older people with mild cases of covid-19 from hospital to nursing homes, to open up beds in hospitals. Since the personnel of nursing homes are not trained in the safety procedures for infectious diseases, and since nursing homes host older, frail people who are the highest-risk category for this illness, the result was a huge number of deaths in these nursing homes. Now nursing homes in New York are being asked to take covid-19 patients from hospitals.

The other is that an analysis of all-cause mortality shows a spike in March such that the difference between the typical March all-cause mortality and the March 2020 all-cause mortality is much bigger than the number of confirmed covid-19 deaths. Now the same phenomenon is being observed in New York City, with numbers similar to Lombardy’s. Notably, the baseline all-case mortality rate in Lombardy is much higher than in New York City (because population growth has stalled and the demographic skews older), so while the absolute number of additional deaths is similar, the relative increase is a much more dramatic 6x in NYC versus roughly 4x in Lombardy.

4 thoughts on “Phase 2, ready or not, here it comes

  1. Thanks for your continued updates. What is the y-axis in your plots? Is it the total number of deaths? Are the populations of the two places about the same?

  2. Yes, in the first chart it is daily death and in the second it is cumulative. Lombardy has a population of 10 million, NYC of about 8.4 million. The reason for looking at deaths is that, even though it is an undercounted parameter, it is not as undercounted as the number of confirmed cases, which mostly depends on how many tests are performed.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s