Report from Penn Epidemiologist Stuck in Northern Italy

This is from my Epidemiology colleague John H Holmes who is right in the epicenter of the Italian outbreak — he nicely describes some of the limitations of available data and gives good reasons why not to over interpret the daily incidence data coming out but instead not lose sight of the big picture: Greetings from Pavia, Italy, where the pandemic still marches on. I hope that what I'm writing about today doesn't sound preachy or mansplain-y. I certainly don't intend it to be either. But the message is rather sobering- I wish it weren't. ---The dynamics of COVID-19 Hopes here (expressed by leaders and the media) in the short decline in incidence over the past days were dashed yesterday, where the slope climbed a bit. We are in this for the long haul, and I hope people don't over-interpret these changes in incidence from one day to the next. The epidemic curve clearly indicates that the dynamics of the pandemic are quite nonlinear. The R-naught, a measure of contagion, is still probably about 2.3, meaning that a single person can infect 2.3 people (on average), and each of those 2.3 people can infect another 2.3., and so on. ---It's not just the epidemic curve There is another really important thing to understand about the count data, too, and that is when you hear about so many thousands of cases, or death- these are typically *cumulative* counts. They don't take into account the number of "recovereds"- people who have lived though the infection- or deaths. In epidemiologiuc analysis, both of these are referred to as "removed"- that is, removed from the susceptible and the infected groups. They are no longer in the counts of currently infected, and don't contribute to the epidemic curve. So when you see that we are edging up to 500,000 confirmed cases, those are not active infections. Unfortunately, we don't know if survivors of the infection are immune, so it's possible that they could be in the susceptible group. ---Testing and reporting rates influence interpretation An apparent slowing of disease spread can be due to many things, not the least of which is ascertainment- which is completely bound up in the testing rate. In other words, the counts completely rely on the testing not just being done, but reported. And we know there is a lag in this, due to how the swabs are batched at collection sites, how the analyses are being run in labs (also batched- it's much more efficient to run a bunch of swabs at once), and then getting those results into a database from which statistical analyses can be run, interpreted, and finally reported. It takes time, and many of these steps occur in bursts of activity. So, take these day-to-day changes (increases or decreases), or no change, with a grain of salt. It's the overall slope that matters, as does the more complicated interpretation for which you need to dust off those calculus textbooks. ---"Flattening the curve" We still don't know if the curve is being flattened- it takes *at least* two weeks to see that, and there's still so very much we don't know about this particular virus. Consider- we don't know if infection confers immunity, or how long the virus can infect after being on various surfaces (we know it survives for hours or even days on certain materials, but we don't know if it can actually infect after that time), or what the effects of mutation (which is happening) are, or if there will be second or third waves of outbreaks. All of this is just conjecture right now. ---Social responsibility The moral of all of this takes us back to what I said about the long haul. Italian Prime Minister Giuseppe Conte and the Consiglio were very smart to extend the emergency here to July 31, with increased constraints on activity and corresponding increases in sanctions. The new decree also provides regional governors the ability to increase even further the constraints, with the approval of the Consiglio. This is sort of happening in the US, with some governors taking the initiatives that the federal government is so gun-shy about. The best we can all do is to continue to embrace and practice a stay-home life style, going out only in times of absolute need and when we do, practicing social distancing and impeccable personal hygiene. We are looking at weeks, perhaps even months, more of this. Stay well (and inside) everyone! — say a prayer for John and all of those who are in the middle of major outbreak areas — including NYC — and don’t let your guard down wherever you are — be safe and smart!

 

©2020 by Covid Data Science. Proudly created with Wix.com