By now, most people are familiar with the concept of the R0, the reproductive value of a pathogen. It’s the number given to say how many people get infected from one person.
The “true” R0 is based on no interventions. Ideally, with interventions, the R0 decreases until it’s less than one. The R0 also varies, because it is dependent upon the environment. A highly contagious droplet-transmitted disease in a congregate setting like a nursing home or war ship will spread faster, leading to a higher R0. But those settings are not the same as typical community transmission. Most of us don’t live on naval ships.
This is the best image I’ve seen yet describing the R0 of different pathogens. It’s old now, published on Feb 20, 2020, but still accurate.
We won’t know the “true” R0 until the pandemic has subsided and we review all the data across the world. But it can be estimated based on pre-intervention transmission in a number of countries. In early March, it was estimated to be about 2.5 (range 1.5–3.5).
Because we learn more as we go along, about the transmission and biology of the virus, these estimates change over time. One more month into this pandemic and nearly all epidemiologists who have reviewed the data available to date have consistently said the R0 is 2-4. This means that for every one person infected, they can infect between two and four susceptible people.
Different environments will change the R0. A cruise ship has a large number of people in close quarters with limited options to move around. The Attack Rate is the number of infected people in a population of susceptibles. The Diamond Princess, the first cruise ship to experience an outbreak of COVID-19, had an attack rate of 9.5% as of Feb 16, 2020. Zhang and colleagues determined an R0 = 2.28. The rate of contagion was likely higher in cabins where there were more people. By late March, 712 people from the ship had become infected. While the overall infection rate was about 19%, these additional cases would be considered part of the secondary attack rate — infection in contacts to the first group.
I’ve only ever been in de-commissioned naval ships on tours, but the close quarters make this claustrophobic person queasy. Isolation in this setting is pretty difficult.
On the USS Roosevelt, the navy ship now famous for its Captain calling out US leaders, the infection rate was about 13.6%. (This figure is based on the overall figure of 585 infected, not a specific time frame. Some of these are likely to fall into the secondary attack rate.) More than 90% of the crew were tested, and while >45% of those positive were asymptomatic or pre-symptomatic, “only” 13.6% became infected. In such close quarters, if the R0 was much greater than 2.5 or 3, especially given the asymptomatic infections, we would have expected a much higher attack rate. About those asymptomatic infections, these are mostly young sailors on the ship. I’m willing to bet that given their age and their jobs, most are otherwise quite healthy individuals with no underlying medical conditions. This means they’re less at risk of developing severe disease.
It’s also important to note that control measures were taken. Had the Captain not made a plea for help, the overall infection rate may have been much higher. The R0 though, is based on the Attack Rate — that is, how many people does one person infect? We know for measles, one person can infect between 12 and 18 other individuals. In the most high contact environment, the USS Roosevelt, that didn’t happen.
Some studies from China have found a higher R0, focusing on household transmission. Like a ship, people sharing a household are in closer contact to each other than they are to those outside the household. Don’t forget eating “family style.” That is, sharing dish with the same utensils, picking up the same spoon, etc. It comes as no surprise that people living together in close quarters would be more likely to become infected.
Some question the “low” R0 given that SARS-CoV-2 has spread so rapidly across the globe. Remember that Influenza spreads rapidly across the globe every year and has an R0 = 1–2. The 1918 Flu Pandemic was 1.8–2.8, H1N1 was estimated to be about 2, and that’s with some degree of immunity. While H1N1 was a novel virus, it is believed that a similar virus may have circulated some years before and that is why the elderly were less at risk than usual for flu complications. They may have had some partial immunity.
Still, H1N1 was first identified in San Antonio in late April and spread across the globe by June, with an R0=2. With a R0=3, SARS-CoV-2 rapid spread is not surprising.
It’s very possible asymptomatic or pre-symptomatic transmission plays a role in this higher contagion value. We know influenza can be transmitted ~24–48 hours before symptoms appear. It’s very likely pre-symptomatic transmission plays a role in COVID-19 as well. Furthermore, the knowledge that many cases are mild (symptomatic but mild enough to not warrant medical care), may indicate a higher R0 value than flu. So R0 = 2–4 for COVID-19 is consistent with the data to date.
That isn’t insignificant. Those of you of a certain age will remember the 80s Faberge shampoo commercial that described this perfectly.
Here’s one person infecting two people. The incubation period is 2–14 days, but the average is 5 days. So by day 5, those two people infect two more, and another 5 days, those four each infect two more. And then we have a pandemic.
post-script: I am an epidemiologist, I am interested in the biology of disease and controlling transmission in the population. I’m not an epidemiologist modeler. For that, I rely on the experts at Yale School of Public Health and the Harvard TH Chan School of Public Health.