As we roll through the fourth week of the lockdown here in Seattle, trapped in our homes and pretending to homeschool our children, it seems reasonable to start thinking that we might have turned a corner. But this is the exact moment to reflect on why things have improved and to understand that we are not going back to normal any time soon. What comes next is about one number, R0, the reproduction number.
R0 refers to the number of new cases caused by each person infected with the disease. As you may recall from earlier posts, we need to keep R0 under one to avoid another explosion in cases. Everything we have been doing and everything we need to do going forward is focused on that challenge. R0 depends on three things: 1) the manner in which the virus spreads, 2) the number and nature of encounters a case has with uninfected people, and 3) the probability that any person they encounter has immunity. Estimates of the R0 for SARS-CoV2 range from 2.0 up to 26, but most settle near 2.5. Until we have a vaccine, the only thing we can control is the number and nature of encounters.
Hence, the lockdowns. And they seem to be working. A British study reported a 73% reduction in personal contacts and an 84% reduction in physical contacts after the lockdown, enough to drop R0 well below 1.0. The data came from questionnaires, which have a dubious reputation among epidemiologists because they rely on the honesty and accuracy of recall of respondents, who might be inclined to overstate their adherence to social distancing. A remarkable Italian study released yesterday gets around that by using cell phone GPS data to track not only the movement of people after the lockdown, but also the number of times each phone was in close proximity to another. They found a 50% reduction in the average range of movement following the lockdown and a 47% reduction in personal contacts, suggesting that either the Italians were less capable of social reserve or the Brits were more likely to lie. A 50% reduction in R0 would reduce it, but not below one. A model based on German data suggests that a 90% reduction in social contacts is necessary to control an outbreak and a delay of five days in implementing once an outbreak begins will result in a failure to prevent a local epidemic.
But most of us have dramatically reduced not only our number of contacts, but also, through distancing, handwashing, and barriers, the likelihood that any given encounter could transmit the virus. And it’s working. So, what next?
To move forward, we will need a few things:
- A cheap universally available antibody test. We still have no good idea how many people have been exposed and who they are. Testing will tell us who needs to worry and who doesn’t and how contagious this disease really is. Those who have been exposed may be candidates for taking on high risk jobs that are putting the lives of others in danger.
- A better understanding of infection control. Without widely available, inexpensive testing, it is hard to understand what strategies truly prevent the spread of disease. We will not be able to return to work without those measures.
- Algorithms to determine who is at highest risk. Most of us will survive an infection. But some have a very high chance of dying. We need to understand the level of risk for each individual and offer protection in accordance with risk.
So, we may be starting to turn a corner, but don’t throw a party just yet. That success is only happening because of the measures in place. We have many more corners to turn until we have a vaccine.