I was going to stop posting coronavirus tracker apps but this one looks really useful. Now that we know most infected people aren’t tested, the number of confirmed cases isn’t all that helpful as a metric except maybe to look at trends over time. The number of people in the hospital, on the other hand, is a hard number, and comparing that number to hospital capacity is very useful. This app from the University of Washington does that. It also forecasts future hospitalizations and gives a confidence range (which is quite wide, but there it is to ponder.)

This is by state, which is a slightly big and arbitrary geographic unit. Looking at my home state of Pennsylvania, things look almost reassuring, but then looking at New Jersey, they look dire. It would take me five hours to drive to Pittsburgh, Pennsylvania but I could almost spit on Camden, New Jersey. There will clearly be pressure to move patients across state lines within and between nearby metro areas, and in fact that is already in the news this morning.

The situation in New York looks just awful. I didn’t look at all 50 states, but a quick sampling suggests that states with large cities (and by proxy, probably large hospital systems), and states that started social distancing relatively early, are likely to do a lot better. People might think they would be safer in more rural areas, and perhaps it is true that your odds of infection are much lower, but your chances of survival if you do get infected could also be much lower. This is partially speculation and based on a few anecdotes I have heard, but I do know that this trend holds for car accidents and gun shot wounds.

To this water resource engineer, the differences in capacity use between states and the differences in the timing of available capacity suggest that you could move patients around, or move equipment and medical staff around, between regions in an organized way and save lives. Maybe somebody should get on that if they haven’t already.