While precise data on the COVID-19 crisis are rather difficult to come by, and are often of rather suspect quality, there are some things that we seem to have learned. The virus is likely considerably more contagious than your typical seasonal flu, with steeply rising mortality rates above age 60 when other health conditions are present. Quite paradoxically, the virus seems to be a mere nuisance for most people who contract it and we don’t have any real understanding of why it becomes a serious health threat for a minority of people.
In recent days, the mayor of Los Angeles has repeatedly warned on his nightly pep talk that LA is the next New York – give it a week and the city medical resources will be overwhelmed. Yet, one has to wonder if the risks really are similar. The city of New York has a population density of nearly four times that of Los Angeles – and Manhattan has over 75,000 population per square mile. New York runs on public transportation which most certainly increases the average number of people with whom an infected individual would come in close contact, in comparison to LA where most travel to work alone in their cars. Density most certainly plays a substantial role here, and in conjunction with the relative risk level of the population, should allow us to map areas where there could be significant outbreaks with dire consequences.
We started with block group level estimates of the expected number of hospital discharges per 100,000 population for the total of respiratory, circulatory, and endocrine problems using our HealthCounts data. While this gives an adequate picture of the “at risk” population in a relative sense, it does not factor in the density of population. We settled on mapping the number of hospital discharges expected “normally” for the population – which is in effect a population count weighted by the prevalence of risk factors — then divided by the area to get a risk density figure.
The results mapped using a consistent national scale clearly show why the New York area was a likely candidate for a widespread outbreak. There are areas of very high risk in all of the largest and dense cities like Chicago, San Francisco, Boston, and Miami. It is in the mid-sized cities where the relative risk varies significantly. Detroit and Indianapolis have limited areas of high density, and Phoenix, despite its relatively older population, is a dispersed city with lower risk of widespread contagion. Portland has, despite its much smaller size, significant areas of high risk.
We again invite researchers, hopefully with some epidemiological experience, to utilize our datasets without charge in the quest to quickly tame this virus that we might minimize both the loss of life and the increasing economic and social distress that many of us are now bearing.
Wishing all both health and peace of mind during this time as we all carry on as best as we can.