Peter Drucker wrote “If you don’t measure it, you can’t manage it.”
The current COVID-19 situation suffers from a lack of credible measurements. For example, we do not seem to have accurate statistics on current cases nor know how many people have already had stealth cases that were never reported, which measurements of antibodies could answer. The early emphasis is on testing only those with deep and critical symptoms, likely caused by an extreme shortage of testing kits. Unleashing the biomedical industry to provide such kits has started to radically change this.
Engineers and scientists, including the information theorists among our colleagues, are extremely good at setting up experiments and estimating variables from noisy measurements. I believe that many physicians are dismayed with the state of our ignorance. With the right information, we might be able to answer important questions, such as:
- Why has the Chinese infection rate plummeted to near zero, if that is true?
- Why has Italy been so hard hit – is it merely a late response, some aspect of social behavior or another factor?
- What is the true reinfection rate (R)?
This situation should be fertile ground for the techniques of analyzing big data. But before it can be validly analyzed, it must be accurately measured in order to predict future impacts.