From here on in, stats on numbers of infected people are useless. Now excuse me as I'm off to panic buy five bananas.
However the experts say:
In 2009, we saw that our previous reporting systems were not dependable; they relied on patients presenting to physicians, which is influenced by public alarm among other factors. Initial case fatality rates for H1N1 differed by up to 50-fold . Conversely, ICU admission criteria are relatively fixed over time. Cases and deaths can be easily tracked, making ICUs ideal places for surveillance of severe pandemic influenza. To use this strategy, it will be important that intensivists understand the size of their catchment (or referral) area so that they can accurately estimate the local incidence. The creation of early warning systems was one of the main goals of the International Forum for Acute Care Trialists (InFACT) and ongoing efforts such as the SPRINT SARI study
https://ccforum.biomedcentral.com/articles/10.1186/s13054-019-2616-1
An interesting article although the implications for where we are now are "thought provoking"..