To encourage someone to do better I did a rough approximation of #Covid19Ireland percentage outcomes of the 9292 new cases arising from 17 July to 23 Sept with outcomes taken from Oct 7th data. Far from perfect but
Hospital 5%
ICU 0.6%
Death 0.9%
11% of hospitalised went to ICU

Oct 7th as an outcome date is over counting hospitalisations but undercounting deaths. But these are ballpark approximations & incidentally match with research that 2 in 3 cases was missed in March/April

Added a new section to this optimised to capture hospitalisation outcomes by reducing case->outcome period to 5 days. That significantly drops hospitalisation to 3% but the previous 0.9% for fatalities is likely to be more accurate because that outcome takes longer

Part of my motivation was to use this calculation to point out that 10,000 a day by Dec 31st would be 310/day to hospital & 31/day to ICU. Given average 2 week stay massively overwhelming the system long before Dec 31 https://twitter.com/StuartDNeilson/status/1315913925385388034?s=20

Don't overly focus in on that as a date prediction. Point is that 10k cases a day anytime will have overwhelmed our healthcare system to the point that a lot of additional people will die on both Covid and non Covid paths. To reach 60% infection - 10k a day for 300 days

The reality is that both hospitalisations & ICU demand would be a lot higher at 10k/day as incidence in older & vulnerable groups would explode with no effecting testing & tracing to locate disease. It would get everywhere whereas thats not true of period data is pulled from