First, the study does not even use the term "mild". They did look at three levels of severity: non-hospitalized, hospitalized, and admitted to ICU. The "non-h

First, the study does not even use the term "mild". They did look at three levels of severity: non-hospitalized, hospitalized, and admitted to ICU. The "non-hospitalized" group surely includes people who would describe their case as "worse than any flu", as well as those would describe theirs as "mild" (e.g. a few sniffles for a few days, like a number of my friends recently.) Second, the hazard ratios differ widely between the three levels of severity. For example, for MACE (major adverse cardiac events) versus the contemporary control group, hazard ratios are: Non-hospitalized 1.26 Hospitalized 2.41 ICU 4.36 One would expect a similar gradation of risk by severity among the "non-hospitalized" group.