[Trigga*happY]

So can someone tell me why epidemiology looks at total cases vs deaths, rather than recovered vs deaths?

To me it seems a results based metric is more useful than a "well maybe these ones might survive" metric?