Not to diminish your loss, but this whole debate is mired in the difficulty in moving from individual outcomes to aggregates that matter for public health polic

Not to diminish your loss, but this whole debate is mired in the difficulty in moving from individual outcomes to aggregates that matter for public health policy. Things got ugly (and will do so again in the future) facing something novel, where many basic assumptions we have about individual results are based on other baseline assumptions that are no longer true. The very expectation of a good surgical outcome for a common stent surgery is based on statistics from "normal" times. But, would the procedure have helped if the surgery was expected to cause COVID exposure, and the patient could have that severe respiratory illness during their surgical recovery? Would it be a good outcome if the surgical staff were dropping like flies with COVID they would get from the regular flow of patients? Alternatively, could the procedure be expected to work as reliably if the staff were wearing all that extra personal protection gear? That is not the conditions under which the procedure was developed and its benefits determined to be worth the risk... I hope that our global experience produced enough data for someone to come up with better answers before the next novel pandemic. But, I don't know how you plow through all the inconsistencies in the data to come to any statistically valid conclusions. I.e. we have different regions/subpopulations who, in effect, ran different arms of an experiment. But, how can we compare their outcomes with sufficient rigor to find clear answers?