Just looking at the cause of death percentages misses the point. Imagine your dream scenario has been achieved and all deaths are caused by errors. 2 hospitals that treat the same number of patients (with similar complexity across the patient populations) are performing similarly on the death metric, 100% of deaths are caused by errors. Never mind that Hospital B kills 100 times as many people.
Putting it another way, without some information about what good performance is, you'd better not make any conclusions about whether this is a good thing or not.
Just looking at the cause of death percentages misses the point. Imagine your dream scenario has been achieved and all deaths are caused by errors. 2 hospitals that treat the same number of patients (with similar complexity across the patient populations) are performing similarly on the death metric, 100% of deaths are caused by errors. Never mind that Hospital B kills 100 times as many people.
Putting it another way, without some information about what good performance is, you'd better not make any conclusions about whether this is a good thing or not.