at least 3 of those 4 cases are related to being twins. (and possibly the 4th a daft Cat A takeoff profile would be an inappropriate assessment of the risk compared to a take off profile suited to climb into IMC) akin to the risk balance between chance of engine failure on takeoff against risk of collision with unseen object when the H-V curve is avoided.
As far as we can establish no-one has ever had an engine failure in a twin whilst climbing upwards backwards in a 'Cat A' t/o, being exposed to t/r failures and other critical components for longer.
Furthermore a twin profile does not have to result in an uncrashed helicopter merely a good chance of survival. Similar to the outcome of an engine failure for a single ofver a hostile (my arse) urban environment. (we have seen it is not so hostile from the relatively good results posted here).
regulatory required fuel systems so complex that they are more prone to Murphy are a consequence, as are ludicrous (critical component) complex gearboxes.
the stats promised are not in practice delivered (you can't deny it). Twin engined helicopter crashes, flown by professional pilots, are giving helicopters a bad name.