One of the lessons we learned from the terror is...we had one of our Ph.D. students do her thesis on friends' grief. We could show that close friends of those who were killed were at the level of siblings when it came to their grief. It went down a bit sooner than for siblings. We identified those. Grandparents have been identified in several studies.
Of course, you need to start somewhere, and I think you start with parents and siblings, and then you have to build on that. It's very flexible, and it has to be flexible for the reasons you gave.
Before I came into the meeting, I saw a couple who lost two children, and they're functioning. Both are back at work. If others had asked if that was possible, we'd say, yes, it is possible for some, but most people will be totally crushed by that. They look favourably at flexible systems with these structures and possibilities built in, but with the length or period, I think you have to base that on what your resources are.
I could tell you that from Swedish research, we see that if there's a sudden death, parents will in general have 10 times as much sick leave, over three weeks, than those who have not experienced it. If you have an anticipated death, it will be five times as much. These are big studies. It is a societal problem that we don't follow up on well enough.