It's a bit of a tricky one. To answer your question on research into how the built environment impacts the behaviour of folks with dementia, that's been done. There's a growing body of it out there that's currently available and looks at the intersection between behaviour and the built environment.
Where it gets a bit trickier—and I think you've actually hit on a weakness that we see across the board—is that there's a lack of research that makes that second leap, looking towards the tie-back with built environment and the actual hard data on workplace injury rates. We can make a secondary leap by saying that fewer responsive behaviours equal fewer incidents of violence, but actually making that direct leap is a bit tricky.
I would say that we can have the same problem when we run into the research looking at models of care. There's a growing body of research that looks at how different models of care impact the quality of the client or patient experience, but how that impacts the workplace safety of the care providers is often a component that's missing. This is actually a piece that we recently went into provincially—just because the body of research is a bit thin on that side—in looking at whether or not the person-centred care model, which is associated with stronger quality of care outcomes for dementia care clients, actually translates into fewer incidents of violence.