Anything to add to it?

I attended a meeting yesterday in Baltimore where we discussed how to go about measuring culture change (i.e., how to operationalize the definition above) and then how we might use those measures to test for a relationship between culture change and clinical quality measures.

To date, we have had single-site demonstrations of success, but no large scale study showing how organizations with the above characteristics produce better clinical outcomes for their residents. Why is that?

In part, it's because our measures of culture change have been lacking. But assuming we had such measures (see above), it's still quite possible that we would not find significant relationships between culture change constructs and all the clinical quality measures.

The fact is, clinical quality is multi-dimensional -- the measures are not very correlated (go to Nursing Home Compare and see how a nursing home can be good at pressure ulcer prevention, but yet bad at pain assessment and treatment). And neither of those aspects of "quality" may be sensitive (in a measurement sense) to things like creating a home mileau, or instituting decentralized management.

Does that mean we should not invest in culture change practices, just because they have yet to be statistically associated (on a large scale) with somewhat problematic measures of clinical quality? We know that culture change practices like consistent assignment lead to lower rates of turnover (and higher resident and staff satisfaction). And we know that higher turnover = lower quality of care. That should be enough for us to find the will to invest in this work.