Physician Services at the End-of-Life - Are We Now Ready for REDEPLOYMENT?

Los Angeles, California; Manhattan, New York; and New Brunswick, New Jersey, had very high physician labor inputs for both Academic Medical Centers' (AMCs) end-of-life cohorts and the region’s Medicare population. Minneapolis and Rochester, Minnesota (home of the Mayo Clinic), had low regional and AMC physician labor effort. This correlation suggests that the variation observed across regions cannot be explained by differences in health status, adequacy of care, or patients’ preferences but is linked to idiosyncratic patterns of labor input found in both AMCs and their associated regions. So concludes a 2006 study "End-Of-Life Care At Academic Medical Centers: Implications For Future Workforce Requirements", David C. Goodman, Thérèse A. Stukel, Chiang-hua Chang and John E. Wennberg.
[Health Affairs 25, no. 2(2006): 521–531; 10.1377/hlthaff.25.2.521]

The authors also observe that in several of the low-input regions, much of the care is provided by large, multispecialty group practices (for example, the Mayo Clinic) or integrated delivery systems (for example, Intermountain Health Care). What is it about such practices and delivery systems that lower the barriers to palliative care, in fact, may encourage more appropriate use of palliative care for those with advanced AND chronic illness? Just a hunch, but I'd surmise that a key element is a shared or common compensation system among palliative medicine physicians, primary care physicians, and specialists. What might support the design of new compensation systems to promote the sort of physician collaboration essential to this redeployment? The likely prospect of payment bundling , a payment reform which has been advocated as a means of reducing readmissions.
Surely there must be other factors besides compensation incentives at play. I'm curious to learn what other possibilities come to your mind?