Regional variations- Two Cheers for Practice Variation

Two cheers for Democracy: one because it admits variety and two because it permits criticism. Two cheers are quite enough; there is no occasion to give three.

Edward Morgan Forster (1879-1970), Two Cheers for Democracy (1951)

It’s been 36 years since Wennberg wrote his classic paper on small area variation. Hundreds of peer-reviewed papers based on the Dartmouth Atlas Project have followed. These papers focus on Medicare patients and differences in care and costs in high and high cost environments and low growth and low cost environments and often mention overspecialization, over-use of resources in doctor-rich cities, doctor generated-demand, and practitioners striving to meet “target-incomes.

Dasrtmouth studies conclude intensive use of resources does not produce better care, and indeed the opposite is usually true. The differences, saith Dartmouth, are due to discretionary decisions by physicians unduly influenced by availability of hospital beds, imaging centers, and a fee-for-service system that rewards fast growth and high utilization. For his work and that of the Dartmouth Group, admirers have recommended Wennberg as a candidate for the Nobel Prize, either in economics or medicine.

Yet regional variation persists. Health costs still vary widely between cities, the countryside, academic centers, hospitals, physicians, North and South, East and West.

Why? Many factors share the blame – self-serving doctors, too many specialists, too few primary practitioners, too little standardization, sparse regulation, excess fragmentation, neglect of prevention, negative fee-for-service incentives, lack of integrated groups run by salaried doctors with no incentives to do more. All may contribute. If only we could somehow use data to”homogenize “ care, stamp out “unwarranted variation,” have salaried doctors compete in integrated groups, we could save 30% in costs. And government and health plans could use variation data as a negotiating chip to bring outliers and cost abusers into line.

But quelling variation hasn’t happened. What are Wennberg and his followers missing? Richard “Buz” Cooper, MD, professor of medicine at Penn and senior fellow of the Leonard Davis Institute of Health Economics Institute at Penn, has suggested, sometimes indelicately, that the Dartmouth data is biased to support their point of view., in articles in Health Affairs and elsewhere in peer-reviewed articles, that these variations are inevitable. In January/February Health Affairs, Dartmouth dignitaries, including the editor of Health Affairs, rebut Cooper's arguments on the basis of flawed methodologies.

Cooper argues that Dartmouth may be missing fundamental realities of human nature – that doctors gravitate to big cities to make a better living and to access better educations for their children,, that populations of these cities have profound socioeconomic circumstances, that it costs more to care for the neglected, the newly arrived, the poor, and the uninsured, that specialists on the whole enhance rather decrease quality, and that there is rampant and growing shortage of doctors, and that we ought build more medical schools and create more residency programs.

This kind of talk is anathema to Dartmouth folk, who insist more care, more doctors, more “intensive” use of resources, has not and will not eventuate in better care.

And so the debate about variation rages.

Impersonal data from on high supports the Dartmouth position that profound variations exist. But human nature at ground zero ignores the data. It is a classic top-down bottom-up reality test. By minimizing variation by trying to bring about uniform prices by bringing down costs in high cost areas to those of low cost areas, one is trying to repeal certain fundamental forces of human nature – that doctors will migrate where they can make better use of their skills and can make a good living, that the cost of delivering care varies by the socioeconomic status of those being cared for, and that doctors and hospitals will do what is needed to satisfy the demands their organizations to be community leaders and to satisfy cultural expectations of their region.

There is nothing mysterious about this – practices and costs vary by region, and bottom-up economic forces are more powerful than top-down calls for compliance to meet arbitrary standards. To paraphrase former House Speaker Tip O’Neill’s maxim about politics, “All health care is local.” Unfortunately, Democracy. like health care, is messy. Both could stand improvement. But neither will perfect or overcome human nature.