Take radical prostatectomies. Please.
The referral region around Baton Rouge, La., has the highest rate in the country at 4.9 procedures per 1,000 population. Yet the region around Lafayette, La., has the lowest at 0.9 per 1,000.
Likewise, back surgeries: New Orleans and Baton Rouge have among the lowest rates in the country, while Shreveport and Monroe, also in Louisiana, are benchmarked at the highest rates in the country. The variance between them ranges from 1.2 procedures per 1,000 to 7.6-a sevenfold difference.
"It's pretty phenomenal in Louisiana," said Jonathan Lord, M.D., chief operating officer of the American Hospital Association. "It runs the entire spectrum."
And no one has a plausible explanation for these variations. There is no evidence that jurisdictional boundaries mean anything to patients when they seek medical care.
This pattern, Lord said, is seen throughout the data in the 1998 edition of The Dartmouth Atlas of Health Care. The Atlas, now in its second edition, charts the disparities in the quantity of healthcare services and spending from region to region across the U.S. It also looks at the implications for cost control and public health. It is published and sold by the AHA.
Clinicians' preferences, not patient needs, have resulted in the variation of utilization, Lord said. He said he believes if patients are given more choice, the variations will normalize.
That was Lord's theme in a speech at the Louisiana Hospital Association's annual meeting in New Orleans recently. "Most of us believe that what we do in our communities is absolutely normative," he told an audience of hospital managers and executives. "Yet you can drive 30, 40, 50 miles and find a practice pattern that is radically different. Why?"
If there's one thing the Dartmouth Atlas makes crystal clear, Lord said, it's that the consequences of higher supply are usually higher utilization. Louisiana has some of the highest hospital bed capacity in the country. Most of Louisiana is in the top quintile of beds per 1,000 residents.
And that tends to push up hospital utilization, especially at the end of life. "Where you live determines not only the kind of care you're going to get but also where you're going to die," Lord says. "Geography is destiny in our business."
In Alexandria, Lake Charles and Shreveport, as many as 50% of Medicare beneficiaries will die in the hospital. In New Orleans, 33% to 36% will.
Not surprisingly, Louisiana has some of the highest per-capita reimbursements by Medicare in the last six months of life among all the states.
"It's very hard to make a value judgment on these statistics," Lord says. "There is some opinion that dying in a family or home situation may be better." The issue of care at the end of life garners high interest from the public. Generally, people prefer less institutionalized settings of death.
One finding from the Dartmouth Atlas is when patients know about the variety of treatment options available and the associated risks, they often opt for less-intensive therapies than their physicians would have chosen for them, Lord said. Patient participation in care decisions reduces not only utilization of resources, but also practice variation, one of the goals of the medical quality movement.
There are some larger lessons to be drawn from these data, Lord said. If Americans truly want to expand access to healthcare and insurance coverage, they must eliminate more waste. Such wide swings in practice patterns are nothing but wasted resources, he said. The biggest cost reductions are to be had in getting people the right care in the first place.
Lord suggests several strategies to apply the data constructively:
Benchmark your community against national norms.
Discuss with your community and physicians the reasons for variations.
Look beneath the surface to see what's driving variations.
"The Atlas is not an answer book," Lord says. "It's a question book for starting a dialogue with your community."