Share this article on:

POLICY WATCH: Why Hospital Prices Vary What Neurologists Should Know

Butcher, Lola

doi: 10.1097/01.NT.0000432689.08794.e8
No caption available

No caption available

Why does Hahnemann University Hospital in Philadelphia charge $234,913 to treat an inpatient suffering intracranial hemorrhage with major complications while Uniontown Hospital in Pittsburgh charges $9,539?

And why does Las Palmas Medical Center in El Paso charge $8,607 for Level 2 nerve injections in the outpatient setting when Central Maine Medical Center in Lewiston bills just $275?

For that matter, why does Palm Springs General Hospital charge just $16,370 for inpatient treatment of a degenerative nervous system disorder diagnosis when Doctors Hospital — located less than 15 miles away — charges $53,501?

Chalk it up to America's strange health care system. In testimony to a US Senate Committee in June, journalist Steven Brill, who examined medical bills for a special issue of TIME magazine earlier this year, explained that the marketplace for health care services does not work the way other markets work. “In a functioning marketplace, prices are based on something that it is explainable — whether it's the cost of producing the product, the laws of supply and demand, or the quality of the product,” he said.

Not so in health care, where hospital officials acknowledge that their charges are not tied to any of those things. Rather, they say their “chargemasters” — official price lists — are vestiges of the way the US health system has developed over the past several decades. The charges are irrelevant because almost no one pays the actual charges, they say.



But as the nation's push to improve the value — defined as quality divided by cost — in health care gets stronger, the call for transparency about health costs is growing. In May, the Centers for Medicare and Medicaid Services (CMS) released a list of the average 2011 charges for the 100 most common Medicare inpatient services at more than 3,300 hospitals. Shortly thereafter, CMS released data on what those hospitals charged for 30 common outpatient services.

Indeed, a greater move toward transparency and accountability are the driving forces behind the “Choosing Wisely” initiative, a campaign by the American Board of Internal Medicine Foundation in collaboration with consumer groups including Consumer Reports, to rein in the skyrocketing costs of medical expenditures projected to comprise 25 percent of the Gross Domestic Product by 2025. The AAN, one of nine specialty societies participating in the effort, released its list of procedures and processes to question in February. (See Neurology Today's article, “Choosing Wisely: Five Things Neurologists and Patients Should Question,”

The high cost associated with many procedures is often not attributable to the professional fees but rather to the facility fees — including the costs of tests ordered — that hospitals might factor into their charges.

According to an analysis of expenditures reported in a 2007 National Ambulatory Medical Care Survey, conducted by the Centers for Disease Control and Prevention, $13.3 billion (95% confidence interval=$10.1-$16.5 billion) was spent on tests ordered at neurological visits. The highest expenditures for procedures billed were for magnetic resonance imaging (accounting for 51 percent of total expenditures at $7.5 billion), electromyography (comprising 20 percent of expenditures at $2.6 billion), and electroencephalography (8 percent of expenditures at $1.1 billion), according to a paper published in May in the Annals of Neurology.

However, in the analysis, the authors offered this caveat about the data: The survey “offers limited clinical detail on the circumstances surrounding why specific tests were ordered, limiting inferences about the value of the tests/procedures identified here. Without such details, we cannot conclude that any test or procedure is overused; nor can we exclude the possibility of underutilization in appropriate clinical contexts. Rather, these data simply establish which tests/procedures account for the most resources.”

While hospital executives maintain that the Medicare data releases cause confusion because charges are meaningless, the public is proving hungry for information. In the week after CMS released inpatient charges, the data were downloaded or viewed more than 200,000 times, said Niall Brennan, director of the Office of Information Products and Data Analytics for CMS.

David van der Goes, PhD, a health economist at the University of New Mexico who studies the cost-effectiveness of neurological treatments, believes that, even though the charge information is not immediately useful to patients, the data release may lead to changes simply because it is prompting people to ask questions.



“I think this is a first and important step in helping people realize that we don't know what health care actually costs,” he said.

Back to Top | Article Outline


By “cost,” Dr. van der Goes means what it costs a hospital to provide the care for, say, a stroke patient. That information is currently not available to the public.

Rather, the CMS data sets show two numbers: 1) A hospital's average 2011 charge for services covered by Medicare for all discharges in a given diagnosis-related group (DRG) such as stroke; and 2) the average Medicare payment plus the patient's copayment and any third-party insurance paid to the hospital for that DRG.

The CMS data show that the relationship between a hospital charge for a given diagnosis and the payment it receives is inexplicable. For example, regardless of what the hospitals charged, Medicare paid Palm Springs General Hospital $7,240 for that degenerative nervous system disorder diagnosis and it paid $7,260 — just $20 more.

That is because hospitals set their own charges, but CMS pays using a fee schedule that is unrelated to those charges. Medicare starts with a payment based on an expert panel's assessment of the resources that hospitals in general use to treat patients with a given diagnosis, John Ney, MD, a member of the AAN Medical Economics Committee, explained to Neurology Today. Then Medicare adds to that payment if a hospital has certain structural factors, such as a high proportion of low-income patients or a residency and graduate medical education program.

Thus, Medicare pays different amounts to different hospitals for a given DRG, but the differential is not related to the quality of care, patient outcomes, or a hospital's actual costs of providing the care.

Although Medicare is the single biggest payer for health care services in America, it is not the only one. Private insurers negotiate the rates they pay to hospitals; in some cases, private insurers pay a discount from a hospital's charge, and in others, they pay a certain percentage over Medicare's pay rates.

Dr. van der Goes, the economist, said neither charge nor payment rates provide the most important information, which is what it costs one hospital to provide care versus another. Until that information is available, he said, no one will know which hospitals truly provide the most high-value care.

Back to Top | Article Outline


Contrary to the way most markets work, the $225,000 difference in charges for stroke care between Hahnemann University Hospital and Uniontown Hospital does not suggest that a patient would fare better in one place over the other.

For one thing, that $225,000 is irrelevant to what either hospital actually got paid. The CMS average payment for stroke care with complications to Hahnemann was $23,126, while the average to Uniontown was $9,810.

For another, the link between the amount paid for services and patient outcomes is not yet clear. Dr. Ney, a Fellow in the Comparative Effectiveness and Cost Outcomes Research Center at the University of Washington and medical director of Surgical Neuromonitoring Associates in Tacoma, said studies seeking to understand the relationship between payments and quality are mixed.

“It's difficult to say that the more you spend on the patients, the better the outcomes or the better the quality of care,” he said.

Beause of the widespread interest in hospital charge data, some patients are likely to ask their neurologists questions. For example, patients may ask their doctors whether they should go to the most expensive hospital in their area to get the best possible care. Dr. Ney advises that neurologists should reassure their patients that what a hospital changes does not reflect its quality of care. A government website — — allows patients (and their physicians) to compare the quality of care of various hospitals. Stroke is the only neurological condition on that website so far, but the site does show rates of hospital-acquired infections, patient experience scores, and other information that may help users decide which hospital is best for them.

Back to Top | Article Outline


To see what the hospitals in your community charge for various diagnoses, check out the dataset available at

  • Click on the “Research, Statistics, Data and Systems” tab.
  • Scroll down to find the “Statistics, Trends and Reports” heading.
  • Click on “Medicare Provider Charge Data” to access the data file.
  • Download the data into an Excel spreadsheet, and click on the tab labeled Top_100_drg.
  • Click “enable editing” at the top of the screen. That will allow you to use the filter embedded in the header of each column.
  • The filters allow you to select a diagnosis-related group (DRG) and the geography (city, hospital referral region, or state) to see what each hospital charges for a diagnosis — and what the Medicare program pays for that diagnosis.
Back to Top | Article Outline


•. Burke JF, Skolarus LE, Callaghan BC, et al. Choosing wisely: Highest-cost tests in outpatient neurology. Ann Neurol 2013:73:679–683.
    •. Medicare Provider Charge Data:
      •. Outpatient Prospective Payment System Provider summary:
        •. Choosing Wisely campaign:
          © 2013 American Academy of Neurology