UCSF study highlights difficulty of knowing health care prices
New UC San Francisco research shows significant price differences for ten common blood tests in California hospitals, with some patients charged as little as $10 for one test while others were charged $10,169 for the identical test.
The analysis of charges at more than 150 California hospitals looked at blood tests that are often required of patients, such as lipid panel, basic metabolic panel, and complete blood cell count with differential white cell count.
Hospital ownership and teaching status help explain a portion of the variation - prices generally were lower at government and teaching hospitals. Factors such as location, labor costs, patient capacity and percentage of uninsured population generally did not account for the price differences, the authors said, making it difficult for patients to know their costs in advance and to "act as rational consumers."
The report will be published in BMJ Open on August 15, 2014.
Charges for a basic metabolic test ranged from $35 to $7,303, depending on the hospital; the median charge was $214. The most extreme price difference was found in charges for a lipid panel: the median charge was $220, but overall charges ranged from $10 to as much as $10,169.
The results are of particular concern, said the authors, since there isn't much room for variability in blood tests. Moreover, because the tests are identical across providers, consumers might be expected to think that hospital charges would be similar.
"You may hear people say that, 'Charges don't matter' or that 'No one pays full charges,'" said senior author Renee Y. Hsia, MD, an associate professor of emergency medicine at UCSF and director of health policy studies in the Department of Emergency Medicine. She is also an attending physician in the emergency department at San Francisco General Hospital and Trauma Center.
"However, uninsured patients certainly face the full brunt of raw charges, especially if they don't qualify for charity care discounts," Hsia said. "And as employers are switching to more consumer-directed health plans with higher deductibles and co-pays, the out-of-pocket costs of even insured patients can be affected by these charges."
The blood test analysis was based on charges assessed in 2011 by general, acute care medical/surgical hospitals. The majority of the hospitals were not-for-profit, urban, non-teaching facilities. On average, 41 percent of their patient populations were on Medicare and 25 percent were on Medicaid.
The blood test charges were based on hospital full rates before pre-payments or contractual adjustments.
The scientists noted that they were unable to quantify some factors which could help influence medical charges for blood tests: "For example, many quality hospitals may choose to invest in higher quality facilities, supplementary services and social services than others," they wrote in their report. "The costs of these differences, while of value to patients, are not easy to measure, However, they likely do trickle down into charges for all basic services, including blood tests."
Still, they wrote, few of these factors were significant predictors of the charges patients ultimately faced.
Price variation similarly occurs when patients are hospitalized, according to a separate study also led by Hsia and published in the August 2014 issue of PLOS ONE. In that analysis, hospital charges for an average California patient admitted for uncomplicated percutaneous coronary intervention ranged from $22,047 to $165,386, with a median charge of $88,350. Charges were higher at California hospitals located in areas with higher costs of living, in rural hospitals, and in hospitals with higher proportions of Medicare patients. However, these factors accounted for less than half the price variation, said the authors.
"To expect patients to be rational consumers is unrealistic when the system itself is irrational," Hsia said. "There is very little that we are able to point to that explains the variation, suggesting that the variations are not predictable and therefore a sign of huge inefficiency within the health care pricing system."
The blood test study data were obtained from the California Office of Statewide Health Planning and Development's Hospital Annual Utilization Data files, the Area Health Resources Files, and the Impact Files from the Centers for Medicare and Medicaid Services.