Medicare officials have quietly posted on the Hospital Compare website a "cost per Medicare beneficiary" score for acute care hospitals. The data shows a 3.1-fold difference in costs across the country.
Anyone may download Hospital Compare data to see comparative risk-adjusted scores for every hospital. Users may compare the most and least expensive among 3,375 hospitals provided for an "episode" of care, described as a point three days prior to admission to 30 days after discharge for care delivered by any provider paid under Medicare Part A or Part B.
This score "is important, because it complements the quality and outcome measures to be used in assessing hospitals’ performance," Centers for Medicare & Medicaid Services spokesman Don McLeod, said in an e-mail.
Hospital leaders worried
But this new and very public way to evaluate acute care has many hospital leaders worried. Some say the scores, which are based on providers' Medicare claims, aren't fair or accurate.
Moreover, under authority from the Affordable Care Act, Medicare's latest Inpatient Prospective Payment System rule proposes to use "cost per Medicare beneficiary" as a significant chunk—20%—of a hospital's value-based purchasing score to determine incentive payments for discharges starting Oct. 1, 2014.
The proposed rule indicates an eight-month "performance period" that would begin a year from now, from May 1 to Dec. 31, 2013.
While CMS officials said the purpose of the new score is to push healthcare cost transparency to reduce Medicare spending, it's unclear how these cost variances equate with patient outcomes, such as infection rates and mortality.
It's conceivable that higher cost hospitals, rather than providing more expensive high-tech care than its less expensive competitors, may be giving unnecessary, invasive care that can introduce infections, exacerbate conditions, or cause other kinds of harm.
Likewise, a low-cost hospital may have poor outcomes because it is not providing necessary care.
Three-fold variation
The scores are based on the variance from 1, which is the value assigned to represent an average cost per beneficiary. For example, for hospitals with 10 or more cases (the number needed for statistical validity), Indian Health Services Hospital in San Carlos, AZ, is the least expensive with a score of .54. Allegiance Hospital of Midland-Permian Basin, Midland, TX is the most expensive, with a score of 1.72.
That score calculates all Medicare Part and B services incurred in the physician's office, for outpatient care, durable medical equipment, imaging, home health services, hospitalization, and surgery.
Other hospitals on the highest end of cost include Harmon Medical and Rehabilitation Hospital, Las Vegas, NV, 1.59; and Cancer Treatment Centers of America at Southwestern Regional Medical Center in Tulsa, OK, 1.54.
The database includes Medicare claims for care provided between May 15, 2010 and Feb. 14, 2011.
Linking efficiency scores to incentive payments
Under value-based purchasing rules, hospitals will have withheld from their payments 1.5% of their Medicare DRG total by 2015. That money will enter a pool to be redistributed as incentive payments to top performing hospitals.
Patient experience as measured by the Hospital Consumer Assessment of Healthcare Providers and Systems stays at 30%, outcome measures including 30-day mortality are weighted at 30%, and clinical process of care measures weighted at 20%.
So the efficiency score, weighted at the remaining 20%, could mean as much as $200,000 for every $1 million of incentive payments in the pool.
For small hospitals, that could be a lot of cash.
'Unfair'
Cheryl Rayl, CEO of the 48-bed Allegiance Hospital in Midland, the most expensive cost per beneficiary hospital, says she was surprised to see her hospital listed at the high end. Allegiance operates primarily as a 39-bed psychiatric hospital, she says, and rarely do its psych patients require acute medical care, which is provided in its nine medical beds.
"The most expensive thing I do over there (for psychiatric patients who occasionally need medical care is insert an IV," she says. "We don't do surgery. We don't deliver babies. We don't have an intensive care unit or telemetry. This doesn't make sense to me."
'Putting reimbursements at risk'
CMS officials say that these cost efficiency scores are risk adjusted for age and severity of illness, so cancer or other specialty hospitals that treat more severely ill patients are fairly compared.
But the measure has not yet received the endorsement of the National Quality Forum, an approval CMS says its seeks. And some hospital leaders are worried about whether the risk-adjustment and scoring system are fair.
Blair Childs, senior vice president of public affairs for Premier healthcare alliance, a network of 2,400 hospitals and other health providers, says he is concerned that "CMS is putting reimbursements at risk based on an unproven Medicare spending per beneficiary measure as part of value-based purchasing. While the measure in principle has merit, it still has not been tested and can’t be replicated," he said in a statement.
"The end result is a lack of national data that hospitals can use to verify CMS’s calculations, determine the appropriateness of the methodology or analyze true differences in performance."
Concerns over quality
Nancy Foster, vice president for quality and patient safety for the American Hospital Association, said in an interview, "We need to pull (the measure) apart and figure out what it is beginning to tell us. And unfortunately, the downloadable file on Hospital Compare does not contain all the detail one would need to more effectively understand what's going on in each of these communities.
"The question in my mind (for those hospitals) at the low end of the scale," Foster continues, "is are those patients able to access all the services they truly need? Or are we seeing very low costs in some cases because patients aren't getting all of the services they need? I think we don't know."
She adds that since these cost variances are being released for the first time, "we need to kick the tires on these measures." She notes that "It seems to me challenging to assume that a patient could get all the care they need (at one hospital in one part of the country) at a cost that's one third that" at a hospital somewhere else.
"I, for one, am looking forward to the plethora of researchers who begin to slice and dice this data and tell us what they're seeing."
Foster predicts that communities will certainly look at patterns. It may be that a community with lower costs uses rehabilitation services or other post-acute care services as their "secret sauce of success," to get patients well on their way to recovery earlier after hospitalization, she said.
"In coupling these payment measures with the quality measures, I think we'll truly begin to learn where the sweet spot is here," she said. "How do you put together this cadre of services where you get high quality good outcomes with the lowest possible cost?"