Why hospitals aren’t cutting costs nearly as much as they could (in one chart)

by Karan Chhabra

The more you read about healthcare delivery, the easier it is to get discouraged about slaying the healthcare cost monster. Absent huge and damaging price cuts, there aren’t a ton of quick fixes. But as you get closer to clinical care, patterns of waste emerge. Those seeing patients every day know them best. But it probably takes a push to get them questioning the dogmas of their practice. Enter this paper (gated) on a cost reduction project from UCSF’s neurosurgery residents, prompted by a hospitalwide incentive program for residents:

OBJECT: Given economic limitations and burgeoning health care costs, there is a need to minimize unnecessary diagnostic laboratory tests.

METHODS: The authors studied whether a financial incentive program for trainees could lead to fewer unnecessary laboratory tests in neurosurgical patients in a large, 600-bed academic hospital setting. The authors identified 5 laboratory tests that ranked in the top 13 of the most frequently ordered during the 2010–2011 fiscal year, yet were least likely to be abnormal or influence patient management.

RESULTS: In a single year of study, there was a 47% reduction in testing of serum total calcium, ionized calcium, chloride, magnesium, and phosphorus. This reduction led to a savings of $1.7 million in billable charges to health care payers and $75,000 of direct costs to the medical center. In addition, there were no significant negative changes in the quality of care delivered, as recorded in a number of metrics, showing that this cost savings did not negatively impact patient care.

CONCLUSIONS: Engaging physician trainees in quality improvement can be successfully achieved by financial incentives. Through the resident-led quality improvement incentive program, neurosurgical trainees successfully reduced unnecessary laboratory tests, resulting in significant cost savings to both the medical center and the health care system. Similar programs that engage trainees could improve the value of care being provided at other academic medical centers.

In a nutshell, each year UCSF offers its residents about $400 each if they can achieve the goals of a quality improvement project that they design and execute. Their neurosurgery house staff, working with their hospitalist service, found that certain lab tests they ordered rarely affected their plan of care. So they designed a protocol describing exactly when those tests ought to be used, advertised the protocol and the incentives, and within a year cut those tests’ use by 47%–without compromising their patients’ quality of care. Considering that it’s just one service with just 18 residents, the cost savings are staggering:

ucsf cost savings

Amazing results. But think about who really won here. The hospital saved $75,000 in lab costs. The payers saved $1.7 million in billable charges. There’s almost no comparison. That $1.7M is actually a loss to UCSF, because if they’d billed it, they would have gotten much (if not all) of it back as revenue. Think about that for a second, because it’s key to understanding much of what’s wrong with American healthcare: this amazing project, which saves gobs of money without hurting patients, probably causes UCSF immediate financial harm.

UCSF is not the VA, or Kaiser Permanente, or the British NHS. It has just one ACO contract for San Francisco city employees, so it’s not a true accountable care model either. For organizations like UCSF–in this sense, for the vast majority of American hospitals—if they don’t do something, they can’t bill for it. Which means they don’t get paid.

Fortunately, UCSF is a forward-thinking institution. I’m sure its leaders are looking to the future and realizing that they’ll have to encourage many more of these projects to make it through the decades ahead. UCSF is an academic institution, too, where quality improvement competitions serve a valuable educational purpose. But let’s not forget what reality is like for the hospital down the street hawking its proton beam facility to patients with prostate cancer. When they save the system money, they lose.

We have reason to believe this will change. As episodic bundling and accountable care spread across the country, hospitals’ incentives will come closer to encouraging what the system needs as a whole. But we need to make sure these progressive payment models aren’t confined to Medicare, and are embraced by enough payers to bring hospitals to the tipping point where lower charges actually are a financial win. Till that happens, feel-good stories like UCSF’s will remain just that.


Karan is a student at Robert Wood Johnson Medical School and Duke graduate who previously worked in strategic research for hospital executives.

Follow him on Twitter @KRChhabra or subscribe to the blog.


10 thoughts on “Why hospitals aren’t cutting costs nearly as much as they could (in one chart)

  1. Aquifer says:

    Another example of why we need Single Payer – as long as charges are mainly in the providers court, there is no real incentive to lower them – in fact, a perverse incentive not to …

  2. Ashwini Batchu says:

    Aquifer, even with a single payer system (be it a universal healthcare plan or private insurer), the hospital would not see returns since they’re not the ones paying the bill. The difference is more fundamental in a fee for service model–misaligned incentives. Whoever is paying (insurer) wants to pay less, whoever is charging (provider) wants to charge more.

    The only example that comes to mind for bucking this trend is Kaiser, where they’re both the payer and provider. This creates a new challenge: how to deliver quality care, without over-treating patients, to decrease costs and increase profit?

    In order for a separated payer/provider system to work, there would have to be incentives for hospital cost reduction efforts built into the payment plan… enter value-based purchasing.

  3. Scott Brody says:

    Would be interesting to know how they arrived at the $1.7 million. Is it based on actual billing procedure or does it simply multiply the foregone volume by HCPCS-level charges from the hospital chargemaster?

    Don’t disagree with the conclusion of the study, simply curious as without digging into the actual foregone reimbursement at the payor level, this might overstate the detriment to the hospital. Particularly with an inpatient-heavy service such as neurosurgery, wouldn’t much of the lab work for a patient be bundled into the DRG payment for a case?

    Would be interesting to see this study performed with a different service line or a more costly subset of potentially extraneous tests (say perhaps imaging)

    • Good question. My reading is that they’re from the chargemaster — which I agree is discounted somewhat when it gets to private payers. Here’s what the paper says:

      To assess the financial impact of any laboratory reduction on the institution, direct laboratory costs and billable charges per test were calculated. The direct laboratory costs represent those expenses incurred to perform the laboratory tests and are inclusive of items such as reagents used in the clinical laboratory, while the billable charges are the costs for the reimbursement agencies, and are designed to cover other hospital expenses incurred, such as facility costs and phlebotomy staff. … Taking into account the direct costs and billable charges attributable to each of these tests, the initiative resulted in a reduction of direct laboratory cost from $137,809 to $63,640, and billable charges from $3.6 million to $1.9 million.

      As I mention in my response to Tony below, private payers don’t rely heavily on DRG-based payment. So while the actual impact to payers may not be as high as the paper states, I think the underlying conflict of interest remains unchanged: healthcare providers have little-to-no incentive to reduce the volume of care they provide under FFS.


  4. Bob LaPorte says:

    the Carolina eHealth Alliance took a different, successful approach regarding tests ordered in the ED, and hit the same barrier. Savings to payers of $1 million a year, costs to the providers to run the data interchange $250k /year for a great ROI.


    From the article: “A meeting at MUSC on Oct. 23, which will include hospital leadership, physicians, private insurance companies and the state Medicaid agency, has been called to discuss the program’s future financing.”

    Perhaps if they come up with a successful approach, they’ll share it. 🙂

  5. I may be missing something, but billable charges says nothing of reimbursement. The hospital gets paid, for most cases, based on DRG. So, for a neurosurgical patient admitted for a spinal fusion, the hospital gets paid based on that DRG, not based on the individual tests done.

    Therefore, for patients in whom this is the case, the hospital loses money for each additional test ordered. They will get paid X dollars (based on the DRG) either way.

  6. whitingnoise says:

    Tony’s comments are accurate – with the exception that not all payors reimburse based upon DRG’s. Billable charges do not reflect reimbursement, that depends upon the payor mix, (Contracts w/private insurers, M’caid, M’care, etc.). You are correct is stating that the hospital may be hurting itself, but to what degree cannot be determined without a really detailed analysis. Most hospitals do not even have cost accounting systems to determine true costs.

    We did a similar exercise in our hospital system, but we focused upon the tests, (lab and imaging) performed on inpatients, that could have been avoided or done in an outpatient setting (reimbursable). Depending upon your payor mix, we were heavily M’care/M’caid, you are really saving costs or gaining reimbursement without loss of revenue.

    Retired H’care Exec.

  7. Brad F says:

    Tony has it right. Bundled reimbursement via DRG for most patients.

    What might need stating is overtesting leads to false (+)’s. Use your imagination at what comes next….

  8. Brad F says:

    On private payers btw, most negotiate per diem or bundled rates (“non-DRG”). MCOs dont pay line item bills.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: