Federally Mandated Postings of Standard Charges by Louisville Hospitals Are Unusable for Their Intended Purpose.

(But will reveal the the unacceptable and unjust absurdity of how we pay for medical services.)

Reporter Gilbert Corsey of WDRB was, to my knowledge, first on the block locally to take public look at the implementation of a newly enforced federal law requiring hospitals to publish their Standard Charges online. Originally part of the Affordable Care Act (a.k.a. Obamacared) as turned into regulation last year,, the stated purpose was to allow the public to compare the cost of services and choose wisely among hospitals before they incur responsibility for payment. An overlying expectation (?dream) was to improve quality and decrease costs. Hospitals bitterly protested implementation of this law.

Mr. Corsey’s reporting verified the expectation that charges amoung neighboring hospitals can vary greatly. For example, an uncomplicated birth at one Louisville hospital was priced twice as high as another, and an injection of a drug used for prostatic cancer varied threefold. Corsey’s report also concluded that the published lists are confusing and difficult to decipher. I agree. I will go further and argue these lists are essentially worthless for their intended consumer purpose – surely knowingly so. Their value however for unintentionally making the policy point that, like pricing for pharmaceuticals, hospital pricing and billing exists in a logic-free, Alice-in Wonderland zone to the detriment of the public. Allow me to explain.

Examination of Hospital charges in 1994
I admit to being chagrined not to pick up on this matter sooner! In fact, one of my very first health policy research studies was a shoe-leather examination of the posted prices of common hospital services mandated by the health care reform efforts of Gov. Brereton Jones in the early 1990s. A copy of an otherwise unpublished paper I wrote describing the effort has been on this website for some time. I literally traveled to all nine of Louisville’s acute care hospitals (there were more individual hospital entities then) to compare the prices of the 39 mandated services to be posted on the wall in their admission areas. It was an interesting experience. Some principal conclusions included:

• Prices varied considerably for virtually all the items, but no one hospital had all the highest or lowest prices. Baptist Hospital East had by far the lowest charges overall and University of Louisville Hospital had the highest. Pricing for an imaginary but possible admission in which 25 of the services were included totaled 37% higher at the most expensive hospital compared to the least expensive.
• There was little evidence that posting these numbers made any difference. In some hospitals the admitting personnel did not know where the list was and could not recall anyone else having asked to see it. In one hospital, the framed list had fallen behind a filing cabinet (where I helped find it) but no one had noticed for weeks.
• Novice healthcare finance person that I was at the time, I recognized that hospitals can define services differently or in other ways make direct comparisons with other hospitals unreliable.
• When the program went on to a second year, I resurveyed the hospitals. No one will be amazed at prices pretty much all went up, but it was clear that the increases were unrelated to the actual cost of providing the services or product. For example, every lab test at one hospital went up the same percent. The hospital lobby was sucessful in terminating the project. No one in Frankfort looked at the data submitted until I did years later.
• The publishing of this information was embarrassing to at least one institution, but the one with the lowest overall charges was proud that their conscious effort to keep charges low was being recognized. (I was discouraged by a University administrator at the time from publishing my analysis with the imprimatur of my membership in the UofL Center for Health Service and Policy Research.)
• For at least a handful of items, the information had potential to meaningfully inform a potential patient. For example, the charges for a screening mammogram were considerably different among the hospitals where the quality of the screening was likely similar.

21st Century not as useful as 20th Century data.
I thought it would be fun, even if not informative, to attempt a similar analysis using the new electronic postings of hospital charges. Sadly, if I had a serious illness and wanted to use this information to facilitate a cost-effective treatment, I would have died first! No shoe leather was necessary, but my computer mouse-hand suffered acute repetitive use syndrome. Sadly, my exploration of the 21st century iteration of a price transparency initiative arrives at exactly the same conclusions as above – this time in spades.

I started this article in my inherent geeky way with numbers, tables and statistics. I quickly realized I would never be able to finish in a timely, brief, or perhaps even understandable way. For those interested in a more technical description of my survey, I may attach an appendix later.

I looked for the postings at 6 of our local hospitals or systems. These included.

Baptist Healthcare Louisville
Norton Healthcare (One list for all hospitals.)
Jewish Hospital
Sts. Mary and Elisabeth Hospital
University of Louisville Hospital
Clark Memorial Hospital (Indiana)
University of Kentucky Hospital (Lexington- added for comparison)

New results.
I was able to find the data I was looking for without too much difficulty – generally somewhere on the hospitals’ billing and payment pages. [See links at end of article.] I initially could not find the charge posting on the UK website and when I called the help number listed on the financial information page, that person did not know what I was talking about. I requested and was promised further information but received none. I subsequently found a link from another site that had reviewed at the posted charges of the hundred largest U.S. hospitals. Their conclusions are the same as mine.

Overview.
Even after finding our local “Charge Masters,” and using a sophisticated database program, they were difficult or impossible to interpret. The average citizen with a desktop workstation and access to a word processor or spreadsheet could probably unpack and view the numbers but they would be essentially meaningless. A smart phone or tablet would be inadequate to the task.  The sheer number of items was huge.  The files provided were typical database files with a minimum of two data columns– one with a very short description of the product or service, and a second with the standard charge. (A few of the hospitals had a third column containing their own unique identifier code which would be of no use to anyone else.) Although the file sizes themselves were smaller than a modest 2 MB, (about the size of a large digital photograph) the files contained huge numbers of individual line items. For example, both Jewish Hospital and Norton described over 75,000 items! The charge data for the Norton Healthcare system was not presented as a data file but as an individually selectable sequence of 1560 screens, each with 50 items to page through. A Microsoft Word version of the Jewish Hospital list would fill 1921 standard pages. None of the hospitals provided a sort capability or search portal on their websites. The very short descriptions of the items are heavily abbreviated and technical. For the average citizen, the descriptions would be as so much Greek. I tried to look for specific items but failed most of the time. (For me they were as if in Latin – the letters readable but requiring frequent use of a dictionary!)

The KentuckyOne Health hospitals included additional elements for each item that would allow some grouping, including some CPT codes and revenue codes that are standard in healthcare billing but not easily sortable by category. For the typical consumer user, these codes would also be so much Greek. The University of Kentucky Hospital provided additional supplementary information labeling individual items with CPT billing codes or DRG (Disease-related Group) codes. These were presented as different tabs in an Excel spreadsheet. Searching for individual items would require looking through different tables and knowing where and what to look for– but kudos to UK for making at least a discernible effort to make their data usable! None of our local hospitals came close.

Even knowing what to look would not have solved the problem entirely for a potential user. For example, in the data from Jewish Hospital, only 50,657 of the 75,054 individual code descriptions listed descriptors were unique. That is to say, many of the descriptions were entered identically numerous times, sometimes with different charges for the same item. This difficulty presented itself in varying degrees across all the hospitals.

Although the whole point of the HHS initiative was to enable comparison of hospitals, allowing each hospital to decide for itself what data to select and how to present it makes comparison extraordinarily difficult and time-consuming– if not frankly impossible. For example, when I tried to compare the 75,054 item descriptions from Jewish Hospital to the 26,413 at Baptist, only a single charge description was identical! (It was a drill bit.) Making the same comparison between Jewish Hospital and University of Louisville hospital identified 9065 identical item descriptions. This degree of overlap probably reflects the fact that until recently, KentuckyOne Health managed both hospitals. It is noteworthy that even for identically labeled items, the charges at UofL Hospital were much higher– often many-fold– than Jewish. For as long as I have been looking at similar data, charges at UofL hospital have been the highest in the state.

Expiration-by dates?
The University of Kentucky Hospital guarantees that their charges will be fixed until June 30 (except for supplies and drugs). I did not see any such declaration at the other hospitals examined.

Redo of 1995 thought experiment.
I had planned to replicate my hypothetical admission of 1995, but even an approximation would take me days of effort, if possible at all. It is my studied conclusion after several days effort, that for an individual seeking financial information about planned elective medical services, that these charge disclosures are useless. In any event, for the individual with acute medical needs, when has discussion about charges ever entered consideration? As it happens, the Kentucky Hospital Association website has been disclosing cost and volume information about hospitals for some time. I understand this is also a requirement of a different federal law. I used this information in my series of articles about angioplasty abuse in Kentucky. For example, in 2017 the average charge for DRG 247, the most commonly performed coronary artery angioplasty (without complications or comorbidity) can be seen to vary by hospital. These range from a low of $61,482 at UK Chandler to $125,511 at Baptist Louisville. The KMA site is worth checking out.

Not only can average charges be compared on the KMA site, but the numbers of discharges are also listed. There is a presumption that hospitals with greater experience offer service of higher quality, but this is by no means a guarantee. In fact, I offer in this discussion that the user should start with the presumption that there is no correlation of charges with quality. Volumes can vary significantly. The number of discharges in 2017 for the DRG 247 mentioned above range from a high of 341 at Jewish to a low of 20 at the University of Louisville. These hospitals have overlapping medical staffs and I would be surprised and disappointed if their quality was meaningfully different for this service. Hospitals performing fewer than 20 of a given service are not included in the KH a website data at all. The sad experience in Kentucky is that for at least two hospitals there was good evidence that many coronary angioplasties were not medically necessary in the first place. A judgement of medical necessity or quality cannot be divined from these lists of standard charges and yet those are the parameters that should– in a perfectly informed world– rank first for consideration by a potential patient. Although discussions and determinations of quality and value are intended to reshape the clinical and financial landscape of healthcare, such valuations are generally difficult to make and probably not yet ready for prime time.

There is at least a fragment of good from this.
Although in my opinion useless for an individual seeking to make a wise financial decision regarding their personal healthcare, the mandatory publication of these “Charge Masters” makes at least one valuable contribution to the public. Much as the pricing and reimbursement schema for drugs disguises any number of shenanigans in the pharmaceutical and health insurance industries, so does the publication of standard charges confirm the absolute opaqueness of how health services are billed. What your hospital bill will look like is predictably unknowable, and those with the least ability to pay will see the highest bill. No wonder that perhaps the major factor driving our rise in personal bankruptcies is healthcare expenses.

Do charges matter or not?
I was told back in 1995 by a high-ranking public healthcare official that I was wasting my time because hospital charges did not matter because no one pays full prices. That argument continues to be made by the hospital industry and others who lobbied very hard against today’s mandated charge disclosure law. This is the same argument made by a predatory pharmaceutical industry– that the prices they charge for drugs are essentially meaningless because no one pays the full price. It is true that many people do not. It is equally true that those without employer-based health insurance, or public insurance, standing between them and industry are in fact presented with the bill for the full amount. There are other reasons why charges are relevant from a public policy perspective but the individual or family facing or placed in bankruptcy is not concerned with public policy at that time.

Real-life examples.
All of us with personal health insurance who receive explanation- of-benefits letters should be familiar with medical charge tomfoolery. For those who have not studied such bills, I offer below two real-time examples why the hospital industry argues that publication of their standard charges would not be helpful to most people. (I am forced to agree with industry in this regards!) When I mentioned to a friend that I was writing about this, they showed me their bill for a commonly performed abdominal operation in a Louisville Hospital last summer.   The outcome was successful, there were no complications, and the patient was more than satisfied with the quality of their care. The itemized bill submitted to the patient’s insurer contained 66 separate items, mostly labeled with standard billing codes that could be used to make comparisons– if a usable comparison database was available. The items were for drugs, lab tests, injection charges, recovery room ($3111), observation charges ($903), and the like. Some significant charges were mysterious such as “Sterile Supply,” otherwise un-coded for $3636. All of the charges for this one-night stay totaled $60,993. However, because of its contractual arrangement with the patient’s insurer, the hospital received only $6911– amounting to only 11.3% of its charges. Because the patient’s insurer was Medicare, Medicare itself paid 80% of the allowable bill and the patient paid the remaining 20% copay of $1364. Medicare’s rules roll all of the hospital charges up into a single standard DRG payment for the specific operation. For payment purposes, all 65 of the other charges were immaterial and were zeroed out. Other payers have the privilege of arranging their own contractual agreements with any individual hospital– or not. For Medicare, the payments are knowable as are the deductibles and copays. For private insurers all bets are off. Wouldn’t you rather have Medicare working for you. What would the presented bill have been if my friend did not have any health insurance at all? Probably rather shocking. Note also that the bill did not include any charges for physician or other professional services which migjht or not have been charged at in-network or higher out-of-network rates.

Outpatient example.
I have a personal example of how Medicare paid for an outpatient stay last summer. I had to stay overnight an excellent large teaching hospital in Providence RI– one with a Leapfrog Grade-A Safety Score. I was having a recurrence of a nearly life-long heart arrhythmia on the way home from a holiday– the first in 20 years. I received excellent care from residents, nurses and physicians, albeit on an uncomfortable stretcher. The hospital itself itemized 15 separate charges (of which Medicare allowed 4). The major item in the charge-list was $3438 for the CPT code for an “ER visit for a problem with significant threat to life or function!” The hospital billed me for $5230, and was paid only $694 by Medicare itself. My AARP Medicare supplement insurance covered my copay of $176 for a total of $867 to the hospital. For its trouble, the hospital received 16.6% of its totaled but immaterial charges

Physicians charge too.
I was billed separately for the services of physicians, the largest amount by the physician in charge of my “life-threatening” condition who billed $867 but received $177; a radiologist who billed $35 dollars for reading my chest x-ray and received $12; a cardiologist who billed $24 for reading my EKG and received $9; and the physician who billed for hospital observation on my discharge day who charged $110 and received $60. I do not believe I received any special financial treatment because I am a physician, nor would I have wanted any. When I saw the bill months later, I felt sorry for the physicians and even the hospital. I did not think their original charges were excessive, especially when I hear about the hospital bills of others for ER and outpatient services. I have written before that a medical insurance structure that does not reimburse providers adequately to cover the cost of their services is doomed to fail. This does not diminish the responsibility of medical providers to provide only the amount of care that is medically necessary, to do so in an efficient manner, and not to tolerate the abuses of other providers.

Why Now?
I find it ironic that the Trump administration has chosen to begin now to enforce this part of the Affordable Care Act! I thought its promise was cut the ACA out “root and branch.” I see a parallel in one of the administrations other initiatives to reduce the cost of medical care– that is to say require pharmaceutical companies to disclose the prices of their drugs more forthrightly. If anything, the pricing and reimbursement system of drugs is more opaque and subject to abuse than any other part of our medical system. However I am not alone in thinking that the proposed disclosure regulations are little more than window dressing with little impact on payment or profits. The administration has to be seen as doing something!  I do agree that prices should be transparent at every level if only to point out the intrinsic unfairness of a system in which every individual seems to pay a different price for the same drug; is unaffordable or bankrupting to most of us; and is a system in which the governing law and policy were written by industry itself through its paid minions. With some luck and public attention, the obvious absurdity of pharmaceutical and other medical charges will precipitate the major transformations that are needed.

Chargemasters: No relation to reality.
There is no doubt that in general, hospital standard charges (a.k.a. Chargemasters) are inflated and bear no recognizable connection to costs. This bizarre and unconscionable practice has repelled outside efforts to make hospital charges more realistic because the inflated charges are the starting point for all other negotiations with insurers and patient alike. Hospitals must like them for some reason!  Without the protection of third parties or government, “standard charges” form the basis of copays and deductibles, insurance premiums based on a percent of charges, and the amounts amounts sent to collection agencies, or reported to credit bureaus. There are probably other reasons I do not know about. [Anyone want to help us?] In its early years, city and state reimbursements to University Hospital through the QCCP program were based on full charges, perhaps explaining why UofL Hospital’s charges have been so high historically. Having make-believe charges also obscures the ability of any of us to understand how our healthcare market is structured and financed and therefore easier for abuses or frank fraud to occur. Of course charges matter!!  The more fortunate of us simply don’t have to notice as we glance over and discard the voluminous paperwork that arrives after every insured medical encounter. The impact is on the less advantaged among us or those denied participation in our multiplicity of third-party payers. It is being said that a major driver of cost of our medical system is increasing prices, not volume or technology. To the extent that high prices keep medical expenses up we are all paying more in our premiums and taxes. From an ethical and philosophical perspective, we tolerate a system that violates the central tenant of health insurance: To spread the risk of the few among the resources of the many.

Peter Hasselbacher, MD
Emeritus Professor of Medicine, UofL
President, KHPI
Jan 23, 2019

Links to Postings of Standard Charges:
Baptist Healthcare Louisville
Norton Healthcare (One list for all hospitals.)
Jewish Hospital
Sts. Mary and Elisabeth Hospital
University of Louisville Hospital
Clark Memorial Hospital (Indiana)
University of Kentucky Hospital

Guidelines given by CMS to Hospitals for postings:
Statutory language from ACA.