Background and Significance
In January 2019, the Centers for Medicare & Medicaid Services (CMS) ruled that hospitals
had to increase health care price transparency by publicly listing their standard
charges (chargemaster) “via the internet in a machine-readable format” and updating
“this information at least annually.”[1] Whereas chargemaster charges vary significantly across hospitals and have been documented
as high as 10 to 20 times the allowable Medicare costs,[2] they do not necessarily reflect the total charge for an episode of care (i.e., the
charge for an appendectomy vs. the individual charges for anesthesia, time in the
operating room, sutures, etc.). Chargemaster charges also often do not reflect the
out-of-pocket costs that insured patients incur. Private payers negotiate payment
rates that are considerably lower than published charges and insured patients often
pay a fraction of the negotiated payment. The amount that an insured patient pays
out of pocket depends upon the cost-sharing structure of the insurance plan, such
as co-pays, deductibles, and networks, in addition to the patient's year-to-date spending.
Insured patients' health insurance literacy is varied[3]
[4]
[5] and, for the most part, insured patients are largely shielded from health care costs.
Chargemaster charges, however, directly affect the consumer cost for uninsured, out-of-network,
auto or casualty insurance, and workers' compensation patients.[6] Chargemaster charges have been linked to surprise medical bills, such as through
the practice of balance billing where out-of-network health care providers bill patients
for the amounts not covered by their insurer; however, the frequency and magnitude
of surprise medical bills is likely to lessen with the surprise billing legislation
passed by Congress in December 2020 and due to be implemented in 2022.[7] Increased private insurance premiums have also been linked to chargemaster charges.[8] Researchers advocate for increased “price transparency” so that patients understand
consumer costs prior to obtaining service. However, the empirical evidence on the
effectiveness of these approaches is mixed.[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18] The CMS chargemaster rule aimed to help patients understand their potential financial
liability and enable comparison-shopping among hospitals. Hospitals are free to choose
how they present the charges as long as they adhere to the aforementioned minimum
requirements.[19] Therefore, the usability of charges or list prices for cross-hospital comparisons
is unclear.
Methods
To explore the rule's utility for patients to comparison-shop medical services, we
identified chargemasters published on the internet for Dallas County hospitals in
May of 2019. We identified Dallas County hospitals by typing “Hospitals in Dallas
County” into Google's search engine and subsequently located hospital chargemasters
by typing in the hospital name followed by “chargemaster” or “price.” We then navigated
the hospital's Web site for a chargemaster. Chargemasters available in a database
compatible format (CSV, XLSX) were aggregated into a single file (see [Supplementary Appendix], available in the online version) including chargeable items, prices, and descriptions.
We opted not to list hospitals by name, as our goal was not to compare institutions
but to provide a general examination of the use and usability of the chargemaster
in general. Readers interested in more detail may download the Appendix (available
in the online version) and request the key for the organizations from the corresponding
author.
A convenience sample of commonly ordered laboratory tests,[20] prescribed medications,[21] and therapeutic or diagnostic procedures[22] across a variety of medical specialties and readily identifiable based upon descriptions
across the hospitals was selected for charge comparison. Multiple experts performed
independent review of the data to ensure that like tests, medications, obtained consensus,
and procedures were comparable. Authors SAA, CUL, and SM performed descriptive statistics
on the charges and compared hospitals based on their overall ranking of charges within
each category.
We elected not to publish the names of the hospitals; however, complying with rules
for good scientific practice, all chargemaster data (including name of the hospitals)
were stored and are available upon request from the corresponding author.
Results
Thirteen hospital chargemasters were identified. One institution requires user affirmation
of a disclaimer explaining the limitations of chargemaster data. Eleven hospitals
provided data in a format compatible for aggregation. Of the hospitals that did not
provide data in an accessible format, one published their chargemaster as a locked
PDF from which the data could not be extracted. Another inaccessible chargemaster
had web-based tables that were not downloadable. An attempt to use web scraping (automatically
pulling information from a Web site and processing it into a searchable format) resulted
in blocked access to the Web site. We aggregated the 11 downloadable chargemaster
files into a single file that included 155,576 chargeable items, their prices, and
descriptions. The procedures and services listed in the chargemasters did not use
standard terminologies, abbreviations, or codes, preventing the automation of price
comparisons among institutions. None of the examined chargemasters had information
on professional fees or discounts offered to self-pay patients.
We selected a convenience sample of common laboratory tests,[20] medications, and procedures across medical specialties to examine the charge variation
among institutions ([Table 1]). The items selected were not universally listed in all chargemasters. The least
frequently listed laboratory test was Rh blood typing (8/11 institutions), the least
frequently listed medications were omeprazole 20-mg and azithromycin 250-mg tablets
(both 8/11 institutions), and the least frequently listed procedure included cesarean
section (C-section; 4/11 institutions; [Table 1]). There was high variation in the charges for the 10 laboratory tests ([Table 1]). The charge for a partial thromboplastin time (2020 Medicare reimbursement: $6.01)[23] ranged from $18.70 to $506.00, a 2,606% difference and an automated platelet count
ranged from $12.00 to $362.00, a 2,917% difference. Two of the 11 hospitals (18%,
hospitals 2 and 9) accounted for the lowest charge in 40% of the laboratory tests
and 1 hospital (hospital 7) accounted for the highest charge in 80% of examined tests.
Table 1
Number of institutions reporting, median price, interquartile range (IQR), lowest
price institution, highest price institution, and percent difference between lowest
cost and highest price institution for common laboratory tests, medications, and procedures
|
No. of institutions reporting
|
Median ($)
|
IQR ($)
|
Lowest price institution (ID)
|
Price from lowest price institution ($)
|
Highest price institution (ID)
|
Price from highest price institution ($)
|
Percent difference between lowest and highest price institution (%)
|
2020 Medicare reimbursement rate ($)
|
Laboratory test
|
Complete blood count
|
11/11
|
162.88
|
43.00
|
2
|
48.00
|
7
|
708.00
|
1,375
|
7.77
|
Hemoglobin
|
11/11
|
57.75
|
62.17
|
9
|
13.13
|
7
|
200.00
|
1,423
|
2.37
|
Hematocrit
|
11/11
|
58.00
|
57.58
|
9
|
13.13
|
7
|
189.00
|
1,339
|
2.37
|
Comprehensive metabolic panel
|
11/11
|
445.90
|
425.14
|
4
|
43.00
|
7
|
1,135.00
|
2,540
|
10.56
|
Automated platelet count
|
9/11
|
83.13
|
83.70
|
9
|
12.00
|
7
|
362.00
|
2,917
|
4.48
|
Troponin, quantitative
|
10/11
|
370.72
|
189.51
|
2
|
73.00
|
7
|
829.00
|
1,036
|
12.47
|
Blood typing (ABO)
|
10/11
|
107.10
|
163.39
|
2
|
54.00
|
10
|
341.61
|
533
|
2.99
|
Blood typing (Rh)
|
8/11
|
113.40
|
68.10
|
2
|
52.00
|
7
|
310.00
|
496
|
2.99
|
Prothrombin time
|
11/11
|
158.55
|
134.00
|
4
|
24.00
|
9
|
450.00
|
1,775
|
4.29
|
Partial thromboplastin time
|
11/11
|
145.00
|
128.53
|
9
|
18.70
|
7
|
506.00
|
2,606
|
6.01
|
Medications
|
Atorvastatin 10-mg tablet
|
9/11
|
16.96
|
34.63
|
1
|
0.75
|
7
|
60.99
|
8,032
|
|
Levothyroxine 88-μg tablet
|
9/11
|
3.25
|
3.28
|
8
|
1.25
|
7
|
14.47
|
1,058
|
|
Omeprazole 20-mg capsule
|
8/11
|
12.28
|
12.02
|
1
|
1.16
|
7
|
114.53
|
9,773
|
|
Lisinopril 10-mg tablet
|
9/11
|
4.05
|
7.34
|
1
|
0.17
|
7
|
21.64
|
12,629
|
|
Amlodipine 5-mg tablet
|
9/11
|
6.38
|
10.05
|
1
|
0.23
|
7
|
43.05
|
18,617
|
|
Azithromycin 250-mg tablet
|
8/11
|
29.91
|
28.87
|
4
|
5.91
|
7
|
228.75
|
3,771
|
|
Zolpidem 5-mg tablet
|
9/11
|
19.88
|
16.37
|
1
|
0.98
|
6
|
56.1
|
5,624
|
|
Furosemide 20-mg tablet
|
9/11
|
3.00
|
4.98
|
1
|
0.20
|
7
|
18.11
|
8,955
|
|
Gabapentin 100-mg capsule
|
9/11
|
3.00
|
5.40
|
1
|
0.23
|
7
|
18.11
|
7,774
|
|
Metformin 500-mg tablet
|
9/11
|
3.00
|
5.34
|
1
|
0.16
|
7
|
21.64
|
13,425
|
|
Procedures
|
Chest X-ray
|
11/11
|
447.00
|
222.00
|
2
|
182.00
|
3
|
719.00
|
295
|
|
Ct abdomen w/o contrast
|
11/11
|
3,332.00
|
1,644.65
|
2
|
927.00
|
11
|
4,330.75
|
367
|
|
Incision and drainage of skin abscess, simple
|
7/11
|
615.18
|
216.08
|
9
|
243.00
|
6
|
966.00
|
615
|
|
Immunization administration
|
6/11
|
132.50
|
114.25
|
2
|
59.00
|
3
|
260.00
|
341
|
|
C-section
|
4/11
|
4,372.00
|
2,000.96
|
11
|
2,890.00
|
1
|
6,669.83
|
131
|
|
Circumcision
|
6/11
|
736.83
|
1,084.66
|
8
|
252.00
|
3
|
7,532.10
|
2,889
|
|
Intubation
|
6/11
|
775.33
|
486.90
|
9
|
260.00
|
7
|
1,181.00
|
354
|
|
Colonoscopy
|
7/11
|
2,513.32
|
915.78
|
10
|
1,081.05
|
8
|
3,425.00
|
217
|
|
Hemodialysis
|
10/11
|
2,184.21
|
1,414.63
|
4
|
1,125.00
|
8
|
4,515.00
|
301
|
|
ECG
|
11/11
|
416.00
|
388.74
|
9
|
198.00
|
10
|
1,022.87
|
417
|
|
Abbreviations: CT, computed tomography; ECG, electrocardiogram; IQR, interquartile
range; Rh, rhesus; w/o, without.
Note: The 11 institutions included in the study were assigned numbers 1–11.
The 10 medications had the highest charge variance of the groups sampled across institutions
([Table 1]). The charge for a single 5-mg tablet of amlodipine ranged from $0.23 to $43.05,
an 18,617% difference. One institution (institution 1) accounted for the lowest charge
in 80% of medications and one institution (institution 7) accounted for the highest
charge in 90% of medications. Notably, institution 7 reported only brand name medications.
Procedures had the highest median charge of the three groups ($775.40, compared with
$133.35 for laboratory tests and $6.45 for medications). They had the smallest median
percent difference of all three groups (347%, compared with 1,423% for laboratory
tests and 8,494% for medications). The largest percent difference for a procedure
was 2,889% for a circumcision, with charges ranging from $252.00 to $7,532.10. Two
institutions accounted for 60% of the lowest charges (institutions and institution
9). One institution accounted for 30% of the highest charges (institution 3).
Across all 30 sampled charges, institution 1 accounted for 27% of the lowest charges
and institution 7 accounted for 60% of the highest charges. Of the sampled laboratory
tests, medications, and procedures, only two items had identical descriptions across
all institutions: the laboratory tests for hemoglobin and for hematocrit. Descriptions
with the largest variation for each grouping included 9 different descriptions for
partial thromboplastin time, 6 different descriptions for 10-mg tablet of atorvastatin,
and 10 different descriptions for computed tomography (CT) abdomen without contrast
([Table 2]).
Table 2
Single greatest variation in description for sampled laboratory tests (partial thromboplastin
time), medications (atorvastatin 10-mg tablet), and procedures (CT abdomen w/o contrast)
Laboratory test
|
Medication
|
Procedure
|
Thromboplastin time, partial, activated
|
Atorvastatin 10-mg tablet (×4)
|
CT abdomen w/o CM
|
Partial thromboplastin time (PTT)
|
Atorvastatin calcium 10-mg tablet
|
CT abdomen w/o dye (×2)
|
Thromboplastin time partial
|
Lipitor 10-mg tablet
|
CT scan, abdomen, w/o contrast
|
Thromboplastin time
|
Atorvastatin 10-mg tablet
|
CT abdomen w/o contrast
|
PTT (×4)
|
Atorvastatin calcium 10 mg
|
CT abdomen w/o contrast
|
Thromboplastin (PTT)
|
Atorvastatin tablet 10 mg
|
CT abdomen w/o contrast
|
PTT plasma fractions EA
|
|
CT abdomen w/o contrast
|
Thromboplastin time (APTT)
|
|
CT abdomen w/ contrast
|
Thromboplastin time partial
|
|
CT abdomen w/o contrast
|
|
|
CT abdomen w/o contrast
|
Abbreviaions: APTT, activated partial thromboplastin time; CM contrast medium; CT,
computed tomography; PTT, Partial thromboplastin time; w/o without.
Discussion
We identified and aggregated hospital chargemasters within a local regional market
in an attempt to compare 10 common laboratory tests, medications, and procedures across
institutions. Our findings highlight a high variation in chargemaster charges as well
as heterogeneous descriptions for identical items across institutions. Similar findings
have been reported by other recent studies using chargemaster data.
Since the implementation of the CMS price transparency rule, several studies have
attempted to compare chargemaster charges among institutions for fewer items over
larger geographic areas.[24]
[25]
[26] One study looked at six advanced diagnostic imaging services across the top 20 ranked
hospitals in U.S. News and World Report.[25] Another study analyzed low-dose chest CT scans for lung cancer screening across
55 American College of Radiology accredited imaging centers across the United States.[24] A third study compared chargemaster data for four items across all 110 acute care
hospitals in North Carolina. All studies found different hospital implementations
of the chargemaster rule, marked variation in prices across institutions, and heterogeneous
descriptions for items. Our study is the first to compare hospital chargemaster prices
for common items within a local hospital market in an attempt to leverage the price transparency rule as intended for use by patients.
Chargemaster Accessibility and Usability
Chargemaster data were available in a format amenable to comparison in 85% of the
hospitals. This accessibility is similar to the study of North Carolina chargemasters
where only 72% (79/110) were accessible.[26] Chargemaster files often used ambiguous terminology and lacked a common standard
to describe charges. This was seen in a prior study with 19 different descriptions
for head CT without contrast across 20 hospitals.[25] When comparing chargemaster items among Dallas County hospitals, we observed markedly
heterogeneous descriptions of synonymous tests and services across institutions, which
created several limitations.
One limitation is that the chargemaster items we compared may not actually reflect
the same items as we intended. For example, for the data we collected about circumcisions,
most hospital descriptions simply listed “circumcision” for this procedure. However,
the large price variation we identified ($252.00–7,532.10) may reflect the difference
between an uncomplicated circumcision of a newborn in the newborn nursery and a circumcision
of an older child/adult with phimosis in the operating room under anesthesia. These
would reflect two different procedures; however, without further specifications, a
consumer would not be able to discern this difference on the chargemasters. Additionally,
for the items that we were not able to identify in a hospital's chargemaster ([Table 1]), we are uncertain as to whether this is because (1) the hospitals failed to publish
their full chargemaster, (2) the hospitals did not offer the items, or (3) we simply
could not find what we were looking for. For example, there were two hospitals that
did not have charges listed for common medications—it can be assumed that this is
because the institutions did not publish their full chargemasters. There were also
several hospitals that did not list prices for common procedures; however, this may
be related to the institution not offering the procedure (e.g., a rehabilitation hospital
did not offer delivery or circumcision). Additionally, we noticed that one hospital
that offered delivery services failed to list C-section charges in its chargemaster.
It can be assumed either that the item was not published or that we (a team of medical
experts) could not find it. Considering these limitations, we found our hypothesis
confirmed that chargemasters published to meet the minimum requirements of the rule
would be difficult for consumers to use and would not be conducive to easy comparison.
Chargemaster Charge Variations
Exploring the chargemaster files in Dallas County, charges varied greatly across the
11 hospitals for the 30 services we examined. We found charge differences of more
than 18,000% for one medication. We observed that some hospitals consistently had
the highest and lowest prices. Although these specific price variations may not be
generalizable to the rest of the United States, previous studies have identified high
variation and prices across different items ranging from 56 to 4,916% difference across
items and studies.[24]
[25]
[26] The study of chargemasters across North Carolina revealed a 358% variation across
institutions ($72.00–258.00) for a complete blood count with differential. This price
variation was not as extreme as was identified in our local market (1,375% variation,
$48–708) for the same laboratory test.
A historic practice reimbursement model in Texas may have contributed to some of the
price differences seen in our regional market. Until 2013, Texas Medicaid reimbursed
inpatient stays based on a percentage of the hospital's standard charges for children's
hospitals, state-owned hospitals, and rural hospitals.[17] Charges were directly linked to payments for certain hospitals and may have contributed
to current chargemaster variations. The contribution of previous reimbursement practices
on the variation of chargemaster charges across states is an area for future research.
Chargemaster Charges versus True Costs to Insured and Uninsured Patients
Comparing the hospital chargemaster charges yields limited information for insured
patients investigating their out-of-pocket costs. In some circumstances, a hospital
with the highest chargemaster charge for a given laboratory test, medication, or procedure
may actually have the lowest out-of-pocket cost for an insured patient. The a priori
contract between the patient's insurance company and the hospital supersedes the chargemaster
charge, and costs usually represent only a fraction of the chargemaster prices. Additionally,
the chargemaster charges for individual hospital services fail to reflect that insurance
companies reimburse hospitals with prenegotiated lump sums for diagnosis-related services
(e.g., total shoulder replacement in lieu of individual components of care). Depending
on the insurance plan and year-to-date health care spending, a patient will pay a
fraction of the insurer's negotiated rate out of pocket, rendering the chargemaster
irrelevant for most insured consumers.
Chargemaster charges potentially yield valuable information for a growing number of
patients in the United States, who are uninsured or financially liable for large portions
of the cost of their care (e.g., patients with high deductibles, or seeking out-of-network
services or services not routinely covered by insurance, e.g., infertility treatments).
The wide variation in charges demonstrates that the comparison of these chargemaster
charges for some patients could represent the difference between financial well-being
and medical bankruptcy. However, due to various hospital practices such as discounted
cash prices, sliding scale payments, and charity care, chargemaster charges are only
an estimation of actual costs to the uninsured. Unlike other studies that have found
that some hospital chargemasters included information about expected costs for patients
without insurance,[25] the hospital chargemasters we sampled did not include this information.
Failure to Achieve Price Transparency
The CMS rule in its current form falls short of its stated goal of price transparency.
Whereas charges for the same service vary across institutions, this information likely
remains unattainable and unusable to most patients. It is likely that most patients
lack the skills and health care insight to find and download the data, analyze them,
and match chargeable items successfully to allow comparison-shopping for all of the
tests, procedures, and medications that will be required during their care episode.
The lack of consistent naming conventions for laboratory tests and procedures, the
use of random abbreviations, and varied terminology require significant “sleuthing”
and medical knowledge to match comparable items across institutions. Chargemasters
are frequently not easily found on the hospitals' Web sites, which frequently also
lack search functionalities. In addition, some hospitals actively block the downloading
of the chargemaster in violation of the rule's spirit. Hospitals have even been found
to have disclaimers discouraging users from using chargemasters to calculate out-of-pocket
costs.[26]
Improving Chargemaster Price Transparency
The utility of CMS's 2019 Hospital Price Transparency rule could be improved by requiring
institutions to publish chargemasters in a consistent manner across all hospitals
and to use standard health care descriptors for communication, such as Healthcare
Common Procedure Coding System (HCPCS) codes for procedures, RxNorm codes for medications,
and Logical Observation Identifiers Names and Codes (LOINC) for laboratory tests,
which would permit better price comparison. Further, publishing the total price of
services for an episode of care (e.g., vaginal delivery) instead of the individual
prices of items (e.g., anesthesia, oxytocin, fluids, room and board, and gauzes) would
be more meaningful to patients. Finally, for insured patients, insurance companies
could be required to provide pricing data and out-of-pocket costs for insured patients.
This requirement could include information on insurance coverage, networks, co-payments,
deductibles, and year-to-date health care spending to allow consumers comparison of
out-of-pocket costs.
Expanded Price Transparency Requirements for Hospitals
CMS released an expanded Hospital Price Transparency rule, which took effect in January
2021. The new rule increases the scope of published price information and mandates
annual publishing of the charges for 300+ care services that patients can schedule
in advance (“shoppable” services) and individual hospital items and services. Descriptions
must include standard charges, gross charges, discounted cash prices, payer-specific
negotiated charges, and payer de-identified minimum and maximum negotiated charges.
Hospitals must also include a description of accounting or billing codes, such as
HCPCS codes. Hospitals must display charges in a consumer-friendly manner and include
an internet-based price estimator tool that allows health care consumers to estimate
accurately the required payment amount in advance for the service.[27]
Limitations of the Expanded Price Transparency Requirements
Although an improvement in the opinion of the authors, even the expanded Hospital
Price Transparency rule for hospitals will have limited utility to insured patients.
Self-pay patients may have sufficient information to calculate out-of-pocket costs
and to comparison-shop for care, whereas insured patients will not be able to use
the pricing information without sufficient knowledge about their health insurance
plans. These deficits are addressed in a proposed “Transparency in Coverage” rule
that may benefit insured patients by requiring payers to publish real-time, personalized
cost-sharing estimates allowing patients to determine their liability for covered
health care items and services. This proposed rule requires payers to publish negotiated
rates for in-network providers and allowed amounts paid to out-of-network providers[28] and would address many shortcomings of the initial 2019 Hospital Price Transparency
rule. However, this rule will likely be ineffective unless strict publication guidelines
are applied and enforced to make the publication of prices and out-of-pocket costs
for consumers comparable and actionable. Further, if the novel price information and
other required data are not presented in a clear and usable format, the proposed rule
could inadvertently further obscure price transparency for patients. Hospitals contract
with numerous health insurers, who in turn offer multiple health insurance plans,
resulting in a plethora of negotiated prices for a single item. In the absence of
defined publication guidelines including a terminology for payers, hospitals, and
plans, the proposed rule will lead to varying implementation across institutions limiting
usability and erecting barriers to price comparison.
The economist Peter Diamond showed that even a small “search cost” could undermine
the competition on price.[29] Therefore, industries (such as health care) where comparison is difficult or costly
can anticipate higher prices and profits.[30] Price comparison intuitively increases competition and reduces prices. However,
if consumers—regardless of their reasons (e.g., personal choice, too much effort,
unable to understand or use)—do not use price data to comparison-shop, publishing
the data could paradoxically increase costs as hospitals and insurance companies become
aware of negotiated rates by competitors that were once trade secrets with the ability
to match competitors' higher prices.
Legal Challenges to Price Transparency
The new Hospital Price Transparency rule for 2021 has been met with litigation. The
suit alleges the rule exceeds the agency's statutory authority and violates the First
Amendment by requiring public disclosure of prices negotiated with payers.[31] CMS predicts that for the first year, the total implementation burden on hospitals
will be 150 person-hours and $11,989.60 per hospital.[27] A federal judge ruled against the lawsuit; however, the plaintiffs plan to appeal
this decision and seek expedited review.[32]
Conclusion
Ultimately, the success of any price transparency rule depends on how easily accessible,
usable, and understandable the price data are. We found current charges difficult
to assess, compare, and apply to real-life scenarios to calculate out-of-pockets costs,
rendering their value less useful. As future iterations of this rule are implemented
and health care price information becomes more readily available from hospitals and
insurers, careful study will be needed to assess its usability and effect on price
comparison and patients' ability to shift care to lower-priced providers. Further,
the effect of price comparison on patients' personal financial burden, stimulating
price competition and minimizing the wide variation in medical prices, health care
spending, and health care outcomes will need to be evaluated.