Price Transparency With Gaps: Assessing the Completeness of Payer Transparency in Coverage Data

ABSTRACT

Objectives: To assess payer-level completeness of 2025 Transparency in Coverage (TIC) negotiated-rate files for physician, hospital outpatient, and hospital inpatient service lines.

Study Design: Cross-sectional descriptive review of 2025 TIC releases from 3 national payers (Aetna, Cigna, and UnitedHealthcare).

Methods: We created cleaned analysis files by ingesting machine-readable files and parsing and deduplicating TIC data sets for the major national plan of each insurer. We compared the network size from the TIC files against marketing materials that reported how many physicians and hospitals were in-network. We calculated what percentage of the most common billing codes had negotiated rate data for large family medicine, cardiology, and orthopedic surgery physician groups, hospital outpatient departments, and inpatient hospitals.

Results: Aetna and Cigna generally listed as many—or more—physicians and hospitals as their marketing materials did, whereas UnitedHealthcare listed fewer. Negotiated-rate completeness was highest for physician specialties and lowest—often minimal—for inpatient files. UnitedHealthcare’s physician groups were near complete, but inpatient data were sparse. Cigna showed high completeness for physician specialties and inpatient data, but limited hospital outpatient rates. Aetna demonstrated moderate to good physician completeness, midrange outpatient hospital data, and heterogeneous inpatient results. Overall, physician and hospital outpatient data were typically usable; inpatient data were insufficient.

Conclusions: For these payers, 2025 TIC files support analysis of physician and hospital outpatient prices but are inadequate for inpatient benchmarking. CMS should evaluate TIC completeness—internally or via an external auditor—and enforce penalties when required information is not published.

Am J Manag Care. 2025;31(Spec. No. 15):SP1121-SP1127

CMS’ price transparency rules require providers and payers to publicly release their negotiated rates for health care services. This started when CMS released the Hospital Price Transparency rule¹ in 2021, requiring hospitals to post a machine-readable file (MRF) containing prices for select shoppable services. Following this, a second rule, called Transparency in Coverage (TIC),² went into effect in 2022, requiring commercial insurers to publicly disclose data on negotiated pricing information (in-network rates) and historical paid amounts (out-of-network rates) in an MRF. These files must include all negotiated rates for all providers and be updated monthly. The sharing of health care prices at this scale publicly is one of the biggest developments in US health care and an important step toward introducing healthy competition in the health care system, but its usability depends on its accuracy and completeness.

Data from both rules have been of interest to health care analysts, researchers, and policy makers who want to better understand health care markets and price variation of health care services. The data contained in TIC files are more comprehensive because they cover a larger set of services and include rates not just for hospitals but also for physicians and other providers. TIC data have been used for analyses such as assessing variation in prices for select services for one plan³ across major payers,⁴ and evaluating hospital market share and prices,⁵ suggesting the high value of TIC data for health care policy in the US.

On the policy front, the White House released an executive order⁶ in February 2025 calling for the strengthening of price transparency initiatives in health care. In May 2025, the departments of HHS, Labor, and the Treasury released updated guidance⁷ to strengthen the quality and usability of TIC data.

We sought to answer 2 questions: (1) What proportion of providers in a payer’s nationwide network are present in their TIC files? and (2) What proportion of commonly utilized billing codes can one find prices for in the TIC data? The first question assessed the completeness of the TIC data at the provider level. The second question went a step further and assessed the completeness of the TIC data at the billing code level.

To our knowledge, this is the first analysis to assess the completeness of TIC data.

We found wide variation in the completeness of data by payer, specialty, and site of service across payers. The completeness of data on services provided by physicians was better than that on services provided in a hospital setting.

DATA AND METHODOLOGY

Data for this study came from the TIC files for the broad national network (preferred provider organization network) for Aetna, Cigna, and UnitedHealthcare. For each insurer, we identified and downloaded the original MRFs from the insurers’ websites in the second quarter of 2025. We did not include any of the Blue Cross Blue Shield plans because they are composed of dozens of individual plans with many different approaches to reporting TIC data, which means they cannot be combined for collective analysis and evaluation.

Completeness of Provider Networks

The TIC data present information for in-network providers at the National Provider Identifier (NPI) level. For each plan, we identified all NPIs that were listed as being in network and matched them to various databases with additional information on each NPI.^8-11 From this list of NPIs, we were able to calculate various specialties of providers and calculate a national estimate for the count of each type of provider. For the denominator, we obtained information from the insurers’ websites to identify the number of physicians, nonphysicians, hospitals, and facilities that each insurer purports to have in its broad networks.^12-16 The insurers grouped providers in different ways (eg, combining hospitals and other facilities), so we aggregated provider categories as needed.

Completeness of Billing Code Negotiated Rates

Estimating the completeness of billing rates requires a fundamentally different approach because there is no clear numerator or denominator for the data. To estimate how complete the data are, we developed an approach to look at multiple common specialties and identify what percentage of common codes were included with billing rates in the MRFs.

We first identified the 100 most frequent Healthcare Common Procedure Coding System (HCPCS) codes billed for 3 common specialties—cardiology, family practice, and orthopedic surgery—and the 100 most common HCPCS codes for hospital outpatient settings based on claims data from Medicare and All-Payer Claims Databases (APCDs) from Colorado and New Hampshire. For the hospital inpatient setting, we included all 771 diagnosis-related group (DRG) codes. The inclusion rate is the percentage of the common codes in our study for which a negotiated rate is found in the MRF. Because these are the most commonly performed billing codes, we expected providers to have negotiated rates for most, if not all, of these codes for their specialty.

Some small practices may only perform a subset of services, so for physician specialties, we limited our analysis to groups with at least 10 physicians in that specialty. We included all hospital inpatient and hospital outpatient locations. In each case, we excluded rates that were outside of a reasonable range (from 50% to 800% of Medicare rates) or had missing or inaccurate provider data (eg, nonexistent NPIs).

RESULTS

Completeness of Provider Networks

Table 1 shows the percentage of various types of providers in the TIC files and the number of providers reported in marketing materials. Aetna had a similar number of providers in its TIC data as in its marketing materials. Cigna had significantly more providers than it reported, whereas UnitedHealthcare had significantly fewer. These differences between the number of providers found in the TIC data (numerator) and the number of providers listed on insurer websites (denominator) could be due to measurement issues in both the numerator and denominator. In the numerator, we estimated the number of providers corresponding to organization NPIs, and our estimates could differ from the actual number of physicians. For the denominator, we relied on numbers provided by insurers on their websites, and it is possible that these do not reflect the present-day number of providers in their respective networks. Additionally, some of the discrepancies may come from variations in how each insurer defines a hospital, a facility, or specialists. However, it is unlikely that the definitions can explain all the differences between the numerator and denominator, and as long as these measurement issues affect all insurers in a similar way, our comparison of completeness of provider networks remains informative.

Completeness of Billing Rates

Table 2 provides summary statistics on the mean, 25th percentile, 50th percentile, 75th percentile, and SD for the completeness of each insurer’s data. For example, for Aetna hospital inpatient data, the 50th percentile hospital has data for all the common inpatient DRG billing codes, which means that more than half of all Aetna inpatient hospitals have full negotiated rate data, whereas the 25th percentile hospital has 11% or fewer of the common codes, indicating bimodal data—either the hospital had very good data or very incomplete data. For Aetna family practice groups, the 25th-percentile group has negotiated rates for at least 54% of codes, indicating moderate completeness for a high majority of groups, and the 50th percentile group has 77% of common rates, indicating better but still not complete coverage.

Cigna has quite complete data for cardiology, family practice, orthopedic surgery, and hospital inpatient data, with each having at least 89% completeness at the 25th percentile (meaning 75% of hospitals have at least that much data), and the 3 physician specialties have at least 99% completeness at the 50th percentile. Hospital outpatient data are significantly less complete, with the 50th percentile hospital outpatient only having 4% complete data.

UnitedHealthcare’s physician group data are very complete, with cardiology, family practice, and orthopedic surgery all having at least 98% of common codes at the 25th percentile. Conversely, hospital inpatient data are very sparse, with the median hospital only having negotiated rates for 2% of codes.

The Figure shows the distribution of data completeness for each insurer by provider specialty and inpatient and outpatient hospitals. The histogram shows the percentage of practices that fall within each of 10 deciles that represent the percentage of common codes for which there were reported negotiated rates. These figures help to visualize the distribution of completeness among groups and hospitals, with a higher density on the right side of the graph indicating more complete data. For example, in the data files released by Aetna, more than 25% of cardiology practices have data for all 100 HCPCS codes for this specialty. On the other hand, for orthopedic practices, more than 50% of providers in Aetna’s data files have complete data (ie, have rates for all 100 HCPCS codes selected in our analysis).

DISCUSSION

This analysis evaluated the completeness of insurer TIC data for 3 national payers and found significant variation in the completeness of the data. Aetna and Cigna generally had network information that was similar to or exceeded their published provider counts, whereas UnitedHealthcare had significantly fewer providers in its TIC data than advertised. In terms of including billing rates for common billing codes, Aetna had moderate to good completeness for each of the physician specialties, hospital outpatient, and hospital inpatient data. Cigna had good completeness for the physician specialties and hospital inpatient data, but very incomplete hospital outpatient data. UnitedHealthcare had very good completeness for physician specialties and hospital outpatient data, but very limited hospital inpatient data.

Improving estimates of the completeness of the data would benefit from more accurate networks (ie, the denominator in Table 1). Currently, it is challenging to identify what the total network should be, as few insurers allow downloads of their full network data with accurate identifiers (such as NPIs), and the data are often inaccurate.¹⁷ The Qualified Health Plan Directory pilot program,¹⁸ if fully adopted, may lead to better data.

It is unclear why there are differences between the insurers, but some factors may contribute. Some differences may arise from the different technical systems used by each insurer, which may make acquiring certain data more or less feasible. Additionally, each national payer has many subsidiaries that have different contracting approaches, and these approaches may influence data quality. With inpatient contracts, in particular, the insurers may not use standardized DRGs as defined by CMS and instead use some other payment methodology; however, if they do, we were unable to find comparable approaches for inpatient care in the data. Finally, insurers may choose, particularly with inpatient data, to not fully release the data, for competitive or other reasons. Future work should seek to better understand these differences.

To date, no insurers have been publicly fined or penalized for failing to comply with the transparency requirements. CMS, working through the Center for Consumer Information and Insurance Oversight, should develop standardized auditing plans to evaluate the completeness of insurer TIC data. It should also perform periodic audits to assess the accuracy of the data, potentially working with various providers to compare their negotiated rates with those that are published. If CMS is not well positioned to audit the plans, this could be outsourced to vendors, either via contract or through programs where those who report noncompliance could share in any penalty payments.

Additionally, an updated schema¹⁹ built on a relational database structure has the potential to improve data completeness and overall quality while simplifying both the creation and evaluation of the TIC data.

CONCLUSIONS

Data on negotiated rates reported in the TIC files offer rich opportunities for future research work. A comparison of prices across insurers is outside the scope of this article, given its focus on data completeness. This is the first study to evaluate the completeness of insurer TIC data and is an important first step in evaluating the quality of these data. Incomplete data can lead to inaccurate estimations of parameters of policy interest, such as average prices or market sizes, and thus these findings offer useful caveats to other users of TIC data. Although we found room for improvement, we also validated that significant amounts of data—particularly for physicians and, to a lesser extent, hospital outpatient services—are complete and available. This suggests that opportunities to help achieve the aims of price transparency are available, while recognizing the significant opportunity for improvement, particularly with hospital inpatient data.

Author Affiliations: Simple Healthcare (DBM, YP), Sanford, FL.

Source of Funding: Peterson Center on Healthcare.

Author Disclosures: Drs Muhlestein and Pathak are employed by Simple Healthcare, which works with price transparency data. Dr Muhlestein is also a member of the editorial board of Population Health, Equity & Outcomes.

Authorship Information: Concept and design (DBM, YP); acquisition of data (DBM); analysis and interpretation of data (DBM, YP); drafting of the manuscript (DBM, YP); critical revision of the manuscript for important intellectual content (DBM, YP); statistical analysis (DBM); and obtaining funding (DBM).

Send Correspondence to: David B. Muhlestein, PhD, JD, Simple Healthcare, 465 N Carolina Run, Sanford, FL 32773. Email: david@simple-healthcare.com.

REFERENCES

1. Hospital price transparency. CMS. Updated May 22, 2025. Accessed August 5, 2025. https://www.cms.gov/priorities/key-initiatives/hospital-price-transparency
2. Health plan price transparency. CMS. Updated September 10, 2024. Accessed August 5, 2025. https://www.cms.gov/priorities/key-initiatives/healthplan-price-transparency
3. Chartock BL, Simon K, Whaley CM. Transparency in Coverage data and variation in prices for common health care services. JAMA Health Forum. 2023;4(10):e233663. doi:10.1001/jamahealthforum.2023.3663
4. Whaley C, Radhakrishnan N, Richards M, Simon K, Chartock B. Understanding health care price variation: evidence from Transparency-in-Coverage data. Health Aff Sch. 2025;3(2):qxaf011. doi:10.1093/haschl/qxaf011
5. Pathak Y, Muhlestein D. Hospital system market share and commercial prices: a cross-sectional approach using price transparency data. Health Econ Rev. 2024;14(1):102. doi:10.1186/s13561-024-00580-w
6. Making America healthy again by empowering patients with clear, accurate, and actionable healthcare pricing information. The White House. February 25, 2025. Accessed August 5, 2025. https://www.whitehouse.gov/presidential-actions/2025/02/making-america-healthy-again-by-empowering-patients-with-clear-accurate-and-actionable-healthcare-pricing-information/
7. FAQs about Affordable Care Act implementation, part 70. US Department of Labor. May 22, 2025. Accessed August 5, 2025. https://www.dol.gov/agencies/ebsa/about-ebsa/our-activities/resource-center/faqs/aca-part-70
8. National downloadable file. CMS. Accessed September 10, 2025. https://data.cms.gov/provider-data/dataset/mj5m-pzi6
9. Medicare Fee-For-Service Public Provider Enrollment. CMS. Accessed September 10, 2025. https://data.cms.gov/provider-characteristics/medicare-provider-supplier-enrollment/medicare-fee-for-service-public-provider-enrollment
10. Medicare physician and other practitioners - by provider. CMS. Accessed September 10, 2025. https://data.cms.gov/provider-summary-by-type-of-service/medicare-physician-other-practitioners/medicare-physician-other-practitioners-by-provider
11. NPI files. CMS. Accessed September 10, 2025. https://download.cms.gov/nppes/NPI_Files.html
12. UnitedHealthcare Choice Plus. UnitedHealthcare. Accessed September 10, 2025. https://www.uhc.com/content/dam/uhcdotcom/en/BrokersAndConsultants/unitedhealthcare_choice-plus-flier.pdf
13. UnitedHealthcare: overview and highlights. UnitedHealthcare. Accessed September 10, 2025. https://www.unitedhealthgroup.com/content/dam/UHG/PDF/investors/2023/ic23/UHG_IC23_UHC_Overviews_Highlights.pdf
14. UnitedHealthcare expands its individual and family plans on the health insurance marketplace to 30 states. News release. UnitedHealthcare. November 1, 2024. Accessed September 10, 2025. https://www.unitedhealthgroup.com/newsroom/2024/2024-11-01-uhc-expands-individual-family-plans-health-insurance-marketplace.html
15. Cigna Healthcare small group health plans for employers with 2-50 employees. Cigna. Accessed September 10, 2025. https://www.cigna.com/employers/small-business/small-group-health-insurance-plans
16. Building a better health care network. Aetna. Accessed September 10, 2025. https://www.aetna.com/individuals-families/health-insurance-through-work/network.html
17. Butala NM, Jiwani K, Bucholz EM. Consistency of physician data across health insurer directories. JAMA. 2023;329(10):841-842. doi:10.1001/jama.2023.0296
18. Qualified Health Plan Directory pilot fact sheet. CMS. June 30, 2025. Accessed September 10, 2025. https://www.cms.gov/files/document/qhp-directory-pilotfact-sheet.pdf
19. Muhlestein D. Improving price transparency data: recommendations from practice. Health Affairs Forefront. March 19, 2025. Accessed August 5, 2025. https://www.healthaffairs.org/content/forefront/improving-price-transparency-data-recommendations-practice