Milliman Transparent unlocks a new frontier of analytics for employers.

The Hospital Price Transparency (HPT) regulation¹ and the Transparency in Coverage (TiC) Final Rules² have changed the U.S. healthcare system by requiring both hospitals and payers to publish unblinded, contractual price information for every provider and service code. This paper specifically focuses on how Milliman Transparent can support self-funded employers with analytics based on price transparency data to enhance their benefit offerings and fulfill their obligations as fiduciaries. In this paper, we provide examples with real data and highlight key questions that employers can answer with Milliman Transparent.

This paper assumes the reader is already familiar with the concepts discussed in the paper, “Price transparency in 2023.”³

Range of opportunities

The transparency data enables employers to pursue a number of actions to support their benefit programs, including:

• Perform unit cost comparisons of networks.
• Request a specific discount guarantee from a third-party administrator (TPA) partner that requires improvements in contracted rates for specific regions or providers.
• Negotiate a direct contract with a provider to achieve better contracted rates and discounts.
• Evaluate Centers of Excellence (COE) models for certain services.
• Develop a tiered network plan design to incentivize utilization of cost-effective and high-quality providers.
• Push information to plan members to support in provider selection.
• Set employee contributions to reflect cost differentials between networks when offering multiple options.

This paper explores how transparency data can enhance the unit cost comparisons of networks for employers. Note that some of the opportunities (e.g., direct contracting) require sufficient employer membership and/or substantial concentrations of employees in a specific geographic area to be feasible.

Network unit cost comparisons

In 2023, over 153 million Americans received health insurance through an employer.⁴ Nearly 65% were covered by employers that managed self-funded benefit programs with third-party administrators (TPAs).⁵ In the self-funded market, the TPA is often also referred to as the payer or carrier. Typically, employers contract with one or more TPAs to provide health insurance coverage to their populations. TPAs develop their own provider networks and negotiate reimbursement levels with each provider in the network. For confidentiality reasons, employers have historically had limited visibility into the actual negotiated rates by provider that they (and their employees) will pay, and how these rates compare across providers and under different TPA networks. Transparency data makes the rates TPAs and providers negotiate publicly available, and has the potential to change the conventional relationship between employers and TPAs. With transparency data, employers can gain insight into these proprietary pricing relationships to better inform benefit decisions.

Discounts as a comparison basis

Evaluating TPA partners is a critical task for self-funded employers and often includes assessments of total estimated cost, provider network breadth, customer service components, and claim adjudication and payment processes. Claim costs are the largest portion of costs to self-funded employers, and, historically, these costs have been assessed by evaluating aggregate discounts off of provider billed charges.

The use of discounts as a comparison basis is a practice that was driven primarily by the simplicity of the calculations and limitations of available data. Assessing claim costs with other methods, such as an average charge, or on a Medicare relative basis, was challenging due to difficulties related to availability and quality of detailed claims and cost data, whereas discounts could be calculated solely based on total dollars. Although a popular method of analysis, the use of discounts has a few drawbacks that should be considered:

• Discounts are calculated from the provider’s billed charges, which can vary substantially among providers. A provider with high billed charges will ”artificially” have a higher discount even if receiving contractual negotiated rates that are similar to another provider that has lower billed charges.
• In many cases, billed charges increase faster than contracted amounts, which results in discounts growing over time, artificially implying an improvement in negotiated rates.
• Discount analyses typically imply that cost differences are universal across an entire market and do not show variation by provider and by service. Healthcare costs vary widely across the country and negotiated rates for a TPA can even vary materially within the same geography.⁶

Common discount-based analysis methods

Discounts are calculated directly from historical claim data. Many TPAs, consultants, and brokers use a standard, defined approach through the Uniform Discount and Data Specifications (UDS). UDS specifies the claims dataset to be used and the methodology for calculating discounts for use in TPA comparisons. UDS data is subject to stringent requirements and actuarial sign-off to ensure accuracy. As such, UDS has long been the industry ”gold standard” for measuring relative price differentials.

These discount-based analyses are typically performed using discount information aggregated to the three-digit ZIP Code level (e.g., “606” in Chicago) from historical claim data provided by health plans across the country. A typical analysis utilizes employer census data to develop a weighted average of the discounts across the geographic regions where their employees reside. The TPA with the highest average discount as determined by this analysis is generally assumed to have the most favorable contracts and, therefore, will result in the lowest claim cost.

A variation on the discount-based approach is for an employer to provide its historical claims data to a TPA for repricing. Often, the TPA is relying on its own internally prepared UDS for this analysis, which likely reflects a different mix of services from the self-funded employer’s claim data and the other TPAs being evaluated. Because the discounts are often self-reported by the TPA, they cannot usually be independently verified.

Discount data compared to transparency data

While achieving the same goal of TPA comparisons, analyses based on discounts and transparency data have several fundamental differences, as shown in Figure 1. The most fundamental difference is that discount-based analyses rely on historical and blinded claim experience, while transparency data contains current provider-specific negotiated rates.

Relying on claim experience, discount-based analysis has the advantage of being based on fully adjudicated claims that reflect all payment adjustments, a TPA’s actual mix of services and providers (although provider-level values are confidential), and the general simplicity of being discount-based. In comparison, transparency data reflects current contract or fee schedule amounts that are shown by provider without confidentiality limitations.

Aggregating transparency results

Because transparency data contains current negotiated rates (and is not based on historical claim experience), it does not have the claim utilization data necessary to aggregate providers together for market-level results. Transparency data allows for fully unblinded provider-level comparisons for employers by service category. However, if market-level aggregations are needed, a provider distribution may be used that is based on the employer’s own experience, or provider market share assumptions.

Other transparency considerations

Despite the advantages of price transparency data, there are challenges that must be addressed. Transparency data reflects the negotiated payment rates by service code, e.g., Healthcare Common Procedure Coding System (HCPCS) or diagnosis-related group (DRG), posted in machine-readable files (MRFs). These files are massive in size and complicated for self-funded employers to process and interpret. Utilization data is not included in MRFs and must be appended appropriately to allow for rolled up comparisons of contractual rates. Some contractual terms, like inpatient stop-loss, are not included in the MRFs. Due to these challenges, most employers and benefit consultants and brokers have not yet been able to readily access or use the data. However, price transparency data provides the opportunity to perform unblinded, employer-specific comparisons with current price information to supplement traditional analyses, improve the reliability of the output, and support decision making. Milliman Transparent was developed to help employer groups and other stakeholders obtain access to the data and review actionable analytics.

It is worth noting that none of the analyses described above capture differences in provider quality. Although cost is important, ensuring employees have access to providers that deliver high-quality, comprehensive care is very important for employers. To support a holistic view of value, evaluations of cost should be married with provider quality metrics, where available. The remainder of this paper will demonstrate the type of analysis employers can perform through Milliman Transparent.

Network comparisons

Which TPA has negotiated the lowest payment rates in my market(s)?

As mentioned above, a key use case for transparency data for employers is to enhance their understanding of relative cost differences between TPAs. The price transparency data provides the opportunity to supplement traditional analyses, which are based on blinded historical claims, to capture additional insights that can be seen at a provider-specific level. Additionally, transparency data offers the ability to further customize the analysis based on an employer’s own utilization either by adjusting weights by service category or by adjusting the results to only focus on the providers that an employer’s population utilizes today.

Figure 2 compares two different TPA networks across select hospitals in the Chicago-Naperville-Elgin, metro area of Illinois and Indiana. Figure 2 reflects inpatient TiC data aggregated into Milliman’s GlobalRVUs™ (GRVUs) Medicare metric. GRVU Medicare is calculated using Milliman’s GRVUs to approximate nationwide Medicare relativities. The GRVUs are a set of relative value units (RVUs) that cover the entire range of healthcare services.

The GRVUs can be thought of as an all-payer version of Medicare because the GRVUs help overcome common limitations of contract comparisons that use Medicare fee schedules. In particular, the GRVUs include adjustments to better account for service categories that are common in commercial populations but are not captured adequately by an age 65+/disabled Medicare population, e.g., maternity, neonatal intensive care unit (NICU), pediatrics. The percentage of GRVU Medicare is calculated by combining the GRVUs with the codes in the data that pass our quality checks and match to our utilization data.

This example shows that, in aggregate, the UHC Choice Plus network appears to have favorable reimbursement rates by about 3% relative to the Cigna OAP network (using number of beds as a proxy for claim weights). However, if an employer knows that its employees primarily utilize providers in the northern suburbs such as Advocate Lutheran General versus facilities in downtown Chicago like Northwestern Memorial Hospital, then the relationship between the TPA networks changes significantly. Under this scenario, Cigna OAP is 2% more favorable than UHC Choice Plus.

This type of analysis is especially powerful in markets where one or two facilities are highly utilized by an employer’s population or where certain facilities have a significant difference in negotiated rates that may be obfuscated when aggregated across a 3-digit ZIP Code or metro area.

How do TPA networks compare?

Another use case of transparency data is the ability to compare cost differentials of various networks from the same payer. Many payers will offer different types of networks based on product, e.g., health maintenance organization (HMO) versus preferred provider organization (PPO), and breadth of the network, narrow versus broad. Payers often have networks that are marketed as ”high performance” or ”narrow” that offer lower negotiated rates with specific hospitals or health systems. The price transparency data provides the opportunity to gain insight into these networks in ways that were not possible in the past.

The table in Figure 3 contains the Percentage of GRVU Medicare⁷ for two networks within the same payer (Anthem) for a sample of hospitals in Cleveland, Ohio. The table contains inpatient (IP) and outpatient (OP) results for Anthem’s broad PPO network and a high-performance network (HPN) from Milliman Transparent. These results are limited to hospitals with credible results for both IP and OP services. The transparency data allows employers to see the breadth of the HPN (i.e., they do not appear to have a contract with the facilities with ”n/a”) as well as the cost differential between the PPO and HPN. In this market, the HPN has lower reimbursement rates of about 10% on IP and 3% on OP as compared to the broad PPO. This type of analysis can provide deeper insight into the various payer networks as well as the differentials across hospitals in a single market, which can enhance TPA evaluations and help support the case to implement an HPN.

Is my TPA publishing data accurately on my behalf?

Under the TiC regulation, payers are required to provide a mapping of the specific Health Insurance and Oversight System (HIOS) IDs and employer identification numbers (EINs) that correspond to each set of negotiated rates that comprise various network products in their MRFs. Therefore, it is possible to obtain the payment rates by service code for every individual and group health plan, including by self-funded employer. This information is important for competitive analysis but should also be of interest for employers that are relying on TPAs to post accurate transparency data on their behalf.

As plan fiduciaries, employers may use transparency data to better understand the contractual payment rates included in their TPA arrangements to support TPA evaluations and due diligence. Employers may be interested to understand the rates posted on their behalf by their TPAs to evaluate compliance with the regulations and understand how the price data may be interpreted by employees and individual consumers, especially in regard to price data from other TPAs. The price transparency data is very dense and cumbersome, and many employers do not have analytics departments with the capacity to analyze the information. With Milliman Transparent, employers can search the data for their EIN and quickly review the prices for top service codes and compare them to other self-funded employers.

Conclusion

With Milliman Transparent, employers have tremendous opportunity to gain a deeper understanding of payer networks, actual negotiated payment rates by TPA and provider, and pricing relationships that historically have been considered proprietary and confidential.

The price transparency data published by hospitals and payers opens a new frontier of analytics to help plan sponsors fulfill their duties as plan fiduciaries and continue to create sustainable, high-quality benefits programs for their employees.

Caveats and limitations

The observations and ideas presented in this paper reflect a point-in-time analysis based on the current information collected and reviewed. Files and file content may have been updated since retrieval.

The analyses presented in this paper are intended to illustrate how transparency data can potentially be used. They are not to be relied upon outside of this illustrative context.

The data presented in this paper is only a subset of the data available at each facility or payer displayed. As such, the results of these limited comparisons should not be interpreted as indicators of any broad contracting relationships or trends.

The estimates included in this paper are not predictions of the future; they are estimates based on the assumptions and data analyzed at a point in time. If the underlying data or other listings are inaccurate or incomplete, the results may also be inaccurate or incomplete. These estimates are based on negotiated payment rate information. Other important considerations for employers such as provider quality and efficiency are not reflected in these estimates.

Throughout this analysis, Milliman relied on data and other information provided by publicly available data sources. Milliman obtained standardized HPT (hospital) and TiC (payer) price transparency data from Turquoise Health (https://turquoise.health/) to support the analyses in this paper. Milliman has not audited or verified this data and other information but has reviewed it for reasonableness. Models used in the preparation of our analysis were applied consistent with their intended use. We have reviewed the models, including their inputs, calculations, and outputs, for consistency, reasonableness, and appropriateness to the intended purpose and in compliance with generally accepted actuarial practice and relevant actuarial standards of practice (ASOP).

Guidelines issued by the American Academy of Actuaries require actuaries to include their professional qualifications in all actuarial communications. Erica Reijula, FSA, MAAA, Spencer Marshall FSA, MAAA, and Mike Gaal, FSA, MAAA are members of the American Academy of Actuaries and meet the qualification standards for performing the analyses in this paper.

¹ Federal Register, Vol. 84, No. 229 (November 27, 2019). Medicare and Medicaid Programs: CY 2020 Hospital Outpatient PPS Policy Changes and Payment Rates and Ambulatory Surgical Center Payment System Policy Changes and Payment Rates. Price Transparency Requirements for Hospitals To Make Standard Charges Public. Final Rule. Retrieved August 20, 2024, from https://www.govinfo.gov/content/pkg/FR-2019-11-27/pdf/2019-24931.pdf.

² CMS. Transparency in Coverage: Final Rule. Retrieved August 20, 2024, from https://www.cms.gov/CCIIO/Resources/Regulations-and-Guidance/Downloads/CMS-Transparency-in-Coverage-9915F.pdf.

³ Reijula, E., Smith, C., Ochsner, A., & Hall, E. (September 2023). Price Transparency in 2023. Retrieved August 20, 2024, from https://www.milliman.com/en/insight/price-transparency-in-2023.

⁴ KFF (October 18, 2023). 2023 Employer Health Benefits Survey. Retrieved August 20, 2024, from https://www.kff.org/report-section/ehbs-2023-summary-of-findings/.

⁵ KFF (October 18, 2023). 2023 Employer Health Benefits Survey. Retrieved August 20, 2024, https://www.kff.org/report-section/ehbs-2023-section-10-plan-funding/.

⁶ Marshall, S., Zhou, D., Anderson, C., & Mills, C. (June 19, 2024). Commercial Reimbursement Benchmarking. Milliman White Paper. Retrieved August 20, 2024, from https://www.milliman.com/en/insight/commercial-reimbursement-benchmarking-medicare-ffs-rates.

⁷ Price Transparency in 2023, op cit.