Edition

Give Me Some Credibility: Addressing the Challenge of Volatility in Value-Based Contracts—Part 2

By James Pisko, Maria Knox and Keith Passwater

Health Watch, May 2023

hw-2023-05-knox-hero.jpg

Value-based care (VBC) in the US health care market is a mechanism that aligns provider and payer financial incentives to reduce cost. This has led to shifting insurance risk[1] from health insurers to health care providers. However, most provider groups do not have much experience managing insurance risk, so in Part 1 of this series, we introduced the unique aspects of VBC contracting that may significantly affect credibility: smaller provider risk pools, indirect risk exposures and constraints around performance measurement choices.

In this article, we build on these concepts and apply them to the key components of a VBC contract:

  • Benchmarks: pros and cons requiring prudent choices
  • Measurement period results: the possibility of decreasing credibility/increasing volatility, especially segmenting the covered population for measurement
  • Limiting risk transfer: various methods that generally reduce the degree of risk transfer and increase credibility
  • Methods for assessing credibility: the same statistical methods for payers and providers

Credibility in VBC: Benchmarks

Sufficient size of the underlying population (credibility) is critical for benchmark accuracy. But two additional elements are just as important for the success of a VBC contract: benchmark appropriateness and provider buy-in. Benchmark appropriateness can be ensured by using a population that is a representative sample.

Just as important, the benchmark development process needs to be transparent and understandable by the provider partner so they will feel comfortable participating. This contrasts with traditional pricing projections, where a member of a health plan has little visibility or concern over how the payers determine credible experience for pricing methodology.

Several approaches enhance credibility in benchmark setting in VBC contracts. The following is a summary of their prevalence and the pros and cons of each.

Retrospective Book of Business Trend

The benchmark is the year-over-year trend of the mix-adjusted larger plan book of business.

Prevalence:

  • A preferred approach by providers

Pros:

  • Enables providers to directly compare their own claim distribution against the rest of the market

Cons:

  • The benchmark is a continuously moving target, which introduces additional volatility in forecasting year-end results.
  • Additional budget risk to the plan when the entire book of business has less favorable experience than the overall budgeted pricing trends. In addition to having to cover these losses, the plan may need to pay on VBC contracts that have performed better than the unfavorable book of business trend.

Prospective Benchmark with Multiple Years Used in the Base Period

The base period for developing benchmarks consists of two or more years.

Prevalence:

  • A preferred approach by both plans and providers

Pros:

  • Increases base period credibility
  • Can improve credibility enough that a larger provider's own prior period experience can be used directly, increasing provider trust in the program

Cons:

  • Many providers may still lack sufficient membership for full credibility.
  • Care cost patterns from previous years may not represent the current state, thus creating an inaccurate benchmark.

Prospective Benchmark with “Manual Rates” Credibility Weighing

The base period is a blend of the providers' own cost-of-care PMPM experience with that of the larger payer book of business. This is a widely used approach in traditional pricing, where the broader book of business is leveraged to develop factors to normalize a less credible population.

Prevalence:

  • Rarely used in the VBC programs of our experience

Pros:

  • Provides a high credibility base for a statistically accurate target

Cons:

  • Providers desire the benchmark to be tied directly to their specific performance. As with commercial groups, providers find it more intuitive to think about projected future costs in terms of their own underlying data. Introducing what actuaries consider relevant experience to enhance the predictive power of projected claims makes it more challenging for providers to trust. Questions of the form “it's not clear or fair that I should be measured based on claims data from other provider systems” are not uncommon.
  • Providers do not have access to the broader plan book of business data. This method makes them reliant on payers to assess credibility of their own historical data and whether it would be appropriate to blend it with a broader data set to increase credibility of the final estimate.

Credibility in VBC: Measurement Period Results

Measurement year provider performance should also consider credibility under VBC contracts. Reasons for this are:

  • The risk pool is limited to members attributed to that provider in the one measurement year.
  • Blending with a manual rate is not performed.
  • Further splits of the provider’s measurement period financials exacerbate the insufficient credibility problem. Common segmentations are by line of business, age groupings, region, funding type and so on.

There are no apparent direct mitigation techniques for the limited risk pool or the lack of manual rate blending once a provider has made the decision to participate in a risk-bearing contract. But there is a mitigation approach to address the lack of credibility due to population segmentations.

Mix Adjustment Factors

Splitting measurement period experience by additional segments is used to address significant changes to the member mix between the base period used in setting the benchmark and the measurement period. This unfortunately produces very narrow “rate cells” with low membership/credibility and volatile claim experience.

A viable mitigation approach is to address changes to member mix by mix adjustment factors that are based on the larger plan’s book of business.

Volatility in VBC: Limiting Risk Transfer

Insufficient credibility due to the factors discussed so far results in relatively higher volatility in performing VBC contracts. As a result, plans have developed methods to limit the level of financial risk to the VBC participating provider, where most of the risk stays with the health plan.

Following are the most common approaches for shared savings/loss arrangements.

Outlier Claims Capping

Claims experience for an episode or member that is in excess of a specified threshold is removed from evaluation in the annual shared savings or losses of the contract. This is due to the relatively less predictable nature of these (outlier) claims that can have a disproportionate impact on average claims PMPM. With that said, certain VBC shared savings programs have attempted to fully incorporate high claims by pricing and evaluating the contract experience separately. Shared savings and loss caps (next section) have always been used in conjunction with high claims to limit their impact on the overall savings or losses.

Shared Savings/Loss Caps

Providers are not responsible for losses (do not participate in savings) outside a specified range. Usually this specified range logic is defined and applied on a per member basis but can also exist for aggregate payments.

Minimum Savings/Loss Rates

The plan and provider do not share in savings and/or losses unless they pass a certain threshold. This approach is designed to ensure there is clear “signal” in the results before parties agree to a financial transfer. This method also estimates administrative expense associated with accounting for small amounts of variation from the target.

Credibility in VBC: Methods for Assessing Credibility

An important function of actuaries on behalf of payers and risk-bearing providers is to evaluate the credibility of historical loss data to understand how useful they would be in estimating future loss data. These methods have generally been based on mathematical models or judgment.

The first step may often be observation based on historical claims PMPM experience. Comparing the PMPM value of claims over time to get a directional level of understanding for volatility can help give an early indication of the credibility of claims data. Note that payers with mature books of business may be able to rely on historical credibility analysis to understand how credible a particular block of business is. However, this credibility assumption may not be valid for providers that are new to risk-bearing VBC contracts.

ASOP 25, “Credibility Procedures,” outlines the scope of credibility procedures that can be relevant for use. It is the responsibility of the actuary to determine which methods are most appropriate.

Generally, these methods are as relevant for traditional insurer settings as they are for provider systems when evaluating cost metrics under VBC contracts. The purpose and metric being evaluated for credibility is largely the same in both situations, with the differences related to the differing degrees of influence providers have over the course of treatment, associated cost and quality for patients attributed to a value-based contract.

Conclusion

In the effort to mitigate high health care costs while maintaining or improving quality, VBC remains a promising mechanism. Still, volatility due to a lack of credibility can discourage both plans and providers from committing to it.

This article has outlined approaches for managing volatility so that both providers and payers can be successful and the main objective of VBC contracts is reached—improving health care affordability while ensuring a high quality of care.

Statements of fact and opinions expressed herein are those of the individual authors and are not necessarily those of the Society of Actuaries, the editors, or the respective authors’ employers.

James Pisko, FSA, MAAA, is an actuary at Nuna. James can be reached at jamesp@nuna.com.

Maria Knox, FSA, MAAA, is an actuary at Nuna. Maria can be reached at maria@nuna.com.

Keith Passwater, FSA, MAAA, FCA, is the CEO of Havarti Risk Services. Keith can be reached at kpasswwater@havartirs.com.

[1] “Insurance risk” is the risk associated with the unknown variation in the utilization and cost of services. From Juliet Spector, Cory Gusland and Carol Kim, Insurance Risk and Its Impact on Provider Share Risk Payment Models (Schaumburg, IL: SOA, 2018). https://www.soa.org/globalassets/assets/files/resources/research-report/2018/insurance-risk-impact.pdf.