How is Your DOFR Doing?

Purpose of DOFRs

A Division of Financial Responsibility (DOFR) is a provision in a contract between a health care provider and a health plan or payer that defines which party is financially responsible for providing specific services. DOFRs are useful for shared savings, shared risk, and global capitation arrangements because they define the services that the provider will be held financially responsible for. A sample excerpt is shown in Table 1 below:

Table 1: Sample Excerpt from a Division of Financial Responsibility

There may be more than two columns for split capitation arrangements where there are more than two parties sharing the risk; for example, there could be columns for the health plan, the contracting hospital, and the contracting medical group or IPA. In such an arrangement, the hospital and the IPA would be responsible for different services and would have separate capitation payments or shared risk arrangements with the health plan.

Negotiating DOFR Changes

The provider should have regular opportunities to renegotiate their contracts with each payer. Before approaching the negotiation table, though, the provider should have an idea of how well they are faring under the current contract. Evaluating performance under a contract can be a complicated process as there are many interrelated provisions that have an impact on the bottom line: there may be withholds for quality incentive programs, stop-loss arrangements, and/or fee-for-service payments received for certain services that are not covered in the DOFR. However, it’s important for providers to know how they are currently faring under a contract so that the current performance can be used as a benchmark by which to evaluate any proposed changes.

The DOFR itself is an essential tool in evaluating the impact of proposed contract changes. A few items are needed in order to complete the contract analysis:

• Claims data for the covered population or a substantively similar population
• An actuarial cost model, which demonstrates utilization and unit cost data on a service-level basis as well as an overall basis
• The current DOFR and the proposed DOFR

First, the claims data must be incorporated into the cost model so that the per-member-per-month (PMPM) results are organized in an easy-to-digest format. Then, for each service category, the analyst can map the historical claims to the financially responsible entity. The result will be detailed PMPM costs for both the health plan and the provider, as shown in Table 2 below. This process can be performed for the current and proposed DOFRs to identify the magnitude of any changes.

Table 2: Sample Excerpt from Cost Model Adjusted to Reflect DOFR Responsibilities

Once this process is complete, it can be used to evaluate proposed rates as well. The percentage difference in the provider’s share of the PMPM between the current DOFR and the proposed DOFR should be reflected in the proposed rate change.

For example, say the universe of potential services is restricted to those shown in the sample excerpt from Table 2. Then the total cost to the provider is $3.10 PMPM. If under the new contract the provider’s expected cost is $3.25 PMPM, the provider should expect about a 4.8% increase to their capitation (calculated as $3.25 / $3.10).  However, the provider should keep in mind that rate changes will also reflect other rating factors, such as trend, demographic or geographic adjustments if applicable, and applicable regulatory changes. It is not unreasonable to ask the health plan for a breakdown of the factors influencing the rate change.

“It is not unreasonable to ask the health plan for a breakdown of the factors influencing
the rate change.”

Evaluating New Contracts

This framework can also be used to evaluate contracts with new payers, rather than just existing contracts. Comparing the DOFRs from two different payers on an apples-to-apples basis using a cost model can help the provider ascertain whether they expect to succeed under a new contract, by comparing the rate relativities and the “DOFR richness” relativity between an existing and a new payer relationship. An example of this comparison is shown in Table 3:

Table 3: Using DOFR Relativities to Evaluate a New Contract (Sample Excerpt from Cost Model)

Again, if we assume this subset of categories represents the entire universe of services, we find that the expected PMPM for Payer 1 is $3.10, while the expected PMPM for Payer 2 is $3.87. All else equal, the provider should expect a capitation rate from Payer 2 that is 24.8% higher than the capitation rate from Payer 1.

Pitfalls to Avoid

There are some caveats to using the above approach for evaluating both existing and new contracts. In particular, the analyst should be aware of the following:

• Claims fluctuation risk(or the inherent volatility of claims): ideally, this analysis would be performed on a pool of members that is large enough to be fully credible, which would mitigate the impact of claims fluctuation risk. This scale might not be possible for smaller providers. The issue with using partially credible data is that it is less likely to accurately predict future performance; there will always be claims volatility, but the expected volatility is reduced as the volume of input data increases. There are a few solutions for cases where data is scarce:
  • The small dataset can be used with an understanding that a wide range of results is possible. The range of results can be estimated using statistical techniques.
  • The provider may pool data from members covered under other, similar contracts. Close attention should be paid when pooling data across separate contracts, as some differences between contracts may preclude the data being comparable.
  • The health plan may have a contractual obligation to deliver the data used to develop proposed payment rates to the provider. However, this is not always the case and the data may not be at the level of detail needed to perform this type of analysis.
  • The analysis can be contracted out to consultants, who generally have access to large datasets for many population types.
  • The provider can establish a risk corridor. While this doesn’t necessarily help with the analysis, it is helpful for providers with risk-based contracts and small populations to mitigate the financial impact of a very bad year.
• Changes in the risk profile: expected claims can be affected by shifts in the risk profile of the covered members between the base period (the coverage period for the members included in the analysis) and the measurement period (the dates the future contract will be in effect).
  • This can be addressed by adjusting for the change in risk score between the base period and the measurement period. However, the analyst should take care to note that while the risk score is appropriate for adjusting on an aggregate level, it may not be appropriate for adjusting on a service-specific level. As an example, think of a case where the overall risk score decreases from 1.0 to 0.9 between periods. It might be appropriate to expect inpatient visits will decrease by 10%. But is it reasonable to expect that maternity services or immunizations will also decrease by 10%?
• Definition of covered services: Some DOFRs include the codes that correspond to each covered service (including HCPCS, CPT and revenue codes). However, most DOFRs do not currently include the codes corresponding to each service. This makes analysis difficult because service categories might be broad and vague, which makes it difficult to accurately map claims to categories. Moreover, different payers might use different logic for which claims to include in which categories, even if the same term for a given category is used in each contract.
  • The issue of broad and/or vague service categories can be alleviated by the provider requesting the payer’s mapping logic.
  • If the payer will not provide the exact logic, the provider can use other tools to map claims to categories. Consultants usually have generic, broadly applicable mapping criteria. There is also a free coded DOFR template available for Industry Collaboration Effort (ICE) members[1].
• Comprehensiveness and accuracy of data: the analysis is only useful to the extent that the data input is accurate and complete. The nature of capitation contracts implies that the available data will generally be encounter data. There are notorious issues with encounter data[2]. These issues can result in incomplete or inaccurate data.
  • This issue can be alleviated by attempting to accurately capture all services performed under capitation contracts. Some contracts include provisions for accurate reporting of encounter data, but there is generally room for further improvement.

Conclusion

Regular evaluation of performance under a given contract is not only useful as a general business assessment. It can also serve as a valuable tool for providers about to enter into contract negotiations. With enough data and supporting analysis, the provider can determine whether the rate they are being paid is generous, insufficient, or relatively fair in comparison to a benchmark.

[1]“ICE – Coded Division of Financial Responsibility Now Available”, Industry Collaboration Effort, 13 February 2017, https://iceforhealth.org/messagedetail.asp?mid=6853

[2]“Challenges in Encounter Data Submissions”, Integrated Healthcare Association, June 2018, https://www.iha.org/sites/default/files/resources/encounter_data_white_paper_final.pdf