Considering trend guarantees in your next TPA selection analysis

  • Print
  • Connect
  • Email
  • Facebook
  • Twitter
  • LinkedIn
  • Google+
By Scott M. Cohen, Paul V. Sakhrani, Brian A. Sweatman | 01 May 2019


Self-insured employers have long evaluated provider discounts offered by medical third-party administrators (TPAs) to assess the opportunity to reduce the cost of providing health insurance benefits to their employees.

A provider discount analysis estimates the difference in expected allowed claims costs by applying provider discounts from competing TPAs to an employer’s historical billed charges. Provider discount analyses are generally thought to do a reasonable job of estimating the expected change in discount upon switching TPAs, but there are several caveats to consider when interpreting the results:

  • The mix of services and/or providers underlying each TPA’s book of business data (and resulting discounts) may differ from an employer's.
  • TPAs may not use the same methodology when compiling discounts. One important methodological component is what claims to include. For example, if one TPA includes claims that are excluded by another (e.g., due to a bundled claims payment arrangements), a true apples-to-apples comparison may be difficult.
  • Provider discount analyses are not designed to measure the impact on utilization that may occur when switching from the utilization and care management programs of one TPA to another.

As a result of these concerns, there is increased interest from both employers and TPAs in incorporating medical trend guarantees into TPA selection analyses. A trend guarantee may be broadly defined as an agreement between a TPA and an employer that compensates the employer in the event that the year-over-year trend in medical claims costs exceeds the negotiated amount. Though employers may intuitively expect that the use of trend guarantees will make the analysis simpler (i.e., take the bidding TPA that guarantees the lowest trend), there are many issues to consider for the self-funded employer:

  • The bidding TPA will generally only be putting a portion of the administrative fees at risk. While administrative fees may be substantial—in the range of about $40 per employee per month (PEPM)—the fees may only represent about 5% of the claims spend of most employers. Even if a TPA is willing to put 20% of its administrative fees at risk, that only represents 1% of total costs. In the event that actual trend is several percentage points higher than the guaranteed trend, the employer would only recoup a small portion of the difference via rebate of administrative fees.
  • The bidding TPA will also likely attempt to limit its exposure on administrative fees via a tiered payout schedule that is calibrated on both the percentage of total fees at risk and the magnitude of the trend “miss.” For example, a payout model may reference payback of 5% of administrative fees if the trend guarantee target is missed by between one and two percentage points and 10% if missed by between two and three percentage points, etc., but capped at a payout of no more than 20% of administrative fees. Given the likely scenario that a TPA will limit its exposure to pay back administrative fees, employers will still need to be prudent in understanding the differential in expected allowed charges among the TPAs.
  • TPAs confident that they have a sizable provider discount advantage versus the incumbent may be willing to offer a greater percentage of fees at risk.
  • TPAs may also be willing to offer a greater percentage of administrative fees at risk if there are mechanisms in place to limit the variability of expected claims (addressed in the next section).
  • TPAs may be willing to offer a multiyear trend guarantee, which can help limit the potential increase in future years.
  • With all guarantees, users must understand the fine print and how the requirements, caveats, and assumptions may affect the overall guarantee.

We expect that TPAs offering trend guarantees will also attempt to control their exposure via reference to the following types of issues:

  • Changes in demographics and/or geographic distribution of enrollees: While expected healthcare costs can vary by age, gender, and geographic location of insureds, the TPA has no influence over the year-to-year fluctuations in enrollment and may wish to apply adjustments to normalize for differences in these variables over time.
  • Risk profile of enrollees: In addition to the adjustment factors that may exist for expected variation in healthcare costs for different age, gender, and geographic location, there may also be adjustment factors used to measure differences due to the clinical conditions present in the population. If the TPA is providing utilization and care management services to the employer, it has some level of responsibility with respect to mitigating increases in the employer population’s morbidity, but the TPA does not have influence over the enrollment and disenrollment of members from one time period to the next. It may wish to use risk adjustment to account for these changes over time.
  • Changes in plan design: Increases or decreases in benefit richness have a direct effect on expected claims costs. The employer is responsible for determining the benefit richness of its plan offerings and, therefore, TPAs may apply adjustments to control for the variation in claims attributable to plan design modifications.
  • Introduction of new services and technologies: While there are likely positive long-term effects originating from these new services and technologies, there are often very high price tags in the short-term. TPAs may apply explicit claims exclusions until these services and technologies are present in both the baseline and performance periods of the trend guarantee.
  • Influence of large claimants: Due to the statistical volatility of large claims, the TPA may apply an exclusion for members who exceed a certain threshold. The TPA may remove the entirety of these members’ claims or they may remove just the portion that exceeds the threshold. The threshold assigned may be variable based on the size of the covered population, with larger populations receiving relatively higher thresholds.


We expect employers to see an increase in the availability of trend guarantees. While trend guarantees may offer a useful hedge against unexpected increases in costs, employers should be diligent in understanding the fine print. A trend guarantee with a lot of caveats and a tiered payout schedule may not have a material impact on the value proposition offered by the TPA submitting guarantees.