The Bureau of Labor Statistics released March 2026 CPI data on April 10. Medical care services rose 3.1 percent year over year, decelerating from 3.4 percent in February, while the hospital and related services subindex climbed another 0.4 percent on the month. The release landed roughly a week after most calendar-year reserve studies were priced, so the new print is the first post-study data point for self-insureds running 12/31/2025 evaluations.
For workers’ compensation self-insureds carrying long-tail indemnity and lifetime medical cases, the headline deceleration narrows the gap between assumed medical trend and realized trend. But the composition matters more than the headline, and hospital services (the component that dominates severity on serious-injury claims) has not decelerated with the blended index.
Who it affects
Self-insured WC programs with meaningful open lifetime medical exposure: public entities, hospital systems, large manufacturers, transit authorities, and single-parent captives writing WC. The sensitivity is highest for programs with a tail of catastrophic claims (quadriplegia, traumatic brain injury, severe burns) where the medical reserve sits on a structured life-care plan and the annual escalator was set when that plan was priced.
The reserve mechanism
The primary effect is on the medical severity trend factor applied to open case reserves, and on the implicit trend embedded in the expected claim ratio (ECR) used for ongoing exposure periods. A 30 bps deceleration in the blended medical index, if it holds, reduces the recalibration gap on claims where the reserve was set against a 4 percent to 5 percent medical trend assumption. It does not close the gap on claims priced in 2022 and 2023, when hospital services was running above 6 percent; those reserves are still catching up, and the chain ladder will read that catch-up as adverse paid development rather than as a trend correction. The Berquist-Sherman restatement is one tool for separating the two.
A secondary effect runs through the NCCI loss cost cycle. Connecticut approved a 3.8 percent decrease for 2026 and New Hampshire approved a 6.1 percent decrease, both effective January 1. When loss cost reductions run ahead of realized medical inflation, the manual premium benchmark that self-insureds use for internal loss picks drifts below the underlying cost trajectory, and that distortion flows through any BF or expected-claims method anchored to a manual-rate ECR.
A third effect sits in Medicare Set-Aside projections. CMS trend factors lag BLS prints, so an MSA priced off 2024 or 2025 CMS guidance will carry a different medical escalator than a reserve set today off the March print. Claims with both a carried medical reserve and a pending MSA can show two different ultimates for the same future medical stream; the pure versus broad IBNR distinction is where that mismatch usually surfaces.
What to ask your actuary
- What medical trend vector are you using for open lifetime WC claims by age of injury, and how did the March print change the central estimate for accident years 2020 through 2023?
- If hospital services continues to run above the blended medical CPI for another two quarters, what IBNR strengthening does that imply on our longer-duration claim segments?
- How does the gap between NCCI’s 2026 approved loss cost decreases and realized medical inflation affect our internal loss pick versus the manual rate benchmark?
What to watch next
The April 2026 CPI release on May 12, and whether the hospital services monthly rate stays above 0.3 percent for a fourth straight month. A sustained reading at that level would argue for decomposing the medical trend selection into a weighted blend of CPI-M and hospital services rather than continuing to anchor on the headline index. Also watch for NCCI’s next State Advisory Forum update, which will signal whether the 2026 loss cost decreases hold for 2027 filings or reverse as the hospital component feeds through to paid medical on accident-year triangles.