The Bureau of Labor Statistics released the June 2026 CPI on July 14, with all-items running 3.5% year over year. Buried under the headline is a split that matters more to a workers’ compensation reserve than the top-line number: medical care services rose 2.9% over the 12 months ending June, while medical care commodities fell 2.1%. Drug deflation pulled the blended medical care index down to roughly 2.0%, from 2.6% in May.
That blended figure is the wrong anchor for a WC medical trend. Prescription drugs are a small and shrinking slice of the workers’ comp medical dollar; physician, hospital, and physical therapy services dominate it. A trend assumption that leans on blended medical CPI borrows the drug line’s deflation, and understates the services inflation that actually drives paid medical development.
Who it affects
Self-insured WC programs carrying open lifetime medical exposure: public entities, hospital systems, universities, large manufacturers, transit authorities, and single-parent captives writing comp. The exposure is highest for programs with a tail of catastrophic claims (traumatic brain injury, spinal cord, severe burns) where the medical reserve sits on a structured life-care plan built almost entirely from physician, hospital, and therapy services, not pharmacy.
The reserve mechanism
The lever is the medical severity trend factor applied to open case reserves, and the trend implicit in the expected claim ratio used for current exposure periods. Anchor that factor to a 2.0% blended medical print and you are selecting roughly 90 basis points below the 2.9% services line that governs where the dollars go. On a long-duration claim discounted and projected over 20 or 30 years, that gap compounds: a mis-selected trend understates both the ultimate and the tail factor that carries the reserve past the last mature development period. The chain ladder will not flag it as a trend error; it reads the eventual catch-up as adverse paid development.
The indemnity side reads off the all-items 3.5% figure, which feeds wage-based benefit escalation and, in states with cost-of-living adjustments, the annual bump on lifetime indemnity. That is a separate trend from medical and should be selected separately.
Where this shows up in your reserves
Open the paid medical development on your last WC review and find the trend exhibit. If the selected medical trend is a single blended number pulled from CPI medical care, it is diluted by drug deflation. The tell is a medical severity assumption near 2.0% to 2.2% while your actual paid-per-open-claim on serious cases is climbing faster. Cross-check the CPI print against NCCI’s Workers Compensation Weighted Medical Price Index, which ran 1.8% in early 2026 but which NCCI itself warns is unlikely to hold. Both the CPI services split and the NCCI benchmark point the same way: the services side is where the severity lives.
What this means for your next review
Ask your actuary to separate prescription drug deflation from services inflation in the severity analysis, and to justify the selected medical trend against a services-weighted benchmark rather than blended medical CPI. This is the same composition problem the March CPI print surfaced, now sharper because commodities turned negative.
Decision-maker checklist
- Confirm which index your latest WC reserve review uses for medical severity: blended medical CPI, or a services-weighted benchmark.
- Run the sensitivity: what does a 100 bps change in the medical trend do to your tail factor and IBNR on lifetime-medical claims.
- Ask whether pharmacy deflation is netted against services inflation inside a single blended trend, and unbundle it if so.
- Watch the July CPI (August release) and the next NCCI medical trend update to confirm the services-commodities divergence persists.