BCBS: Hospitals Use AI to Inflate Bills via 'Bump Codes' — SkimNews
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- Blue Cross Blue Shield published research finding that hospitals increasingly use AI tools to document secondary diagnoses like anemia, malnutrition, and low sodium, costing the insurer hundreds of millions of dollars over the past two years.
- Hospitals exploit "bump codes" — secondary diagnoses billed under Medicare's diagnosis-related groups (DRG) system — to charge more for a hospitalization even when the secondary condition isn't actively treated.
- The Centers for Medicare and Medicaid Services last substantially overhauled its list of qualifying bump codes in 2007; a 2019 CMS proposal to downgrade or eliminate many bump codes stalled after hospital industry pushback.
- The physician author describes automated clinical-documentation-integrity (CDI) tools that scan patient charts and prompt doctors to record secondary diagnoses, saying they answered honestly but still worried the outputs inflated patient bills.
- BCBS stops short of alleging fraud — the central question, per the article, is whether documented lab abnormalities like borderline low sodium genuinely cost hospitals more to treat, or merely trigger higher reimbursements.
- The author calls for a structural fix: CMS overhauling which secondary conditions qualify for bump coding, and hospitals building more nuanced documentation tools that distinguish clinically significant findings from incidental lab values.
Why it matters: The article reframes the AI-and-health-costs debate: the real driver is a CMS billing system whose bump-code list hasn't been meaningfully updated since 2007, and both hospitals and insurers are now deploying competing AI tools (documentation queries vs. payment-integrity audits) that add administrative cost without improving care — costs that ultimately pass to patients via higher premiums.
Ask SkimNews
