AI-Powered Billing Tools Drive Healthcare Costs to 17-Year High, PwC Report Finds
核心洞察
Commercial healthcare costs are projected to rise 9% in 2027, the highest medical cost trend in nearly two decades, according to PwC (搜索)'s analysis of 27 health plans covering over 103 million members.
Nearly 70% of surveyed health plans ranked providers' use of AI documentation and coding tools as a top-three cost inflator, with about 20% calling AI the number one inflationary trend.
Blue Cross Blue Shield (搜索) analysis found approximately $663 million in inpatient spending and $1.67 billion in outpatient spending may be tied to AI-powered coding practices.
Commercial healthcare costs are projected to climb 9% next year, the highest medical cost trend in nearly two decades, driven in part by providers' rapid adoption of artificial intelligence billing and coding tools, according to a new report from professional services firm PwC (搜索) released June 11.
The analysis, which surveyed and interviewed actuaries at 27 U.S. health plans representing more than 103 million employer-sponsored members and 8 million individual ACA marketplace members, identified five key inflators pushing costs upward. Among them, AI-enabled revenue optimization tools emerged as a notable and somewhat unexpected contributor.
Nearly 70% of surveyed plans ranked providers' use of AI documentation and coding products as a top-three inflator for next year, while approximately 20% identified AI as the number one inflationary trend. The individual market is projected to see an 8.5% cost increase.
How AI Tools Increase Reimbursement
The mechanism behind AI-driven cost inflation is straightforward: AI-powered documentation tools allow clinicians to record greater specificity in patient encounters, capturing more diagnoses, comorbidities, and complexity that support higher-level billing codes. More complete documentation translates to higher reimbursement — even when the underlying clinical work or treatment remains unchanged.
"Providers are using AI-enabled documentation and coding tools to record greater specificity and reimbursable severity, which means payers see higher paid amounts per claim," the PwC (搜索) report states.
Glenn Hunzinger, U.S. health industries leader at PwC (搜索), emphasized that this does not necessarily indicate inappropriate coding. "The ability to use technology and AI to more appropriately code or code things that they were never able to, that's the trend we're seeing," Hunzinger said. "It does have an impact on that 9%, albeit it's not the biggest piece."
Many health systems were likely missing correct codes previously due to the volume of care and the complexity of their internal systems, Hunzinger noted. Additionally, thin hospital margins and increased reimbursement pressure — particularly amid federal healthcare spending cuts to Medicaid — may motivate providers to adopt AI billing tools to remain financially viable.
Quantifying the Impact
A separate analysis from Blue Cross Blue Shield (搜索)'s Blue Health Intelligence unit examined de-identified claims from tens of thousands of maternity admissions and found sharp increases in acute posthemorrhagic anemia coding at some hospitals, without a corresponding increase in transfusions. The rise in that single diagnosis added $22 million to maternity admission costs in one year.
The Blue Cross report further estimated that approximately $663 million in inpatient spending and at least $1.67 billion in outpatient spending may be tied to AI-powered coding tools. The analysis also cited a 2023 HFMA and AKASA survey finding that 46% of hospitals and health systems already used AI in billing, coding, and claims, while federal data showed AI use for billing jumped 25 percentage points year over year.
Broader Cost Drivers
AI is not the largest contributor to rising healthcare costs. Labor and supply cost inflation, lingering from the COVID-19 pandemic, remains a major driver. Provider consolidation also fuels the trend, with nearly 65% of survey respondents ranking contracting pressure from larger, merged health systems as a top-three inflator.
The Independent Dispute Resolution process established by the No Surprises Act — the 2022 law protecting patients from unexpected out-of-network bills — represents another inflationary force, as providers win the vast majority of disputes and earn more than they typically would for care.
Rising pharmacy spending, including for expensive GLP-1 medications, and increased demand for behavioral healthcare further contribute to the upward trajectory. Meanwhile, deflationary forces such as biosimilars, generic drugs, and site-of-care shifts to cheaper facilities are already factored into current projections and are unlikely to drive additional improvement.
The Incentive Problem
The findings highlight a structural challenge: AI enters a healthcare system still largely built on fee-for-service payment models that reward doing more and billing more. As the PwC (搜索) analysis notes, a startup that helps a hospital code more precisely may describe itself as a productivity company, but the buyer may experience the product as a revenue tool.
Venrock partner Bob Kocher articulated the economic tension: if AI is applied inside fee-for-service medicine, it will likely increase units of care and coding. If payment models instead reward lower total cost, the same technology could be pushed in a different direction.
Harvard Medical School associate professor Hossein Estiri suggested that attention may eventually shift away from administrative AI toward tools that improve patient outcomes.
The Payer Response
PwC (搜索) recommends that payers improve coding-intensity surveillance, severity-shift monitoring, payment integrity, and pre-payment review of high-dollar claims. The goal, the firm states, is more accurate payments rather than more denials — a distinction that carries weight given that insurers themselves face lawsuits over alleged algorithmic claim denials, with UnitedHealth Group and Cigna both facing legal challenges.
Hunzinger maintained that AI could eventually help bend the cost curve, particularly by automating administrative tasks that account for a significant portion of healthcare spending and by reducing provider burnout. However, he acknowledged that innovation takes time to reduce costs in a highly regulated industry. "It's a tough environment to operate in, in a flawless fashion," he said. "Because at the end of the day, it's patients, it's people. So that's why it probably takes a little bit longer."