Misreporting and/or under-reporting of hospital acquired infections (HAIs), particularly surgical site infections (SSIs) is a well-known phenomenon (see Mahmoud, Surgical Infections 2009 and Smith, Annals of Surgery, 2004). Although it cannot be said that hospitals intentionally misreport these statistics, it is clear that with the arrival of reimbursement penalties and public reporting of complication rates on HospitalCompare.gov, there are significant pressures to minimize reported infection rates.
A recent study conducted at the Stanford Graduate School of Business led by Mohsen Bayati further confirms this reality, and points to a practice of “upcoding,” where HAIs are misreported to increase reimbursement or avoid financial penalties. The study looked at state-level variations in adverse event reporting regulations to discover discrepancies between HAI and present-on-admission (POA) infection reporting rates that might be suggestive of upcoding. “Strongly regulated” vs. “Weakly-regulated” state results were compared while controlling for a wide range of patient risk factors and demographic characteristics, demonstrating that rates of POAs were significantly less than the rate of HAIs in state with stronger regulations. Of critical importance here is the fact that POA infections are not included in the database that driving reimbursement penalties and public reporting.
It is concerning that cases of CA-UTI (the most common type of hospital-acquired infection) are rarely identied in [claims data]… In addition, coders often listed [infections] as present on admission, although the medical record indicated that it was hospital acquired… Because coding seems to be fraught with error, nonpayment according to CMS policy may not reliably occur.” – Meddings et al. (2010)
The pressure to “upcode” is associated with a very significant negative externality, penalizing truthful hospitals with higher financial penalties and tarnished public reputation. This practice is doubly impactful from an policy and economic perspective. Not only does it lead to under-reporting of true HAI rates, reducing potential reductions in Medicare expenditures and all but eliminating the financial incentives intended to decrease HAIs, but because POA infections are over-reported and reimbursed normally, it results in a further increase in overall Medicare expenditures. Bayati estimates this increase to total $200 million across 10,000 instances of over-reported POAs.
Putting our finger on the true magnitude of HAIs, specifically SSIs, is a key goal of this blog. Though we’ve found that many surgeons have a realistic view of their infection rates, all too often we hear surgeons and hospital administrators boast a common refrain: “We rarely have infections here.” Really?