Research Activities, June 2013
Hospitals rarely describe urinary tract infections as catheter-associated urinary tract infections when coding claims
Patient Safety and Quality
Nearly a third (32 percent) of all hospital-acquired infections originate in the urinary tract. Of these, the majority are catheter-associated urinary tract infections (CAUTIs). Yet a new study found that claims data identified a much lower rate of urinary tract infections (UTIs) as CAUTIs than what was expected from surveillance data.
As a very common, expensive, and potentially preventable hospital-acquired infection, hospital-acquired CAUTI was among the first conditions selected for the 2008 implementation of the Hospital-Acquired Conditions Initiative (HACI). This initiative denied hospitals extra payment by Medicare for hospital-acquired CAUTIs and other hospital-acquired conditions. This policy to not pay additionally for several hospital-acquired conditions rapidly expanded to many other payers beyond Medicare, including Blue Cross Blue Shield nationwide.
The study also found that nonpayment for hospital-acquired CAUTIs was low, resulting in minimal financial impact to hospitals, due to rare application of the catheter-associated diagnosis code to describe UTIs as CAUTIs. What’s more, patients also frequently had other coexisting conditions that generated the same additional payment even after removal of CAUTI as a diagnosis for payment.
The researchers examined State-wide claims data to determine how often UTIs were described as CAUTIs in the claims data, and how often payment was reduced to the hospital because of no additional payment for a hospital-acquired CAUTI diagnosis (that, prior to 2008, was eligible to generate extra payment as a coexisting condition).
The statewide study included 767,531 adult discharges from Michigan hospitals in 2007 before the HACI policy was implemented requiring nonpayment for hospital-acquired conditions and 781,343 adult discharges in 2009 after the ruling went into effect. All had been admitted to 96 hospitals in 1 State. All of the 96 hospitals frequently requested payment for UTIs (listed as a diagnosis in 10 percent of all adult discharges). However, these UTIs were rarely coded to show that these infections were actually catheter-associated (which required an additional catheter-association diagnosis code) or hospital-acquired conditions (using the mandatory "present-on-admission" variable required since 2008).
According to epidemiologic studies, the majority of hospital-acquired UTIs are catheter-associated, with rates ranging from 59 percent to 86 percent. In this study, only 2.6 percent of all hospital-acquired UTIs were coded in claims as being catheter-associated in 2009. Nonpayment for hospital-acquired CAUTI reduced hospital payment for only 0.003% of all hospitalizations.
The researchers recommend improving how hospital-acquired and device-associated events are reported in claims. They also suggest developing more rigorous data sets to measure these events accurately before using claims data to generate hospital rates of CAUTI events for public reporting (as these CAUTI events from claims data are now reported on Medicare’s Hospital Compare Web site) or penalizing hospitals with high rates (as expected in 2015 by the Affordable Care Act). Hospitals with high rates in claims data may simply document CAUTI events better in claims data.
The study was supported in part by AHRQ (HS18344). See "Effect of nonpayment for hospital-acquired, catheter-associated urinary tract infection," by Jennifer A. Meddings, M.D., M.Sc., Heidi Reichert, M.A., Mary A.M. Rogers, Ph.D., M.S., and others, in the September 4, 2012, Annals of Internal Medicine 157(5), pp. 305-312, W74-W77.