Medicare claims data identify hospital-acquired catheter-associated urinary tract infections with limited accuracy
Research Activities, August 2009, No. 348
In October 2008, under Congressional mandate, Medicare revised its payment policy with the intent of reducing reimbursement to hospitals for treating a patient's urinary tract infection (UTI) when the infection was linked to a catheter inserted during a hospital stay. However, several problems exist that undermine the intended effects of the new policy. Medicare claims may not be able to identify which patients used catheters and which patients had UTIs, note Agency for Healthcare Research and Quality researchers Chunliu Zhan, M.D., Ph.D., Anne Elixhauser, Ph.D., and William B. Baine, M.D. They used Medicare data to locate the two most common procedure codes for catheter use: "insertion of indwelling urinary catheter" and "replacement of indwelling urinary catheter." They found that less than 2 percent of catheter use was coded in Medicare claims, making these codes poor indicators of which patients received catheters during their hospital stays.
The researchers suggest that treating all patients who undergo major surgery as having a catheter is a good proxy for catheter use, and would identify nearly 50 percent of catheter use. The positive predictive value (PPV) of Medicare claims data for identifying which patients contracted UTIs from hospital-inserted catheters was just 30 percent. However, the authors suggest that addition of a present-on-admission code (to identify patients with UTIs at the time of hospital admission) in Medicare claims in October 2007 could boost the PPV up to 86 percent.
See "Identification of hospital-acquired catheter-associated urinary tract infections from Medicare claims: Sensitivity and positive predictive value," by Drs. Zhan, Elixhauser, Baine, and others in the March 2009 Medical Care 47(3), pp. 364-369.
Reprints (AHRQ Publication No. 09-R046) are available from the AHRQ Publications Clearinghouse.