New federal penalties for catheter-associated urinary tract infections that limit reimbursement to hospitals where they occur may be ineffective because hospital staffs often don't accurately code those cases.
That's the finding in a recent study by University of Michigan Health System's Jennifer Meddings, MD, and others who wrote that hospital staff accurately used the correct ICD-9 code for CA-UTI, 996.64, only 1% of the time in a sample of patients hospitalized in 2007.
Accurately recording such hospital-acquired infections requires a complicated three-step process that is seldom completed, the authors said. That would result in the hospitals receiving full federal payment, which defeats the purpose of the federal penalty to improve care, they wrote in a recent edition of the journal Infection Control and Hospital Epidemiology.
"If no improvements are made in applying diagnosis codes to accurately identify hospital-acquired CA-UTIs, the nonpayment policy will have little if any effect on hospital reimbursement for this complication, because hospitals will continue to be eligible for additional payment for UTIs," Meddings and colleagues wrote.
"Furthermore, hospitals that face little financial incentive (or penalty) may be unlikely to invest in the processes necessary to improve care."
Additionally, the researchers discovered that "hospital coders often code UTIs as present on admission when the medical record indicated that the UTIs were hospital- acquired."
The researchers looked at medical records at the University of Michigan Health System in 2007, a period that ended nine months before the Centers for Medicare and Medicaid Services' (CMS) Hospital Acquired Conditions Initiative's list of eight such conditions not eligible for payment, as of Oct. 1, 2008.
The researchers sampled medical records and found 80 patients who had evidence of a catheter-linked urinary tract infection during their hospitalization. Of those, 37, or 46%, were hospital-acquired, according to their review, although none were identified as "hospital-acquired" or "nosocomial."
"Yet no cases in our sample, and only 1.2% of cases of secondary-diagnosis UTI at UMHS during 2007, were identified as Catheter-Associated Urinary Tract Infections by hospital coders in the administrative discharge data set.
"Similarly, our exploratory analyses of Michigan, California, and nationwide discharge data also indicate that the catheter association code is applied to only approximately 1% of all secondary diagnosis UTIs submitted for payment."
Nancy Foster, vice president for quality and patient safety policy for the American Hospital Association, said she agrees that accurate coding is a complex process. But she says that even if hospitals fail to correctly code, it would not have much impact on payment.
That's because the bulk of catheter-associated urinary tract infections would not result in reduced payment to hospitals because the patients that got these infections already have so many complications that they are already in highest payment categories. "It isn't having a very significant impact on payment," she says.
Interviewed by phone, Meddings agreed that it may not have much of an impact because the sample of patients was very very sick, for example, many of them were on the list or had already had an organ transplant.
But she said that urinary tract infections are common, as is the use of catheters in hospitals. And often, the use of a catheter is not reflected in the medical record.
"Catheters are considered part of the general supply, like bandages. When they're used they're not scanned for a bar code, and there's no tracking system to link patients with urinary tract infections to catheters," she said.
When auditors and hospital coders look through discharge records to decide on what codes to use for billing, often the catheter link is not found, she said.
Meddings and others wrote in the article that they did not think CMS had explored or anticipated this phenomenon during the development of the policy.
"Because coding of hospital-acquired CA-UTI seems to be fraught with error, nonpayment as required by CMS policy for this complication is unlikely to occur reliably," the authors wrote.
"Accordingly, the effective implementation of this policy will require an enhanced auditing process by the CMS, which could vitiate any financial gains anticipated from nonpayment for this complication."