Many hospitals still don’t do enough to prevent UTIs from catheters

No hospitalized patient loves getting a urinary catheter. 

But if they must have one, the last thing they want is a urinary tract infection caused by bacteria that seize their chance to grow on the catheter. 

Worse yet, some of those microbes might even use the catheter, or a catheter-related skin break in a sensitive area, as a highway into their bladder or bloodstream.

That’s why patient safety experts have worked for 25 years to reduce this risk. 

They’ve  tested and promoted strategies that encourage hospital teams to use catheters only when patients need them the most, and to take other steps to prevent and watch for catheter-associated urinary tract infections or CAUTIs. 

Now, a new study shows that while hospitals across the country have made progress over the last 20 years, they still have far to go on some key strategies.

It’s published in JAMA Network Open by a University of Michigan Medical School and VA Ann Arbor Healthcare System team that has led the urinary catheter safety charge for two decades.

In all, only 49% of general hospitals and 32% of hospitals run by the United States Department of Veterans Affairs have a system in place to remind doctors that their patient has a catheter, or require doctors to proactively renew a patient’s catheter order after a certain amount of time. 

Higher percentages of hospitals (58% and 38%, respectively) let nurses remove catheters without a doctor’s order if they feel a patient can try urinating on their own.

Those three practices have the strongest evidence behind them for their power to reduce catheter-associated urinary tract infections. 

And while the study shows that a higher percentage of hospitals use one or more of the three options now than 15 or 20 years ago, going from about 23% to almost 70% in non-VA hospitals, the percentage has plateaued or even dropped slightly in recent years. 

The study is based on 20 years’ worth of surveys of hospital infection prevention staff carried out every four years since 2005.

It shows growth over time in the percentage of hospitals that monitor which patients have catheters, and how many days they’ve had them.

The percentage of VA hospitals with these systems was lower than the percentage of non-VA hospitals, but a VA-wide electronic health record system now rolling out may close the gap, the researchers say. 

The study was led by Sanjay Saint, M.D., M.P.H., Professor Emeritus of Internal Medicine at U-M, now at Northwell Health and the Donald and Barbara Zucker School of Medicine at Hofstra/Northwell, and by Jennifer Meddings, M.D., M.Sc., a Professor of Internal Medicine and Pediatrics at the U-M Medical School, primary care physician at U-M Health, and Research Investigator at the VAAAHS.

“Looking back over 20 years, it is encouraging to see how far hospitals have come in recognizing that urinary catheters are not benign devices and in building systems to use them more thoughtfully,” said Saint.

“At the same time, our findings are a reminder that progress is not self-sustaining. The most effective strategies – such as prompting clinicians to reassess whether a catheter is still needed and empowering nurses to remove it when appropriate – should be embedded in everyday care.”

Meddings said, “It is quite surprising that reminders to prompt clinicians to reassess catheters’ necessity have not improved significantly in recent years, given these strategies are low-cost, effective, and many health systems have already embedded these types of reminders into electronic medical record workflows.”

In much the same way, she said, hospitals can use their electronic medical record systems to restrict the use of catheter placement for specific, appropriate clinical reasons. “This can include reminding and prompting clinicians, within the electronic order for inserting catheters, to only place indwelling catheters for a list of clinically appropriate reasons, and easily reminding them of potential alternatives such as external catheter placement,” she said. “These types of orders serve as just-in-time education or re-education for clinicians, which is particularly important for trainee physicians who place the majority of catheter orders in many hospitals.” 

Free resources to help hospitals use catheters appropriately

The U-M/VA team has launched a website, bladdersafe.org, to share free resources to help hospitals anywhere understand what they can do to prevent CAUTIs in medical and surgical patients. 

They also offer a mobile app, BladderSafe, to help clinicians decide when to use or avoid common devices to manage adult patients’ urinary needs. 

The site even includes a calculator to help hospitals understand how the cost of implementing those changes can ultimately save money, by reducing cases of catheter-associated bacteriuria, symptomatic urinary tract infection and bloodstream infection. 

Medicare penalizes non-VA hospitals financially if their patients covered by traditional Medicare have high CAUTI rates, and does not reimburse hospitals for the cost of caring for such patients’ catheter-associated infections. 

The public can even see and compare hospitals’ rates of CAUTIs and other patient safety issues among patients with traditional Medicare coverage. 

In addition to reminders, automatic stop orders and nurse-initiated catheter removal, another key strategy for reducing CAUTI is using a portable ultrasound device to scan the bladder and guide decisions about catheter use.

On this measure, more than 70% of VA and non-VA hospitals are using the scanners.

Backed by decades of work

The researchers behind the new study, and their colleagues, created many of those resources as part of years of research and quality improvement projects related to urinary catheters and infection prevention. 

In fact, their work showed it was possible to drive down CAUTI rates at hospitals and nursing homes across Michigan and nationwide. 

“Through numerous local, regional, and national implementation projects, we have assessed how collaborative efforts positively impact both process measures and traditional outcome measures like infection rates,” said epidemiologist M. Todd Greene, Ph.D., an Associate Research Scientist in the U-M Division of Hospital Medicine and Research Investigator at the VAAAHS.

“Moving forward, in addition to continuing to monitor appropriate catheter utilization and CAUTI rates, expanding our tracking systems to better monitor, recognize, and prevent non-infectious complications will help to ensure we capture the full spectrum of device-related harms.” 

Because hospital infection prevention staff are often at risk of high job stress and turnover, better support for them could help ensure continuity in data collection, and in patient safety and infection prevention efforts.   

The authors are members of the U-M/VA Patient Safety Enhancement Program, the U-M Institute for Healthcare Policy and Innovation, and the VA Center for Clinical Management Research. In addition to Saint, Meddings and Greene, they include Sarah Krein, Ph.D., R.N., Kristin Chrouser, M.D., M.P.H., Karen Fowler, M.P.H. and David Ratz., M.S.

Saint recently won a national patient safety award, called the  John M. Eisenberg Patient Safety and Quality Award in the Individual Achievement category, from Joint Commission and the National Quality Forum in recognition of his work on CAUTI prevention. The award is named for a former administrator of AHRQ.

The survey project was funded by the Department of Veterans Affairs, and used survey resources through the Michigan Institute for Clinical Health Research, which is supported by a National Institutes of Health National Center for Advancing Translational Sciences grant (UM1TR004404).

U-M and VA research and quality improvement projects on CAUTI prevention have been funded by the Agency for Healthcare Research and Quality, the NIH and Blue Cross Blue Shield of Michigan through the Collaborative Quality Initiatives based at U-M.

Citation: "Catheter-Associated Urinary Tract Infection Prevention Practices in US Hospitals," JAMA Network Open. DOI: 10.1001/jamanetworkopen.2026.37540