AHRQ toolkit helps hospitals improve antibiotic selection to reduce deadly C. difficile infections

Feature Story

Belinda Ostrowsky, M.D., (rt) and Priya Nori, M.D., of the Montefiore Medical Center/Albert Einstein College of Medicine, review an antibiogram (chart summary of percentage of bacteria susceptible or resistant to a range of antibiotics used in the hospital), which is used to help select optimal antibiotic therapy.

Your very sick hospital patient has just finished 7 days of the antibiotic ciprofloxacin (cipro) for a urinary tract infection (UTI). The lab has found more bacteria in his urine. But he no longer has a fever or back pain indicating a UTI. The followup urine sample may simply have been contaminated. On the other hand, maybe he needs another course of antibiotics. But that may put him at risk of a potentially deadly Clostridium difficile infection. That was a recent dilemma faced by George McKinley, M.D., Infection Disease Specialist at St. Luke's Roosevelt hospital in New York City, and a typical one faced by today's doctors as they battle sometimes deadly healthcare-associated infections like C. difficile.

"The patient is debilitated, with multiple problems," recounts McKinley. "I am very concerned that if he gets another course of antibiotics he will get a C. difficile infection. By trying to avoid that problem, could I be delaying the antibiotic treatment in a way that will put him at risk for another type of infection? These are the types of decisions we have to make all the time, calculating the risk and benefit of antibiotics."

C. difficile infection is a potentially deadly infection that has been linked to certain antibiotics that are typically the broad-spectrum antibiotics most often used at a hospital, which target a broad range of bacterial infections. People most at risk of getting C. difficile infections are those in health care facilities where C. difficile bacteria can be transmitted to them from health care workers' hands, other patients, or surfaces such as toilet seats, bed rails, and food trays. Also at risk are patients taking antibiotics that destroy good intestinal bacteria that can defeat dangerous bacteria like C. difficile and sicker and older patients. Generally, the longer patients are on antibiotics, the greater their risk of getting C. difficile.

Infection with C. difficile, which produces toxins in the colon, can range from mild diarrhea to damage to the bowel severe enough to lead to colon removal, sepsis, or even death. The number of patients hospitalized with C. difficile doubled between 2000 and 2007, with rates highest in the northeastern United States, according to data from the Agency for Healthcare Research and Quality (AHRQ) Healthcare Cost and Utilization Project (HCUP). Rates seem to have leveled off from 2008 to 2010, but remain disturbingly high.

AHRQ's new online "Toolkit for Reduction of Clostridium difficile Infections Through Antimicrobial Stewardships" helps hospitals improve the use of antibiotics by implementing an Antimicrobial Stewardship Program (ASP) directly targeting C. difficile infections. It builds on a general antimicrobial stewardship toolkit developed by the Greater New York Hospital Association/United Hospital Fund. The toolkit helps individual hospitals identify antibiotics most linked to C. difficile infections at their hospital so they know which ones to target. The toolkit also guides hospitals in development of strategies to improve appropriate use of antibiotics. Sometimes this may involve consultation with the hospital infectious disease specialist before prescribing a targeted antibiotic, review of patients on broad-spectrum antibiotics like cipro that combat several types of bacteria to see if they can be put on more narrow-spectrum antibiotics (audit and feedback), and computerized alerts asking clinicians about the need to continue antibiotics for a patient. Finally, the toolkit helps hospitals consider organizational changes and resources needed to create and sustain an effective ASP; provides instructions on how to plan, implement, and adjust an ASP; and describes lessons learned from the 10 hospitals that participated in the C. difficile antibiotic stewardship project.

All the project hospitals that put in place elements of the antibiotic stewardship program directed at C. difficile decreased use of at least one targeted antibiotic associated with C. difficile, notes Belinda Ostrowsky, M.D., M.P.H., Director of the Antimicrobial Stewardship Program at the Montefiore Medical Center/Albert Einstein College of Medicine and clinical principal investigator for the AHRQ toolkit project. "Each of the hospitals has already tried techniques to clean the environment and improve infection control, such as hand washing and signage," says Ostrowsky. "The toolkit adds a layer of being more sensible in antibiotic prescribing."

"There are antibiotics that are workhorses at many hospitals," explains Ostrowsky. "For example, the quinolones, like cipro and moxifloxacin, are often used as antibiotic workhorses because they can be given orally. And in the past, they have been associated with the hyperendemic strain of C. diff., and we did see within the project hospitals that they were associated with many cases of C. diff."

But restricting hospital antibiotic use is not easy, notes McKinley. Some doctors find it difficult to restrain themselves from prescribing antibiotics. "Antibiotics are life-saving interventions and many patients need antibiotics, and yet those patients get C. diff.," says McKinley. "But to have a patient get C. diff. from an antibiotic that they didn't need or could have received an alternative antibiotic that might have been less prone to give them C diff., that is something we continually educate doctors about."

Power of toolkit strategies and infection control

McKinley's hospital has been targeting the use of certain antibiotics for years in an effort to reduce healthcare associated infections. However, AHRQ's toolkit enabled the hospital to target specific antibiotics linked to C. difficile infections at the hospital (cipro, a quinolone, and cefepime, a cephalosporin). Using the C. difficile toolkit, immediate patient isolation with onset of diarrhea without waiting for laboratory confirmation of C. difficile, contact precautions (use of gowns and gloves and strict hand washing), and stringent surface disinfection policies, his hospital was able to reduce hospital-onset C. difficile infections by 50 percent in August 2012 compared with 2009.

The Bronx-Lebanon Hospital Center, another project hospital, did a case-control study that found that cipro (and the quinolone antibiotics in general) and Zosyn® (a broad-spectrum antibiotic from the pencillin family) were the two most often used antibiotics at the hospital and most linked to C. difficile infections. Frank Palmieri, Ph.D., R.Ph., Antibiotic Clinical Pharmacist at the hospital, Frances Petersen, R.Ph., M.P.H., Director of Infection Control, and their team started an intravenous (IV) to oral program. The key question to clinicians was, "Do you really need to continue the antibiotic?"

With the hospital programmer, they built a program to work with the hospital system's computerized physician order entry program. For example, after physicians ordered an initial 3-day supply of IV cipro and went on the system to reorder the antibiotic, a popup message would ask them to consider if they really needed to continue with IV cipro or might be able to switch to oral cipro according to the guidelines. Many times the physician did switch to the oral medication or stop the antibiotic altogether after the 3 days.

HCUP Data on C. difficile Hospitalizations

  • In 2009, there were 336,600 hospitalizations that involved C. difficile infection (CDI)—nearly 1 percent of all hospital stays.
  • In 2009, patients 85 years and older had the highest rate of CDI hospitalizations—1,089 versus only 11 per 100,000 population for patients younger than 18.
  • CDI hospitalizations doubled between 2001 and 2005, then leveled off during 2008 to 2010.
  • Among patients diagnosed in 2009 with CDI during hospitalization, 4.8 percent were readmitted to the hospital within 30 days and 6.9 percent within 90 days principally for CDI. Another 12.8 percent were readmitted within 30 days and 17.2 percent within 90 days for any listing of CDI.
  • Nearly half (49 percent) of long-term care patients hospitalized with CDI were readmitted to the hospital within 90 days for any cause.

For more information, on the HCUP report on CDIs in hospital stays in 2009 go to http://www.hcup-us.ahrq.gov/reports/statbriefs/sb124.pdf. You can view the HCUP report on rehospitalizations for Clostridium difficile at http://www.hcup-us.ahrq.gov/reports/statbriefs/sb145.pdf.

C. difficile cases drop by half

"In addition to asking if the continuation of ciprofloxacin was necessary, we also looked at our treatment guidelines that included this antibiotic," adds Palmieri. "This led to a major change in recommending quinolones [including cipro] upfront for urinary tract infections and even the pneumonias. The result has been that we are now using half the amount of ciprofloxacin and have seen our number of C. diff cases drop by the same amount."

The hospital received a letter from the New York State Department of Health recently asking them how they were able to significantly lower their C. difficile rates from 2007 to 2011. Petersen attributed this success to a bundle of strategies. They included special protocols for cleaning patient isolation rooms, in-service training of staff on proper room cleaning, placing patients with positive C. difficile labs on immediate contact precautions with a special sign outside the patient's door, and their ASP.

The hospital had a general ASP that developed and matured, but working on the toolkit project helped them focus. Says Petersen, "When you start a program, you want to tackle 8 million issues at one time. Focusing on one thing and seeing that it works, leads you to the next step. You have to take one step at a time."

Petersen recently had a relative at the hospital who had C. difficile infection. "He was being treated [with antibiotics] for pneumonia," she recalls. "Honestly I thought he was going to die. I've never seen anything like that in my life. The fact that they discontinued antibiotics, changed them, and medically managed him made a difference. He was delirious. I didn't think he was going to make it, and he's not that old. They saved him." Palmieri's father was at another hospital with C. difficile infection in the past year. It was close and took his father 6 weeks to be clear of the infection. A small group of patients with C. difficile infections are repeatedly readmitted to the hospital—even over a period of years—for relapse of the infection.

Toolkit helps hospitals modify antibiotic selection

AHRQ's toolkit helps hospitals modify antibiotic selection to prevent such devastating infections. Palmieri sees great promise in the toolkit, especially for hospitals that are starting ASPs. "They don't know what direction to go or how to start, and the toolkit gives great guides to get them going," he says. While team membership will vary among hospitals, the traditional core ASP team includes an infectious disease physician and a pharmacist (Pharm.D.), and perhaps a clinical microbiologist, infection preventionist, and hospital epidemiologist. An information technology representative and senior administrator often act as liaisons to support and supplement the core ASP team members. Ostrowsky and her team have a close relationship with the infection control people and nurses on the floor taking care of the patients to make sure they are collecting the C. difficile samples and asking prescribers if they still need to have the patient on an antibiotic or whether the course can be shortened.

"Part of teamwork is having the prescribers believe there's a problem," says Ostrowsky. "We work with the chief residents, head of the hospital service, and people who coordinate the emergency doctors so we can get the educational information out there. It's changing the culture."

Ostrowsky points out that some bacteria are already developing resistance to broad-spectrum antibiotics like cipro due to their overuse. The contribution of commonly used antibiotics to C. difficile infections is also a problem. "If there were a lot of new antibiotics, we wouldn't be having this conversation," she asserts. "But the problem is that we don't have that many new antibiotics coming down the road, so we need to be smart about how we use the ones that we have."

Ostrowsky believes the toolkit project has raised awareness of the problem of certain antibiotics and C. difficile infections and the importance of ASPs. "Our antimicrobial stewardship programs look at antibiotic use in the whole facility, and a lot of times physicians will look at one patient at a time and don't see the whole picture. What's nice about the antimicrobial stewardship programs is that people who develop them are often the people who have a more global view."

The toolkit's description of lessons learned and strategies developed by the project hospitals can help hospitals think about what may work for their circumstances. Says Ostrowsky, "The toolkit gives examples of ways you might start to tackle these problems so you don't have to reinvent the wheel." The toolkit is available at ERASE C. difficile .

GSM

Page last reviewed February 2013
Internet Citation: AHRQ toolkit helps hospitals improve antibiotic selection to reduce deadly C. difficile infections: Feature Story. February 2013. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/news/newsletters/research-activities/13feb/0213RA1.html