Published in the February 2017 issue of Today’s Hospitalist
WHEN HOSPITALISTS at the University of Michigan decided to embark on an antibiotic stewardship program a few years ago, they scoured a week of data from their own hospital charts. The goal was to get an idea of how well physicians were documenting their antibiotic orders. (See what our readers had to say about hospitalist antimicrobial stewardship.)
The hospitalists were shocked to find that only 4% of the charts contained the three basic elements experts say can help clinicians manage such orders appropriately: the antibiotic’s indication, how long the therapy should last, and start and stop dates. These documentation steps may seem small, but experts say they help with handoffs, discharge planning and appropriate use. They also make shortcomings visible, such as doctors not planning when to end patients’ therapy.
“We were doing terribly,” admits Megan R. Mack, MD, a University of Michigan hospitalist and lead author of an August 2016 Journal of Hospital Medicine (JHM) article describing a multi-site effort to engage hospitalists in antimicrobial stewardship. “Physicians couldn’t argue with the data, and the results hammered home the concept that antibiotic stewardship is the responsibility of us all.”
“We need to create a habit of re-evaluation.”
The hospitalists at the University of Michigan are far from alone. The CDC, for example, estimates that between 20% and 50% of all antibiotics prescribed in U.S. hospitals may be inappropriate.
And a study published in the Sept. 19, 2016, issue of JAMA Internal Medicine found that “overall days of therapy of all antibiotics among hospitalized patients in U.S. hospitals did not change significantly” between 2006 and 2012. Researchers also found that the use of broad-spectrum agents in hospitals had grown significantly over those same years.
But hospitals haven’t been very quick to respond. Only 39% of hospitals report having an antimicrobial stewardship program that meets all of the CDC’s criteria. Those include having a program backed by hospital leadership that can boast of pharmacy expertise, educational components, the ability to track and report prescribing and resistance patterns, and engagement in at least one prescribing improvement action.
But pressure from regulatory agencies will likely boost the number of programs. California already requires hospitals to launch stewardship programs, while the Joint Commission this year has begun assessing such programs as part of its accreditation survey. Further, the Centers for Medicare and Medicaid Services (CMS) last summer proposed new “conditions of participation” for hospitals and critical access facilities, which include having a stewardship program in place. Such a condition may be implemented as early as 2018.
While policy prescriptions focus on hospitals, ID experts say that stewardship initiatives need to target antibiotic prescribers. That’s where hospitalists come in.
“Improving antibiotic use has to come from the people who actually prescribe the antibiotics,” says Arjun Srinivasan, MD, associate director for health care associated infection prevention programs at the CDC in Atlanta. Because smaller and even mid-sized hospitals don’t have regular access to ID subspecialists or ID-trained pharmacists, Dr. Srinivasan adds, hospitalists will need to step up and lead.
For hospitalists to lead stewardship efforts, however, they’ll need to overcome significant barriers. Those include a lack of time, interest, urgency, and the feeling that while better use of antibiotics is important generally, it is not necessarily hospitalists’ responsibility.
“I know how many DVT and stroke fall-outs I have, but I don’t know how many UTI patients I have on unnecessary antibiotics.”
~ Michael Goonewarde, MD
Tamar Foster Barlam, MD, director of antimicrobial stewardship for Boston Medical Center and lead author of new guidelines on how to implement stewardship programs, says there’s also a persistent misconception among physicians that more and broader antibiotics can be better than less and narrower. For physicians, Dr. Barlam notes, that thinking particularly applies to the patient sitting in front of them. “They don’t want to miss something in the moment.”
And some clinicians take a negative view of antibiotic stewardship, seeing such initiatives as nothing more than a way for administrators to cut costs. While that may have been true in some hospitals in the past, that’s no longer the case. “The emphasis now has moved away from cost and toward reducing C. diff and antibiotic resistance,” says Dr. Barlam. “The issues are medical, not fiscal.”
The good news is that obstacles can be overcome by seemingly small steps, like asking prescribers to state their reason for ordering antibiotics. Data show that such efforts can go a long way to getting hospitalists engaged.
Dr. Mack and her colleagues at the University of Michigan described three effective “best practices” in their JHM study. Those included enhancing antimicrobial documentation, improving local clinical guidelines and making them accessible, and instituting an antimicrobial timeout after 72 hours.
The study concluded that those practices resulted in “a gradual shift in thinking” among hospitalists, “from initial skepticism about embedding stewardship within their daily workflow to general acceptance that it was a worthwhile and meaningful endeavor.”
A project underway at Baltimore’s Johns Hopkins Bayview Medical Center takes a different approach: It is designed to engage hospitalists in antibiotic stewardship by concentrating on improving adherence to local guidance for one common condition that accounts for a lot of inappropriate antibiotic use: urinary tract infections (UTIs).
The project—led by Jennifer L. Townsend, MD, an ID subspecialist who is part of the hospitalist service, and Robin McKenzie, MD, an ID stewardship leader with an interest in UTIs—is educating hospitalists about new evidence-based ways to diagnose UTIs. That should lead to less antibiotic use.
Dr. Townsend has also been pulling routine urinalysis testing out of most order sets because, she notes, when physicians see bacteria in urine, they want to treat it. Moreover, she is recruiting “peer promoters” among hospitalists and housestaff who will be charged with reviewing some charts of patients taking antibiotics for UTIs. They will then “call their friends, and say, ‘This is now Day 3 of your patient being on an antibiotic. Do you really think this is an UTI?’ ”
At Reading Hospital, Reading Health System’s 600-plus-bed hospital in West Reading, Pa., hospitalist Michael Goonewardene, MD, sits on the central antibiotic stewardship committee and advocates for antibiotic practice changes among the 55-hospitalist group. He has had more success convincing hospitalists to improve their documentation of antibiotic orders, but less getting them to participate in antibiotic timeouts.
One lesson he has learned: Any proposed changes “can’t interrupt hospitalists’ workday and have to be beneficial to them,” or they won’t get on board.
With documentation, for instance, hospitalists appreciate the benefits of including an indication and information about therapy duration. Having that information readily available “saves them time and hassle digging back through the EMR, trying to add up the days,” Dr. Goonewardene explains, “particularly when they want to discharge the patient.”
To make life easier for hospitalists, he recruited the hospital’s clinical documentation specialists—who were already combing through charts, making sure the hospitalists properly code and bill—to start noting whether antibiotic orders include indication, duration and proposed number of treatment days. They then query hospitalists when information is missing.
“Part of what the documentation specialists did was to remind us about key parts of antibiotic documentation,” says Dr. Goonewardene. “It is important for patient care that we assess appropriate antibiotic use, which is now mandated for 2017.”
On the other hand, he still struggles to get colleagues comfortable with antibiotic timeouts. (The CDC and other stewardship experts recommend timeouts to prompt providers to stop and rethink orders 48 or 72 hours into a hospitalization.)
“We are still trying to incorporate this in our workflow,” Dr. Goonewardene says. “We thought we could do it through the multidisciplinary rounds, but we have had mixed results.” There are so many things to talk about during those rounds, he adds, “anything that makes rounds longer is very much frowned upon.”
Changing the “Wild West for antibiotics”
When hospitalist Nisha Viswanathan, MD, talks about antimicrobial stewardship, she describes her practice’s culture as “the Wild West for antibiotics.” That’s because the hospital has allowed clinicians to order whatever antibiotics they wanted without any restrictions or need to justify their use.
Dr. Viswanathan, who chairs the antimicrobial stewardship committee at the 200+bed White River Medical Center, a community hospital in Batesville, Ark., has started with baby steps. Her first project was to better understand prescribing habits. She tackled that by requiring doctors to choose an indication from a dropdown menu for every antibiotic order.
That step actually helped improve patient care. “We’ve had multiple situations where we discovered that the wrong antibiotic was being used to cover the bacteria cultured,” she points out.
More recently, her program began having hospitalists implement antibiotic timeouts for ICU patients. During each timeout, doctors consider pharmacist recommendations to deescalate, set an end date, escalate, or switch drugs to something more appropriate, based on culture results and clinical data.
“Some physicians wanted to continue the medications they were using,” says Dr. Viswanathan. “But we have noticed that by day 3 or 4, they have switched patients to more appropriately narrow regimens, where previously they may not have. This is making our physicians think about antibiotics, and then they start making such decisions on their own.” She’s also noticed other improvements, including fewer days of therapy ordered.
The program plans to implement antibiotic timeouts hospital-wide. But at the community hospital, time and resource constraints are big barriers to monitoring everyone’s antibiotic choices. As a result, the program has decided to limit that monitoring to the seven broad-spectrum antibiotics most frequently used in the hospital.
Measure and report
Hospitalists who don’t embrace antibiotic stewardship on their own may find another motivation in the form of public reporting. In Pennsylvania, Dr. Goonewardene thinks hospitalists will truly come on board only when health systems collect—and report—data on antibiotic prescribing.
“Nobody wants to be the under-performer,” he explains. “As a hospitalist, I get feedback on everything. I know how many DVT and stroke fall-outs I have, but I don’t know how many UTI patients I have on unnecessary antibiotics.”
Such measurement—and reporting—may be coming sooner than many hospitalists think. Anurag Malani, MD, is medical director for the antimicrobial stewardship and infection prevention programs for three community teaching hospitals within the Saint Joseph Mercy Health System in southeastern Michigan. He cites a measure developed by the CDC known as the SAAR (standard-zed antimicrobial administration ratio), which will enable hospitals to compare their antibiotic utilization to that of other facilities. Last summer, the CMS proposed using such a measure as part of its Hospital Inpatient Quality Reporting program.
An infectious disease subspecialist, Dr. Malani notes that “it’s going to be difficult for smaller hospitals, especially if they are not part of a larger health system, to have a way to measure antibiotic use.” But, he adds, “you can’t improve unless you can measure.”
Engaging hospitalists may also require a bigger change in the culture of medicine. Johns Hopkins Bayview’s Dr. Townsend points out that challenging a colleague’s decision or changing a treatment plan can be perceived as being “disloyal.”
“If your colleague evaluated a patient and says ‘UTI,’ you don’t question that,” she says. “We need to create a habit of re-evaluation.” Changing an antibiotic order should not be seen as criticism of a colleague, she explains. “We should instead say that a patient’s course has evolved, and ‘I know more now than you did, so I’m going to stop the antibiotics.’ ”
Creating such a culture, however, still runs into the very real barriers of time and money. While Dr. Viswanathan in Arkansas says that stewardship efforts enjoy moral and political support at her hospital, it is very difficult to balance that responsibility with her clinical duties.
“I really enjoy making a difference in our quality of care, but I have to fit stewardship efforts in around my role as clinical provider,” she points out. The Arkansas state health department recommends that stewardship chairs should receive protected time or extra compensation. But having that additional time may cost her hospitalist program.
“If I see fewer patients,” Dr. Viswanathan says, “my colleagues have to see more. I am willing to do what I can to champion stewardship. But I don’t want to take on projects that I can’t commit to, and at the moment I see a full patient load every day.”
Deborah Gesensway is a freelance writer who covers U.S. health care from Toronto.
The more you do, the bigger the impact
WHEN IT COMES to antimicrobial stewardship programs, most evidence shows that more is better: Multifaceted efforts reduce antibiotic consumption better than limited ones, even in small hospitals.
In an abstract presented at the IDWeek 2016 meeting last fall, Edward Stenehjem, MD, an infectious disease specialist who is medical director of antimicrobial stewardship at Intermountain Healthcare, described how some of the 15 small community hospitals in his health system, each with less than 150 beds, reduced their antibiotic usage. Over 15 months, those hospitals cut that usage by between 20% and 30% with a robust stewardship program modeled on ones in large academic centers.
The program included prospective audit and feedback, ordering restrictions enforced by trained physicians and pharmacists, ID physicians reviewing clinically important blood cultures, and a telephone hotline that gave remote facilities 24/7 access to ID specialists. By comparison, Dr. Stenehjem says, a group of other small hospitals in the system, which were given only a bare-bones stewardship program with basic education and access to the ID hotline, accomplished much less.
“We showed that the more you did, the bigger the impact on antibiotic usage,” Dr. Stenehjem explains. As a result of that trial, Intermountain has now developed an ID telehealth program that primarily serves those 15 small hospitals. Intermountain has also hired two physicians, an ID pharmacist, a director and an analyst to operationalize that program. The team works with onsite physician and pharmacy champions to provide stewardship interventions and support.
“We are able to do this because we are part of a centralized health system, and Intermountain Healthcare has provided us with the resources,” he notes. He adds that he believes small hospitals that are not part of a bigger system will find antibiotic stewardship difficult, due to a lack of ID providers and administrative support. (See “Stewardship challenges for smaller hospitals,” below.)
“There are a lot of challenges when you branch out into smaller hospitals,” says Tamar Foster Barlam, MD, director of antimicrobial stewardship for Boston Medical Center and lead author of the April 2016 guidelines from IDSA and SHEA (published in Clinical Infectious Diseases) on how to implement such programs. Hurdles include limited (or no) access to infectious disease physicians or ID pharmacy specialists, bare-bones microbiology labs without rapid testing or the expertise to analyze local antibiograms, inadequate computer systems, scant data collection ability, and few resources to compensate clinicians—with time or money—to do the work.
“Giving lectures to your hospitalists isn’t an effective stewardship program and shouldn’t be interpreted as such.”
But unlike guidelines published nearly 10 years ago, the new recommendations recognize these limitations, Dr. Barlam says. They do not specify, for instance, that antibiotic stewardship programs must be led by ID doctors and clinical pharmacists with ID training who should be compensated for their time.
Most experts think that’s probably the best model, but hospitals “are going to put the team together based on their own resources and who they can get hold of.”
Plus, it’s more important to create a program with mechanisms shown to effectively reduce inappropriate antibiotic use. Key interventions include:
- preauthorization and/or prospective audit and feedback for targeted antibiotics, particularly those used to treat drug-resistant infections;
- syndrome-specific initiatives, such as focusing only on UTI or pneumonia care;
- implementing antibiotic timeouts or other strategies to encourage doctors to reconsider planned therapy; and
- rapid diagnostic testing to determine quickly if infections are viral rather than bacterial.
The guidelines also recommend against “relying solely” on educating providers. “Giving lectures to your hospitalists isn’t an effective stewardship program and shouldn’t be interpreted as such,” Dr. Barlam says. While the guideline authors weren’t able to spell out what hospitals of specific sizes should be able to implement, “we did make the point that this is all dependent on resources.”
Dr. Barlam also offers small hospitals the same advice she gives larger ones: “Don’t try to improve prescribing for every type of infection at the same time,” she points out. “Hospitals that don’t have resources should focus on the interventions they think will have the biggest bang for their buck. Choose what you think you have the resources to do, and start there.”