Prescribing Antibiotics Just Got More Complicated

No one knows for sure how long a patient should take an antibiotic, new research suggests.

Conventional wisdom is that patients who fail to complete a prescribed course of antibiotics put themselves and others at risk of antibiotic resistance.

But a recent analysis fails to find evidence to support the notion and throws conventional thinking into question.

The World Health Organization endorses the full course of infection-fighting drugs, as do clinical recommendations in Australia, Canada, the United States, and Europe.<

The authors of a recent BMJ analysis point out, however, that in most cases, no one really knows how long a patient should take an antibiotic.

They write that the idea of "stopping antibiotic treatment early encourages antibiotic resistance is not supported by evidence. At the same time taking antibiotics for longer than necessary has the opposite effect, increasing the risk of resistance."

Tim E.A. Peto, a professor of infectious disease at Oxford Biomedical Research Centre in Oxford, England, is one of the study's authors. His group contends that there is a need for more evidence on the optimal course of antibiotic treatment.

A competent physician, for example, may have a good sense of an individual patient and know when it makes sense to stop treatment. Others will need guidance.

Peto's team is looking at one approach, instituting a formal review 24 to 48 hours after starting antibiotics, to assess the need to continue.

"What we haven't got is a nice algorithm that people can follow without thinking," he said. "That needs to be developed."

A hospital may have an antibiotic stewardship program, but "there may not be a good evidence base behind what they are prescribing," Peto says.

Deborah Pasko is the director of medication safety and quality for the American Society of Hospital Health-System Pharmacists. She says there is some evidence on how long to continue antibiotic therapy.

"It's just that most of the guidelines thus far have been based upon older studies," Pasko said. Those studies don't include the advanced antibiotics in use today.

Pasko says her group is optimistic about newer studies suggesting a shorter course of antibiotics might be appropriate for certain bacterial infections, such as sinusitis. If additional research shows good outcomes with shorter durations, then her group would partner with other organizations, such as the Infectious Diseases Society of America, to consider changes to the guidelines.

Evidence seems to be moving in the direction of the shorter course.

Several studies have suggested patients with community-acquired pneumonia can be treated for five days, not ten. Clinical researchers at several US academic medical centers are currently recruiting volunteers for trial that will test a short course for children with the condition.

An announcement of the trial by the National Institute for Allergy and infectious Disease notes that the approach "could also help conserve the long-term effectiveness of available drugs."

Once the evidence is in, getting providers to apply it is the next task. That involves changing long-entrenched ideas and behavior, and that's never easy.

A paper published in June reported that fewer than half of antibiotics prescribed to veterans with respiratory infections or cystitis at one VA center were done so according to guidelines.

Most cases involved the use of an antibiotic when it was not needed.

Daniel Livorsi, an infectious disease specialist at University of Iowa Hospitals & Clinics, is one of the authors on that study and several others looking at physicians antibiotic prescribing patterns.

Doctors tend to give longer courses of antibiotics longer because they are more concerned about an infection than about a patient developing adverse effects, he said. In some cases, doctors are not aware of the evidence.

Even when they were "we found that it is really hard to get doctors to change their prescribing behavior," Livorsi says. One way to do it is through audited feedback for physicians.

Livorsi thinks focusing on the length of therapy is useful. Doctors can agree on a diagnosis of an infection. A conversation about how long to treat is an opportunity to bring evidence into the discussion, he said.

Even without definitive evidence, prescribers can rely on diagnostic tools and expert opinion to help guide decisions about how long to continue care, he says.

After interviewing 30 physicians in two Indianapolis hospitals, Livorsi and his team concluded that " that antibiotic decisions are not entirely based on reason." Over use is "recognized, but generally accepted," according to a 2016 paper on the study.

Here's how one resident put it: "When it is 3:00 in the morning, depending on how busy you are, the easiest solution is to throw vancomycin and piperacillin-tazobactam at every patient because you do not have time to read the confusing guidelines that tell you 16 different things you would potentially do." 

Pasko of ASHP recommends bringing the hospital pharmacist into the decision-making process. In some cases, "we have resources such as diagnostic tests that we can use to determine whether to stop or extend therapy," she says.