20% of outpatient antibiotics prescribed without a patient visit

SAN FRANCISCO — Another large study has spotlighted the rampant overprescribing of antibiotics to outpatients.

This one showed that one in five prescriptions is written without an in-person encounter with a patient and close to half are not written for an infection-related diagnosis.

Jeffrey A. Linder, MD, MPH, chief of the division of general internal medicine and geriatrics and professor of medicine at Northwestern University Feinberg School of Medicine in Chicago, summarized the findings during a news conference at IDWeek. The results were based on an analysis of more than 500,000 antibiotic prescriptions written by more than 2,000 clinicians.

“We continue to be pretty cavalier about the use of antibiotics,” Linder said.

According to experts, antibiotic use — most common in the outpatient setting — is a primary driver of antibiotic resistance and a focus of antimicrobial stewardship efforts aimed at limiting their use and shortening the duration of treatment.

Despite stewardship efforts, however, research has showed that outpatient antibiotic prescribing has not declined in recent years. In fact, although overall antibiotic spending in the United States decreased by more than 16% between 2010 and 2015, prescribing in outpatient clinics swelled by 148%, according to researchers. Linder noted that 80% of all antibiotic prescriptions are made in the outpatient setting, and the CDC has previously said that 30% of them are unnecessary.

A recent review of national data found rampant antibiotic misuse for respiratory infections, and another study showed that nearly half of patients who received antibiotics did not need them, including almost one-third of patients without pneumonia.

There is no shortage of literature documenting the problem, but Linder said many studies and interventions involving the prescribing habits of physicians have focused on in-person visits and a narrow set of diagnoses.

“The way we have been looking at outpatient antibiotic prescribing — only focusing on in-patient visits and a restricted number of diagnoses — has been missing a lot of antibiotic prescribing,” he said.

For the current study, Linder and colleagues analyzed all outpatient prescriptions made from November 2015 to October 2017 in 514 outpatient clinics connected with the Northwestern health care system to see if the prescriptions were made during an in-person visit or in some other context, such as over the phone or through an online patient portal. They checked more than 90,000 ICD-10 codes to see if the prescriptions were justified and related to an infection. Any infections not associated with an infection-related code were considered noninfection-related prescriptions.

PAGE BREAK

Among 509,534 antibiotic prescriptions made to 279,169 patients by 2,413 clinicians, Linder and colleagues found that 20% were written without an in-person encounter — 10% over the phone, 4% during order-only encounters when prescriptions were entered into the system with no explanation, 4% via refill orders and 1% through online patient portals. Moreover, up to 46% of the prescriptions were not associated with an infection-related diagnosis. The mean age of the patients was 43 years, 60% were women and 75% were white.

Among the clinicians writing prescriptions, 63% were attending physicians, 18% were residents or fellows, 10% were nurse practitioners and 7% were physician assistants. Almost half — 41% —were medical specialists, 21% were primary care clinicians and 7% were surgical specialists.

Linder noted some of the reasons that a patient might receive an antibiotic prescription in the absence of an in-person visit.

“Two examples that come to mind are women who have frequent urinary tract infections, classic symptoms and have had documented urinary tract infections in the past. It might be perfectly reasonable for them to call in and get an antibiotic prescription. Another example might be a teen or young adult who is taking antibiotics for acne,” he said.

“We have a little bit more work to do on the data to understand if that accounts for a lot of what we’re seeing. First look is that it really doesn’t. The type of antibiotics that we’re seeing and the frequency with which antibiotics are being prescribed don’t really fit with those two examples.”

According to the analysis, the most commonly prescribed antibiotic classes were penicillins (30%), macrolides (23%), cephalosporins (14%), fluoroquinolones (11%), tetracyclines (10%) and sulfonamides (6%).

Ebbing Lautenbach, MD, MPH, MSCE, chief of the division of infectious diseases in the Perelman School of Medicine at the University of Pennsylvania, said the study shows that the problem of inappropriate antibiotic prescribing in the outpatient setting “is even bigger than we anticipated.”

“Efforts to improve how we use antibiotics in the outpatient setting are urgently needed,” Lautenbach, who was not involved in the study, said during the news conference. “One of the goals of antibiotic stewardship is figuring out how often antibiotics are used, in which situations they are used and how often are they appropriate. Generating the types of data that help to underline those types of initiatives are still a work in progress.”

Linder said one limitation of the study is that it was confined to mostly urban practices in the Northwestern health system.

“I don’t see any reason to think it’s going to be markedly different somewhere else,” he said. – by Gerard Gallagher

Disclosures: Linder and Lautenbach report no relevant disclosures.