Focusing ambulatory antibiotic stewardship interventions on in-person visits only may miss a substantial proportion of inappropriate prescribing, according to a study published yesterday in PLOS One.
The study authors say the findings, which are based on ambulatory antibiotic prescribing data collected prior to the COVID-19 pandemic, could take on greater significance with more outpatient healthcare visits now being conducted virtually.
Smaller proportion of appropriate prescriptions
For the study, a team led by researchers with Northwestern University Feinberg School of Medicine used electronic health record data to analyze all ambulatory, oral antibiotic prescriptions ordered at Northwestern Medicine from January 2016 through 2019. They classified encounters as either in-person or not-in-person, with telephone, e-patient portal, orders-only, and refill encounters considered not-in-person.
Previous research has shown that at least one-third of all ambulatory or outpatient antibiotic prescriptions are inappropriate, which has led to nationwide efforts to promote more judicious antibiotic prescribing in these settings. But the authors note that not-in-person encounters, which may be hampered by the inability to conduct a physical examination, are often overlooked in outpatient antibiotic stewardship initiatives. So they set out to assess the prevalence and appropriateness of not-in-person antibiotic prescriptions to get a more complete picture of ambulatory prescribing.
The researchers used to diagnostic codes to categorize the prescriptions into 5 mutually exclusive appropriateness groups: chronic antibiotic use (for conditions like chronic osteomyelitis and cystic fibrosis), antibiotic-appropriate, potentially antibiotic-appropriate, non-antibiotic-appropriate, and antibiotics not associated with a diagnosis.
These findings suggest that ambulatory stewardship interventions that focus only on in-person visits may miss a large proportion of antibiotic prescribing, inappropriate prescribing, and antibiotics prescribed in the absence of any diagnosis.
Over the 4-year study period, there were 714,057 antibiotic prescriptions ordered for 348,739 patients at 467 clinics. Patients were 61% female, with a mean age of 41 years. Overall, 81% of antibiotic prescriptions were ordered during in-person visits and 19% during not-in-person visits. The most common types of not-in-person visits were telephone (10% of all not-in-person prescriptions), orders-only (5%), and refills (3%).
Across all encounter types, chronic antibiotic use accounted for 16% of antibiotic prescriptions, antibiotic-appropriate and potentially antibiotic-appropriate for 54%, and non-antibiotic-appropriate for 22%. Eight percent of prescriptions were associated with no diagnosis. The authors note that if the absence of a diagnosis were to be considered non-antibiotic-appropriate prescribing, that would bring the total proportion of inappropriate prescribing observed—30%—in line with national estimates.
Antibiotics prescribed during not-in-person visits were more likely to be chronic (20% vs 15%) compared with in-person visits, but less likely to be associated with appropriate or potentially appropriate diagnoses (30% vs 59%) and more likely to be associated with no diagnosis code (42% vs <1%). An estimated 58% of not-in-person encounters had a diagnosis code, compared with 99% of in-person encounters.
"These findings suggest that ambulatory stewardship interventions that focus only on in-person visits may miss a large proportion of antibiotic prescribing, inappropriate prescribing, and antibiotics prescribed in the absence of any diagnosis," the study authors wrote.
The authors add that the findings are even more important given COVID-19 pandemic-related shifts in outpatient healthcare visits. Within the Northwestern Medicine system and other healthcare systems around the country, in-person visits declined during the pandemic as virtual visits became the norm, and that trend may continue.
"The importance of considering prescribing across all encounters may only grow if ongoing shifts towards remote care continue," they concluded.
