Secondary Analysis Investigates Spillover Effects of Antibiotic Audit and Feedback

Antibiotics revolutionized medicine, but their overuse has caused a global crisis of antimicrobial resistance (AMR), which threatens public health through treatment failures and resistant pathogens, particularly from inappropriate outpatient prescribing. To investigate the broader impacts of antibiotic prescribing, researchers conducted a secondary analysis that followed previous trial results, which demonstrated that audits and feedback on antibiotic prescribing ultimately reduced prescribing in older adults. This post hoc secondary analysis, published by investigators in JAMA Network, aims to determine if this intervention had a spillover effect on antibiotic prescribing across the entire patient population.¹

Image credit: I Viewfinder | stock.adobe.com

According to WHO, AMR accounted for 1.27 million global deaths and contributed to 4.95 million deaths in 2019 alone. The condition occurs when bacteria, viruses, fungi, and parasites stop responding to antimicrobial medicines, which creates a drug resistance, making infections difficult to treat. The misuse and overuse of antimicrobials to treat, prevent, and control infections has influenced the spread of AMR.²

In the previous pragmatic randomized clinical trial conducted in Ontario, Canada, researchers aimed to evaluate the effectiveness of mailed audit-and-feedback reports on antibiotic prescribing to primary care physicians, using existing administrative data. The trial was conducted from January 2021 to December 2022, with a feedback letter sent on or after January 2022, including physicians that were randomly assigned in a 4 to 1 allocation. Results from the previous study revealed a 4% relative reduction in antibiotic prescribing rates at 12 months and a 14% relative reduction in prolonged duration prescribing among individuals aged 65 and older.¹

The current post hoc secondary analysis was conducted from March to June 2024 and aimed to further evaluate the spillover of an audit-and-feedback intervention that targets individuals aged 65 and older on a broader population of all age groups. Differing from the previous study, researchers used the IQVIA Xponent antibiotic database, which collects prescription data from nearly 62% of Ontario pharmacies. The study authors noted that this database allowed them to capture antibiotic dispensing outcomes for individuals of all ages.¹

The primary end point of the study was the total number of antibiotic prescriptions issued by physicians 12 months post-intervention for individuals across all age groups. The researchers evaluated this by using the new administrative data source that provided antibiotic prescription counts for every patient age group. Additionally, to account for baseline prescribing and patient demographics, they analyzed the data using Poisson regression models, stratified by patient age and sex.¹

Out of 5097 individuals, a total of 4964 were included in the analysis, with 3967 in the intervention group and 997 in the control group. Additionally, 2766 of the physicians were male, and 2549 have practiced for over 25 years. The results demonstrated similar findings to the previous study, with a 7% relative reduction in antibiotic prescription counts and an 18% relative reduction in prolonged duration prescription counts. However, the intervention group demonstrated a reduction in antibiotic prescriptions at 12 months after intervention compared with the control group, and the proportion of prescriptions exceeding 7 days decreased. Reductions in age and sex were displayed for antibiotics commonly used for respiratory infections.¹

The findings suggest that a reduction in antibiotic prescriptions across all patient ages within the intervention group was shown. These results demonstrate that routinely collected administrative data can effectively support the implementation and evaluation of antibiotic audit and feedback, even when limited to an older patient demographic.¹

“Future research should explore the long-term sustainability of the intervention’s associations with antibiotic prescribing. Additionally, investigating the underlying mechanisms driving behavior change in response to peer comparison audit and feedback could provide valuable insights for refining and optimizing such interventions,” the study authors concluded.¹

REFERENCES

1. Saqib K., Ivers N., Brown K., et al. Spillover From an Intervention on Antibiotic Prescribing for Family Physicians. JAMA Network. (2025) doi:10.1001/jamanetworkopen.2025.18261

2. Antimicrobial Resistance. World Health Organization. News release. November 21, 2023. Accessed July 16, 2025. https://www.who.int/news-room/fact-sheets/detail/antimicrobial-resistance