The goal of this study is to evaluate an educational intervention that aims to reduce the number of unnecessary antibiotics prescribed by family physicians and nurse practitioners in British Columbia, Canada. The intervention materials include a confidential personalized prescribing "portrait" and an evidence-based educational summary (therapeutics letter), accompanied by an introduction letter. The main research questions are: 1. Will the intervention lead to a reduction in the overall number of antibiotics prescriptions started? 2. Will the intervention lead to a reduction in the proportion of antibiotics prescribed that are likely unnecessary, especially prescriptions for upper respiratory tract infections, acute bronchitis, acute sinusitis? Researchers will conduct an intervention study with family physicians and nurse practitioners in British Columbia, Canada. Participant clinicians will be randomly assigned to one of two groups. The Early Group will consist of 80% of the participants and will receive the intervention (prescribing portrait, evidence summary, and introduction letter) at the start of the study. The Delayed Group will consist of 20% of participants and will receive the intervention about nine months later. This study design allows most practitioners to receive the intervention early while still allowing time to compare the two groups to assess the impact. To estimate the impact of the intervention, researchers will use administrative health data to compare the prescribing of the Early Group with prescribing of the Delayed Group.
Audit and feedback is a well-established strategy for improving the quality of care in primary healthcare settings. It involves the systematic collection and analysis of healthcare data, followed by the provision of information to healthcare providers to improve the quality of care. Antimicrobial resistance is a growing global health problem, fueled in part by unnecessary antibiotic use. Most antibiotics are prescribed in the community, especially for respiratory infections, and many of these prescriptions provide little or no benefit, particularly for viral illnesses. Effective antimicrobial stewardship in primary care is essential. Based in British Columbia, Canada, the Therapeutics Initiative's Portrait program provides family physicians and nurse practitioners with individualized prescribing data compared with their peers, paired with evidence-based messaging, and a concise statement summarizing the desired change in prescribing behavior. The Portrait program implements a series of pragmatic trials to improve prescribing where clinicians are randomized into an early intervention and a delayed control group. The current study evaluates the impact of an antibiotic prescribing portrait and evidence summary focused on reducing overall antibiotic use and avoiding routine prescribing for common respiratory tract infections (upper respiratory tract infections, acute bronchitis, acute sinusitis) in primary care. The prescribing portrait title is "What are your antibiotic prescribing rates?" and is accompanied by a therapeutics letter (evidence summary) entitled "Avoid antibiotics for self-limited viral infections" and an introduction letter. The portrait contains two key recommendations based on recent systematic reviews on the effect of antibiotic prescribing on antimicrobial resistance in primary care: 1) Reduce the number of your patient visits that end with an antibiotic prescription, and 2) Avoid routinely prescribing antibiotics for upper respiratory tract infections, acute bronchitis, and acute sinusitis. The first recommendation is accompanied by a bar graph showing the prescriber their number of antibiotic prescriptions per 100 patients in 2024 in comparison with the median practice in BC and an achievable target. The second recommendation is accompanied by a bar graph showing the prescriber's percentage of antibiotic prescriptions identified as likely unnecessary in 2024 in comparison with the median practice in BC and an achievable target. The prescribing portrait data were generated using linked administrative health claims data from the BC Ministry of Health. Data access is provided according to the Information Sharing Agreement with the BC Ministry of Health (ISA 16-036). To maintain data security, data will be accessed via the Ministry of Health's Secure Analysis Environment, and include: * BC Medical Services Plan (MSP) physician services data * Hospital separations from the Discharge Abstracts Database (DAD) * BC PharmaNet data (prescription drugs dispensed at community pharmacies) * Client Roster (client demographic data) * Practitioner Roster (practitioner demographic data) 7,626 physicians and nurse practitioners in British Columbia were randomized in an 80:20 ratio, with 80% assigned to receive the intervention materials early and 20% assigned to a delayed control group (6,100 intervention; 1,526 control). Clinicians in the early group will be mailed the intervention at study initiation (February 2026), while those in the delayed group will be mailed the same materials after an approximate 9-month delay. Study group assignment will be conducted using a random number generator by an independent analyst. Researchers selected this unequal allocation to facilitate earlier dissemination of the intervention to a larger proportion of clinicians while maintaining a control group for estimation of the intervention effect. Researchers will use a modified intention-to-treat approach for all analyses, with clinicians analyzed based on their randomized assignment to early or delayed receipt of the prescribing portrait. Clinicians who opt out will be excluded from analysis. Intervention effects will be estimated using Poisson regression to estimate rates, rate differences and rate ratios for the outcome, with the delayed group serving as a contemporaneous control group to account for secular changes in prescribing. The primary outcome is the weekly rate of new (incident) antibiotic prescriptions per practitioner. A prescription is considered "new" if the patient has not received an antibiotic from the same prescriber in the previous 30 days. Repeat prescriptions within an ongoing course of care are excluded. Secondary outcomes are: 1) the weekly rate of likely unnecessary incident antibiotic prescriptions per practitioner, and 2) the weekly rate of likely unnecessary, condition-specific incident antibiotic prescriptions per practitioner for upper respiratory tract infections (URTI), acute bronchitis, and acute sinusitis. Researchers will summarize baseline clinician characteristics by study group using descriptive statistics. To evaluate the durability of the intervention effect, researchers will assess primary and secondary outcomes at 1, 3, and 6 months, as well as over the full delay period (approximately 9 months).
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
HEALTH_SERVICES_RESEARCH
Masking
SINGLE
Enrollment
7,626
The prescribing portrait audit-and-feedback intervention consists of a confidential, personalized prescribing audit-and-feedback report (Portrait), an evidence-based educational summary (therapeutics letter), and an introduction letter. The prescribing portrait uses linked BC Ministry of Health administrative claims data and presents two key recommendations: (1) reduce the proportion of patient visits resulting in an antibiotic prescription, and (2) avoid routinely prescribing antibiotics for upper respiratory tract infections, acute bronchitis, and acute sinusitis. Each recommendation is accompanied by a bar graph showing the prescriber's 2024 performance compared with the median prescriber in British Columbia and an achievable target. The intervention will be delivered using a secure paper mailing process. During every aspect of production and delivery, prescribing data are confidential and protected.
Therapeutics Initiative - Dept of Anesthesiology, Pharmacology & Therapeutics, Faculty of Medicine, University of British Columbia
Vancouver, British Columbia, Canada
Weekly rate of incident oral antibiotic prescribing
The primary outcome is the number of incident oral antibiotic prescriptions per practitioner per week. Follow-up time will be calculated for each practitioner in the 52 weeks before and 39 weeks after the portrait is mailed to the Early group. A practitioner will count as contributing a full week (7 days) of follow-up if, during that week, they either: 1) wrote at least one prescription (recorded in PharmaNet), or 2) billed the Medical Services Plan for at least one patient visit. Weekly prescribing rates will be plotted by Early and Delayed groups and analyzed using Poisson regression with generalized estimating equations to account for repeated measures. Models will adjust for cohort (Early vs. Delay), age, sex, rural practice, and provider type. The intervention effect will be estimated using a pre-post level-change variable and its interaction with cohort status, representing the adjusted mean impact of the portrait.
Time frame: From trial initiation to 30 days, 3, 6, and 9 months post-intervention
Weekly rate of 'likely unnecessary' antibiotic prescriptions
Weekly rate of 'likely unnecessary' incident oral antibiotic prescribing per practitioner. An antibiotic prescription is defined as 'likely unnecessary' if the dispensing occurs within 5 days of a physician visit with a diagnosis where oral antibiotic treatment is 'never appropriate'. The 'achievable target' comparator is defined as the top 25% of physicians with the lowest proportion of likely unnecessary prescriptions. is defined as the top 25% of physicians with the lowest proportion of likely unnecessary prescriptions. Analyses will be conducted and reported using the same methodology as the primary analyses.
Time frame: From trial initiation to 30 days, 3, 6, and 9 months post-intervention
Weekly rate of 'likely unnecessary' condition-specific antibiotic prescriptions
Weekly rate of 'likely unnecessary' condition-specific incident oral antibiotic prescribing per practitioner for the following conditions: 1) upper respiratory tract infections (URTI), 2) acute bronchitis, 3) acute sinusitis. An antibiotic prescription is defined as 'likely unnecessary' if the dispensing occurs within 5 days of a physician visit with a diagnosis where oral antibiotic treatment is 'never appropriate'. The 'achievable target' comparator is defined as the top 25% of physicians with the lowest proportion of likely unnecessary prescriptions. Analyses will be conducted and reported using the same methodology as the primary analyses.
Time frame: From trial initiation to 30 days, 3, 6, and 9 months post-intervention
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.