Behavioral Interventions Help Reduce Inappropriate Antibiotic Prescriptions for Acute Respiratory Infections
EBM Focus - Volume 11, Issue 9
- Antibiotics are often unnecessarily prescribed for acute respiratory tract infection. To address this issue, 4 points of advice for reducing antibiotic misuse were recently published by the American College of Physicians and Centers for Disease Control and Prevention.
- A recent cluster randomized trial also evaluated 3 behavioral interventions for reducing antibiotic misuse during acute respiratory tract infections.
- Two interventions, requiring physicians to provide justifications for antibiotic prescriptions and emailing physicians reports comparing their performance to other physicians in the region, effectively reduced inappropriate antibiotic prescriptions for acute respiratory tract infections.
It is estimated that up to 50% of antibiotics prescribed at more than 100 million ambulatory care visits in the United States each year are unnecessary (J Antimicrob Chemother 2014 Jan;69(1):234, CDC 2013 PDF). This misuse of antibiotics contributes to higher health costs and increasing antibiotic resistance. Acute respiratory tract infections (acute uncomplicated bronchitis, pharyngitis, rhinosinusitis, and the common cold) are the most common reason for outpatient antibiotic prescriptions even though most are caused by viruses. The American College of Physicians and Centers for Disease Control and Prevention recently published a clinical guideline addressing inappropriate prescribing for acute respiratory tract infections (Ann Intern Med 2016 Jan 19 early online). The four items of high-value care advice are as follows:
- In patients with bronchitis, do not perform testing or give antibiotic therapy unless pneumonia is suspected.
- In patients with pharyngitis, use a rapid antigen detection test or culture for group A Streptococcus in patients with symptoms suggestive of streptococcal pharyngitis and only prescribe antibiotics if a Streptococcal infection is confirmed.
- In patients with acute rhinosinusitis, only prescribe antibiotics if patients have 1 of the following: persistent symptoms for > 10 days, severe symptoms, high fever (> 39 degrees Celsius, 102.2 degrees Fahrenheit) plus purulent nasal discharge, facial pain for ≥ 3 consecutive days, or worsening symptoms after a typical viral infection lasting ≥ 5 days that was initially improving.
- In patients with the common cold, do not prescribe antibiotics.
While guidelines such as these are important, it can be difficult to change common prescribing practices. To help find interventions effective at reducing inappropriate antibiotic prescribing, a recent cluster randomized trial evaluated 3 behavioral interventions for reducing inappropriate antibiotic prescribing at 47 primary care practices including 248 clinicians in Boston, Massachusetts and Los Angeles, California.
Practices were randomized in a factorial design to 0, 1, 2, or 3 behavioral interventions. Interventions included suggested alternatives where an acute respiratory tract infection diagnosis triggered a warning against antibiotic use in the electronic health record and presented alternative nonantibiotic treatment regimens, accountable justifications requiring physicians prescribing antibiotics to provide a free-text justification for the prescription in the electronic health record, and peer comparison providing email reports comparing physician rates of inappropriate antibiotic prescribing in a given region. Antibiotic prescribing practices for adult patients with “antibiotic-inappropriate acute respiratory tract infections” were evaluated for 18 months before and 18 months after the intervention. Patients with guideline-specified comorbidities or concomitant infections were excluded. A total of 31,712 outpatient visits were included in the analysis (14,753 during the baseline period and 16,959 during the intervention period). The rates of inappropriate antibiotic prescribing at the start of the intervention period compared to the end of the intervention period can be found in the table below. To assess efficacy, the differences between inappropriate antibiotic prescribing with each intervention vs. control were compared over time to determine a trend. In this analysis, the accountable justifications and peer comparison interventions were both significantly associated with reduced inappropriate antibiotic use over time (p < 0.001 for each). The difference in the suggested alternatives vs. control trends did not reach significance.
While the rate of inappropriate prescriptions gradually declined in all interventions as well as the control group, accountable justifications and peer comparisons were both associated with significantly lower rates of inappropriate prescriptions compared to control at the end of the intervention period. Although these results are not definitive, they do show that insights from behavioral science may be helpful in developing interventions to effectively translate clinical guidelines into everyday practice. The results of this trial may aid initiatives attempting to reduce antibiotic overuse.