The multiplex polymerase chain reaction respiratory pathogen panel (RPP) is used frequently in emergency departments (EDs) for the rapid identification of viruses and atypical bacteria of the respiratory tract. Its clinical value is unclear, as numerous studies have demonstrated that its use has a limited impact on antibiotic prescribing. We aimed to describe the relationship between RPP results and antibiotic prescribing rates for ED patients in our large academic medical center.
We retrospectively analyzed the charts of 1,061 patients aged 18-90 who were treated and released from two EDs from 1/1/2015 to 1/31/2018 and underwent RPP testing. Patients with evidence of bacterial infection were excluded based on RPP detection of atypical bacteria and microbiological analysis of blood, urine, wound, and sputum specimens. The results of the RPP and the rates of subsequent respiratory pathogen-directed antibiotic prescribing (including ED and outpatient pharmacy orders) were compared.
Antibiotic prescription rates were 21.5% in patients who tested negative for any respiratory virus, compared to 14.5% in patients who tested positive (OR 0.70, p<0.01). When positive RPPs were subdivided based on virus type (influenza and non-influenza) and compared to negative RPPs, only influenza-detection was associated with a significant reduction in antibiotic prescriptions (table 1).
In our study population, the presence of a respiratory virus detected by the RPP was correlated with a significant decrease in antibiotic prescribing. This effect was largely driven by influenza detection. This demonstrates that at our institution, the RPP may have a role in reducing unnecessary antibiotic utilization, but providers need further guidance in the interpretation of non-influenza respiratory virus positivity.
N No. of patients given antibiotics Odds Ratio (95% CI) P-value Negative 628 135 (21.5%) Reference Positive 433 63 (14.5%) 0.70 (0.56-0.88) <0.01 Influenza* 169 20 (11.8%) 0.49 (0.30-0.81) <0.01 Non-influenza virus(es) 264 43 (16.3%) 0.71 (0.49-1.04) 0.08 *Includes RPPs that
were positive for multiple viruses if influenza was present
No. of patients given antibiotics
Odds Ratio (95% CI)
*Includes RPPs that were positive for multiple viruses if influenza was present
D. Ding, None
S. Hochman, None
M. Phillips, None