1022. Antibiotic Prescribing for Acute Respiratory Infections Increases as the Clinic Session Wears On
Session: Poster Abstract Session: Stewardship: Implementing Programs
Friday, October 4, 2013
Room: The Moscone Center: Poster Hall C
  • 1022_IDWPOSTER.pdf (528.1 kB)
  • Background: Fatigue and stress may deplete clinicians’ capacity to resist prescribing antibiotics for acute respiratory infections (ARIs). We hypothesized that primary care clinicians would be less likely to prescribe antibiotics for ARIs at the beginning and more likely to prescribe antibiotics for ARIs at the end of clinic sessions.

    Methods: We compared the antibiotic prescribing rates for all ARIs, antibiotic-appropriate diagnoses (e.g., sinusitis or pneumonia), and non-antibiotic-appropriate diagnoses (e.g., acute bronchitis) for the first visit of a half-day clinic session and the last hour of a half-day clinic session (i.e., the 11 am and 4 pm hours) to the remaining visits. We used multivariable generalized estimating equations to identify independent predictors of antibiotic prescribing after adjusting for patient demographics, antibiotic-appropriateness of diagnosis, and clustering by clinician.

    Results: There were 678,982 total visits and 31,838 ARI visits scheduled from 8 am to 5 pm, Monday to Friday, by patients aged 18 to 64 years old without chronic lung disease to 561 clinicians in 23 Boston-area primary care practices, between May 2011 and September 2012. The antibiotic prescribing rate for the first visit of the session was significantly lower than later visits for all ARIs (55.2% vs. 58.7%; p<0.001), antibiotic-appropriate diagnoses (74.2% vs. 77.4%; p=0.008), and non-antibiotic-appropriate diagnoses (45.1% vs. 48.8%; p<0.001). The antibiotic prescribing rate was significantly higher during the last hour of each session compared to the remainder of the session for all ARIs (60.3% vs. 57.6%; p<0.001), antibiotic-appropriate diagnoses (78.8% vs. 76.4%; p=0.01), and non-antibiotic-appropriate diagnoses (50.6% vs. 47.6%; p<0.001). In multivariable modeling, clinicians were less likely to prescribe antibiotics at the first visit of the session (odds ratio [OR], 0.88; 95% confidence interval [CI], 0.81 to 0.96) and more likely to prescribe antibiotics in the last hour of a clinic session (OR, 1.12; 95% CI, 1.04 to 1.19).

    Conclusion: Primary care clinicians are slightly, but significantly less likely to prescribe antibiotics at the beginning of a clinic session and more likely to prescribe antibiotics at the end of a clinic session.

    Jeffrey Linder, MD, MPH, FACP1, Jason Doctor, PhD2, Mark Friedberg, MD, MPP3, Harry Reyes Nieva1, Caroline Birks, MD4, James Falcone5, Yelena Kleyner5, Daniella Meeker, PhD6 and Craig Fox, PhD7, (1)Brigham & Women's Hospital, Boston, MA, (2)University of Southern California, Los Angeles, CA, (3)RAND, Boston, MA, (4)Massachusetts General Hospital, Boston, MA, (5)Partners Healthcare System, Wellesley, MA, (6)RAND, Santa Monica, CA, (7)University of California Los Angeles, Los Angeles, CA


    J. Linder, None

    J. Doctor, None

    M. Friedberg, None

    H. Reyes Nieva, None

    C. Birks, None

    J. Falcone, None

    Y. Kleyner, None

    D. Meeker, None

    C. Fox, None

    Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.