
Background: Acute pharyngitis is a
common diagnosis in ambulatory pediatrics. While group A streptococcus (GAS)
is the most common bacterial cause of acute pharyngitis in children, the vast
majority of cases are caused by viruses. Nationally, antibiotics are
prescribed in excess of the maximum expected prevalence of GAS pharyngitis. The
Infectious Diseases Society of America (IDSA) clinical practice guideline for
GAS pharyngitis recommends strict criteria for testing to avoid misdiagnosis of
colonized children. The majority of testing (~60%) for GAS pharyngitis in this
pediatric practice did not meet IDSA criteria prior to intervention. Methods:
Charts were included from patients with a procedure code for GAS rapid antigen
detection test (RADT). Retrospective chart review was performed. The outcome
measure, unnecessary GAS testing, was defined as any patient meeting 1 or more
of the following criteria: 1) age < 3 years 2) presence of viral symptoms
3) absence of sore throat 4) absence of expected GAS pharyngitis exam
findings. A convenience sample of 20 charts per month was included for October
1, 2013-January 31, 2016. Charts were excluded if RADT results were unknown. Iterative
Plan-Do-Study-Act cycles were completed. Interventions include: 1) educational
webinar 2) review of current state with clinic providers and development of new
office procedure 3) updated data report to providers. Results: Over
the project period, 60.3% of GAS pharyngitis testing was unnecessary (Figure 1).
The primary reason for unnecessary testing was presence of viral symptoms,
followed by absence of sore throat. The unnecessary tests resulted in 87
antibiotic prescriptions that were not indicated. Lower proportions of
unnecessary testing were seen following Intervention #2. Conclusion: The
majority of GAS pharyngitis testing in this practice was not clinically
indicated prior to intervention. Testing for GAS in children when it is not
clinically indicated can result in unnecessary antibiotic exposure. Discussion
of appropriate testing practices and adoption of new office procedures may
decrease unnecessary testing. Further tests of change are needed to determine
if sustainable improvement was made. Future tests of
change will be designed based on feedback from the third intervention.
Figure 1. Control chart of outcome measure

L. Norton,
None
L. Harte, None
A. Myers, None