The use of audit and feedback to improve patient care and reduce practice variation in paediatric tonsil surgery

Rationale: In international guidelines, it is recommended to only perform a tonsillectomy for
recurrent throat infection if there have been more than seven episodes in the past year, more
than five episodes per year in the past two years, or more than three episodes per year in the
past three years. Additionally, it is recommended to perform a watchful waiting strategy
instead of adenoidectomy in case of upper respiratory tract infections only. Despite these
recommendations, practice variation remains problematic. It is thought that awareness can
reduce practice variation. Therefore, this study aims to reduce practice variation in
tonsillectomies and adenoidectomies by using audit and feedback to otolaryngologists and
general practitioners (GPs). The study consists of 2 phases: 1. Development and
evaluation of an implementation strategy to reduce practice variation in tonsillectomy and
adenoidectomy using data from the LOGEX Benchmark Database. We will perform a
randomized controlled trial on the effect of the implementation strategy. Fifteen randomly
selected hospitals from the LOGEX Benchmark Database will receive an invitation to
participate in our study. We will compare clinical practice before and after the intervention in
the participating hospitals, and we will compare the intervention hospitals with the control
group, which consists all other hospitals from the LOGEX Benchmark Database, 2. Analysis
of practice variation in GPs referral pattern using data from the ELAN Datawarehouse (Den
Haag). We will present the outcomes to the GP practices included in the analysis. We will
evaluate the effectiveness of audit and feedback on practice variation, process variation,
outcomes and costs in the treatment of tonsillitis and upper respiratory infections in children
to otolaryngologists and GPs.
Main study parameters/endpoints: Main outcome is practice variation in tonsillectomy and
adenoidectomy treatment for children with upper respiratory tract infections before and after
giving audit and feedback to otolaryngologists. Also, changes in clinical practice (both
treatment choices and treatment processes) will be analysed using difference-in-difference
analysis comparing intervention hospitals with the control group. Hospital characteristics will
be compared with the opinion of physicians on the topic. Furthermore, we will evaluate
practice variation in GP practices concerning the treatment of children with upper respiratory
tract infections.

We hypothesize that providing feedback to hospitals will reduce practice variation

We measure ongoing (per months) from 6 months before until 6 months after the last feedback cycle

Audit and feedback, 4 times, during 1 year