This secondary analysis includes all providers from the Use of Behavioral Economics to Improve Treatment of Acute Respiratory Infections (BEARI) trial who completed the exit survey following 18 month trial completion. The objective of this study is to test the hypothesis that the Peer Comparison intervention in the BEARI trial (Meeker et al. 2016) promoting antibiotic stewardship did not adversely impact physician job satisfaction. Detrimental impacts on job satisfaction is a phenomenon that was observed in a randomized controlled trial using a Peer Comparison intervention with different characteristics from the BEARI trial. (Reiff et al. 2022) The BEARI trial sample size, intraclass correlation, and measurement of job satisfaction are comparable to Reiff et al. 2022.
The differences in Peer Comparison interventions between BEARI and Reiff et al. include the following:
- Differences in physicians' control and agency over patient adherence to screening recommendations vs. their own antibiotic prescribing
- Relatedly, in the antibiotic stewardship study, there were achievable benchmarks for improvement, including a feedback and ranking framework allowing all physicians to attain the highest status and thus obtain positive feedback.
- Differences in framing and presentation of messages
The investigators define a clinically significant detrimental effect on job satisfaction as 27% of individuals reducing job satisfaction ratings by one point on a 5-point likert scale. This shift is equivalent to a mean difference of 0.32 and a Cohen's d of 0.36. This difference corresponds to approximately a ⅓ reduction in job satisfaction on a 5-point likert scale.
H0: The BEARI Peer Comparison intervention had a clinically and statistically significant detrimental effect on physician job satisfaction. Control-PeerComparison>=0.32
HA: The Peer Comparison intervention had no clinically significant negative impact on physician job satisfaction. Control-PeerComparison<0.32
The investigators will conduct a traditional hypothesis test Control-PeerComparison= 0.0 as a secondary analysis.
