Back to History

Fields Changed

Registration

Field Before After
Trial End Date September 15, 2024 August 01, 2025
Last Published July 01, 2024 12:58 PM May 17, 2025 09:05 AM
Intervention End Date September 15, 2024 July 31, 2025
Primary Outcomes (Explanation) Quality score and binary indicators related to (i) antibiotic dispensing practices and (ii) health condition behavior & monitoring Binary indicators related to (i) antibiotic dispensing practices
Planned Number of Observations 1,600 (4 observations per pharmacist (N=400). Original registration: 1,600 (4 observations per pharmacist (N=400). Revised registration: 3,325 (7 observations per pharmacist (N=475))
Sample size (or number of clusters) by treatment arms 800 (2 observations per pharmacist (N=400). Original registration: 800 (2 observations per pharmacist (N=400). Revised registration: 1662 (2 observations per pharmacy (N=475)).
Power calculation: Minimum Detectable Effect Size for Main Outcomes Comparison across alternative diagnostic uncertainties: In a basic group means comparison with 80% power, a significance level of 0.05, and 0.8 unit standard deviations, the MDE is 18pp for the total sample of 1,600 observations (4 per pharmacy). (Assumption: Baseline mean of 0.3 & 1 treatment vs. 1 control group comparison). Comparison across alternative diagnostic uncertainties: In a basic group means comparison with 80% power, a significance level of 0.05, and 0.8 unit standard deviations, the MDE is 18pp for the total sample of 1,600 observations (4 per pharmacy). (Assumption: Baseline mean of 0.3 & 1 treatment vs. 1 control group comparison). Revised registration: MDE is 7.8pp keeping all parameters as they were except for the newly registered changes in sample size
Intervention (Hidden) • Vignettes: Individuals will receive 4 vignettes that aim to vary patients’ symptoms to signal more/less clear underlying medical conditions. The order of the vignettes is randomized. • Audit study: We employ an audit study in which characters portray different systems with characters being linked to diagnostically more or less clear medical conditions. The order of characters visiting pharmacies is randomized. • Vignettes: Individuals will receive 4 vignettes that aim to vary patients’ symptoms to signal more/less clear underlying medical conditions. The order of the vignettes is randomized. Correct answers to the vignettes are financially incentivized. • Audit study: We employ an audit study in which characters portray different systems with characters being linked to diagnostically more or less clear medical conditions. The order of characters visiting pharmacies is randomized.
Back to top