Skip to main content
Fig. 6 | BMC Medical Research Methodology

Fig. 6

From: Simulating complex patient populations with hierarchical learning effects to support methods development for post-market surveillance

Fig. 6

Adverse outcome rate in the presence of treatment effect and learning with varying sample sizes. Curves show smoothed probability (and 95% confidence intervals with corresponding shaded regions) of an adverse outcome over the case series. Provider learning-associated risk was specified as starting at 50% of the mean patient-associated risk among patients assigned to the novel treatment and declined toward 0 over the provider’s first 25 cases. Number of institutions simulated was set to 30 (a), 10 (b), and 5 c. In each simulated dataset, 50% of the population was assigned the novel treatment, the population adverse outcome rate was 10%, and the odds ratio of an adverse outcome for the novel treatment compared to the reference treatment was 2.0

Back to article page