Fig. 2

a Net benefit curves of three clinical trial enrollment strategies, enrolling all sepsis patients (black line), enrolling patients with septic shock (blue line), and enrolling IL-8-positive patients (red line), in a population restricted to immunocompetent patients. The x-axis represents the threshold probability, which is the probability of sepsis mortality that a hypothetical trial would require for enrollment. The y-axis is the net benefit, which represents the tradeoff between true positives and false positives, and is described further in the legend of Fig. 1. The net benefit curves are interpreted vertically, such that at each threshold probability, the strategy with the highest net benefit is the optimal strategy for enriching a trial with high-risk subjects. For example, if a trial sought to enroll patients with at least 35% mortality risk (dotted vertical line), enrolling only IL-8 positive patients is the optimal strategy. b Intervention curves comparing enrolling all sepsis patients (reference, not shown), enrolling patients with septic shock (blue line), and enrolling IL-8-positive patients (red line), in a population restricted to immunocompetent patients. The y-axis represents the number of survivors that avoid the intervention, which in this case is enrollment and exposure to unproven therapy. Intervention curves are also interpreted vertically. For example, at a threshold probability of 35%, enrolling only IL-8-positive patients would lead to a greater reduction in the number of survivors unnecessarily exposed compared to enrolling patients with septic shock