Figure 4: Type I (alpha) and Type II (beta) error boundaries over the accrued sample size. Note:Treatment effect of IV acetaminophen versus placebo plotted with the pre-determined boundaries for efficacy, harm, and futility for hypoxia. The vertical axis is the standardized effect size (Z), defined as the difference in group proportions divided by the standard error of the difference. Z = 0 indicates no treatment effect, while Z > 0 indicates that IV acetaminophen has more hypoxia than placebo, and Z < 0 indicates that IV acetaminophen has a lower incidence of hypoxia than placebo. The horizontal axis indicates the number of planned (N = 580) patient enrollment. At each of the interim analyses (25%, 50%, 75%, and 100% of the total sample size), crossing a blue dot (into the blue shaded regions) indicates efficacy while crossing a red dot (into the red shaded region) indicates futility.