Table 4.

A summary of the output from 5000 simulated trial datasets under the ‘null hypothesis’ scenario for the PIMS simulations

Variable	Description of the variable	Minimum	Q1	Median	Mean	Q3	Maximum
nevents0	Number of events in the control group	2	26	29	29	33	49
nevents1	Number of events in the treatment group	2	26	29	29	33	47
pevents0	Proportion of events in the control group	0.03	0.09	0.10	0.10	0.11	0.23
pevents1	Proportion of events in the treatment group	0.02	0.09	0.10	0.10	0.11	0.25
sample_size	Sample size	141	584	584	583	584	584
interim_time	Time at the interim (days)	80	169	197	199	227	378
interim_or	Odds ratio at the interim	0.10	0.78	1.00	1.12	1.26	11.55
interim_lci	Lower CI for OR at interim	0.02	0.30	0.39	0.43	0.50	3.15
interim_uci	Upper CI for OR at interim	0.34	2.00	2.55	2.99	3.31	74.69
interim_p	P-value for any difference between treatments at the interim analysis	0.00	0.33	0.62	0.52	0.66	1.00
interim_stop	Whether the trial would have stopped at the interim (based on decision criteria at interim)	0.00	0.00	0.00	0.00	0.00	1.00
final_or	Odds ratio at final analysis	0.36	0.83	1.00	1.04	1.21	2.74
final_lci	Lower CI for odds ratio	0.19	0.48	0.58	0.60	0.70	1.52
final_uci	Upper CI for odds ratio	0.66	1.43	1.73	1.81	2.10	5.15
final_p	P-value for treatment difference at final analysis	0.00	0.25	0.49	0.50	0.77	1.00
final_stop	Whether the trial is conclusive at final analysis	0.00	0.00	0.00	0.04	0.00	1.00
stop	Whether the trial was conclusive (at the interim or at the final analysis)	0.000	0.000	0.000	0.043	0.000	1.000
flipflop	The probability of trial flip-flopping	0.000	0.000	0.000	0.001	0.000	1.000

Q1: 1st quartile; Q3: 3rd quartile. In bold is the proportion of trials that conclude as a success when there is no treatment effect (i.e., type I error)