TABLE A4.

Proportion of times (out of 2000) the stopping trigger was met under Bayesian and frequentist monitoring approach. The frequentist looks are happening less frequently that the Bayesian looks to limit $\alpha -$spending. The stopping rules are Bayesian: $P({\text{OR}}_{co}<1)\ge 0.95$ and $P({\text{OR}}_{co}<0.8)\ge 0.50$ and $P({\text{OR}}_l<1)\ge 0.95$ and $P({\text{OR}}_l<0.8)\ge 0.50$ Frequentist: O'Brian‐Fleming approach with 5 data looks preserving overall $\alpha =0.05$, with the respective $\alpha$ values at each look shown under the % information available

	Bayesian approach		O'Brian‐Fleming approach
	Information	% of simulation trigger met	Information	% of simulations $P<\alpha$
Effect	$({\delta }_0,{\delta }_1,{\delta }_2)=(0,0,0)$
	20%	0.64	20%, $\alpha =0.000005$	0
	33%	0.2
	40%	0.2	40%, $\alpha =0.0013$	0
	50%	0.39
	60%	0.54	60%, $\alpha =0.0085$	0.2
	67%	0.39
	80%	0.34	80%, $\alpha =0.0228$	0.25
	90%	0.25
	100%	0.15	100%, $\alpha =0.0417$	1.23
Type I error	Total	3.1	Total	1.68
Effect	$({\delta }_0,{\delta }_1,{\delta }_2)=(0.1,0.2,0.3)$
	20%	2.18	20%, $\alpha =0.000005$	0
	33%	3.77
	40%	3.33	40%, $\alpha =0.0013$	0.2
	50%	2.73
	60%	3.72	60%, $\alpha =0.0085$	2.28
	67%	2.58
	80%	3.97	80%, $\alpha =0.0228$	5.36
	90%	2.98
	100%	2.83	100%, $\alpha =0.0417$	8.24
Power	Total	28.09	Total	16.08
Effect	$({\delta }_0,{\delta }_1,{\delta }_2)=(0.4,0.5,0.6)$
	20%	15.81	20%, $\alpha =0.000005$	0.1
	33%	22.13
	40%	12.58	40%, $\alpha =0.0013$	7.88
	50%	12.68
	60%	9.97	60%, $\alpha =0.0085$	29.91
	67%	4.91
	80%	6.78	80%, $\alpha =0.0228$	28.13
	90%	4.28
	100%	2.51	100%, $\alpha =0.0417$	17.85
Power	Total	91.65	Total	83.87