Two-stage Bayesian hierarchical modeling for blinded and unblinded safety monitoring in randomized clinical trials

Table 8 The probability of flagged trials for the AEs under flat effect where both the control and treatment arms are the same but higher than the expected incident rate for scenario V

Example AEs (True Proportion; Expected Event Rate)	Blinded: Proportion with signal for early termination Unblinded: Termination Proportion with confirmed by unblinded data	(P_crit1, P_crit2)
Example AEs (True Proportion; Expected Event Rate)		(0.9, 0.7)	(0.9, 0.8)	(0.9, 0.9)
Pneumothorax Induced by HBO therapy^a (π = 3%; π_M = 2%)	Blinded	0.32	0.33	0.33
	Unblinded	0.45	0.24	0.09
	Overall Rate	0.15	0.08	0.03
Signs of Pulmonary Dysfunction^a (π = 32%; π_M = 25%)	Blinded	0.60	0.60	0.59
	Unblinded	0.40	0.25	0.12
	Overall Rate	0.24	0.15	0.07
Pneumonia^a (π = 49%; π_M = 40%)	Blinded	0.65	0.66	0.64
	Unblinded	0.40	0.27	0.13
	Overall Rate	0.26	0.18	0.08
Critical decreased CPP^a (< 60 mmHg) (π = 81%; π_M = 75%)	Blinded	0.62	0.63	0.62
	Unblinded	0.42	0.28	0.13
	Overall Rate	0.26	0.18	0.08
Critical hypotension^a (MAP< 70 mmHg) (π = 81%; π_M = 75%)	Blinded	0.61	0.62	0.61
	Unblinded	0.42	0.28	0.12
	Overall Rate	0.26	0.17	0.07
Seizures during HBO treatment^a (π = 1%; π_M = 1%)	Blinded	0.28	0.28	0.28
	Unblinded	0.44	0.26	0.09
	Overall Rate	0.12	0.07	0.02
Hypercarbia during transportation^a (PaCO2 > 45 mmHg) (π = 14%; π_M = 10%)	Blinded	0.49	0.49	0.49
	Unblinded	0.41	0.27	0.11
	Overall Rate	0.20	0.13	0.05

ISSN: 1471-2288