A Bayesian non-inferiority approach using experts’ margin elicitation – application to the monitoring of safety events

Table 5 Final decision rule retained through the sensitivity analysis: thresholds to be applied at each interim analysis and final overall rates of misclassifications, according to the event

	Event
	Death	IVH ^[2]	NEC ^[3]	Retinopathy
Thresholds \(\tau *^{L}_{j}\)^[1], to applied in analysis
1	0.95	0.95	0.95	0.95
2	0.95	0.95	0.95	0.95
3	0.95	0.95	0.95	0.95
4	0.95	0.95	0.95	0.95
5	0.949	0.949	0.950	0.949
6	0.948	0.948	0.949	0.948
7	0.944	0.944	0.948	0.946
8	0.934	0.933	0.944	0.938
9	0.907	0.903	0.932	0.918
10	0.832	0.821	0.902	0.862
11	0.63	0.60	0.82	0.71
Overall rate of errors
Class a misclassifications ^[4]	0.09	0.09	0.10	0.10
Class b misclassifications ^[5]	<0.01	0.07	0.20	0.27

¹\(\tau *^{L}_{j}\) : threshold to apply in the Bayesian decision rule for the event k in the subgroup j, at the l interim analysis: the rule will conclude that their is an unacceptable excess if \( {P (\delta ^{l}_{j}) \geq \tau ^{l}_{j}}\)
²IVH: Intraventricular haemorrhage
³NEC: Necrotizing enterocolitis
⁴Class a misclassifications: Trials that conclude that the difference between arms is Unacceptable, while it is not true
⁵Class b misclassifications: Trials that conclude that the difference between arms is Acceptable, while it is not true

ISSN: 1471-2288