Using Bayesian statistics to estimate the likelihood a new trial will demonstrate the efficacy of a new treatment

Table 1 Description of the retrieved trials

First author	e-date	Imaging	Female/male^a	Age (mean)	BMI (mean)	RSG	AC	BOPH	BOA	IOD	SR	OB
Chareancholvanich	02/03/2013	mri	70/10	70	28	low	uk	high	low	low	low	uk
Victor	26/04/2013	mri/ct	86/42	67	-	uk	low	high	low	high	low	high
Roh	03/08/2013	ct	82/8	70	27	low	low	high	uk	high	low	uk
Hamilton	06/08/2013	ct	31/21	68	31	uk	uk	high	low	high	low	uk
Boonen	10/08/2013	mri	106/74	67	30	low	low	high	low	low	low	low
Parratte	15/08/2013	mri	24/16	71	29	high	low	high	low	low	low	uk
Chotanaphuti	04/09/2013	ct	70/10	70	25	uk	uk	high	low	low	low	low
Woolson	07/03/2014	ct	0/63	66	33	low	low	high	low	high	low	low
Kotela	28/06/2014	ct	66/29	67	30	uk	uk	high	low	low	low	low
Pfitzner	16/07/2014	mri/ct	51/39	65^b	30	low	low	high	low	low	low	low
Yan	14/09/2014	mri	41/19	69	-	low	uk	high	low	low	low	uk
Abane	09/01/2015	mri	88/52	69	29	low	uk	high	low	high	low	low
Molicnik	04/03/2015	mri	31/7	67	33	uk	uk	high	uk	uk	low	uk

RSG random sequence generation, AC allocation concealment, BOPH blinding of participants/care providers, BOA blinding of outcome assessors, IOD incomplete outcome data, SR selective reporting, OB other biases. ^a sex ratio in shown for patients randomized and outcome is shown for patients analyzed, therefore numbers may differ. ^b mean of group’s median

ISSN: 1471-2288