. 2015 Jun 30;26(9):1966–1973. doi: 10.1093/annonc/mdv283

Table 3.

Risk factors for non-reporting of QoL outcomes

	Univariable effect		Multivariable effect
	Odds ratio (95% confidence interval)	P	Adjusted odds ratio (95% confidence interval)	P
Planned target sample size^a	0.98 (0.91–1.06)	0.63	1.00 (0.93–1.08)	0.95
Discontinued RCT (versus completed)	2.62 (0.95–7.20)	0.062	2.65 (0.93–7.52)	0.067
Single-centre status (versus multicentre)	1.48 (0.13–16.98)	0.75	1.16 (0.13–20.31)	0.71
Investigator sponsorship (versus industry)	1.66 (0.69–4.00)	0.26	1.56 (0.62–3.91)	0.34

Of the 90 RCT that pre-specified QoL outcomes, we excluded seven RCTs with unclear completion status (none of them reporting any QoL outcomes) thus leaving 83 RCTs and a total of 48 events for analysis.

See supplementary Tables 1–3, available at Annals of Oncology online, for sensitivity analyses.

^aIn increments of 100.

RCT, randomized clinical trial.