. 2017 Oct 19;187(5):1113–1122. doi: 10.1093/aje/kwx344

Table 5.

Estimated Ratios of Odds Ratios and Between–Meta-Analysis Heterogeneity in Mean Bias Associated With Risk-of-Bias Judgements in Randomized Trials and Meta-Analyses, by Type of Outcome Measure (Multivariable Analyses (Model B)), in the ROBES Study^a

Risk-of-Bias Domain and Outcome	No. of MAs or RTs Contributing to Analysis		Average Bias		No. of MAs Contributing to κ Estimation	Within-MA Heterogeneity		Between-MA Heterogeneity
Risk-of-Bias Domain and Outcome	MAs	RTs	ROR	95% CrI	No. of MAs Contributing to κ Estimation	κ	95% CrI	φ	95% CrI
Sequence generation: high/unclear risk of bias vs. low risk of bias
All outcomes	189	2,158	0.95	0.88, 1.03	142	0.08	0.02, 0.18	0.11	0.03, 0.22
Mortality	34	363	0.92	0.75, 1.18	27	0.14	0.02, 0.36	0.14	0.03, 0.42
Other objective/semiobjective outcome	47	523	1.06	0.90, 1.28	38	0.14	0.03, 0.33	0.20	0.04, 0.44
Subjective outcome/mixture^b	108	1,272	0.94	0.84, 1.04	77	0.08	0.02, 0.18	0.11	0.02, 0.24
Allocation concealment: high/unclear risk of bias vs. low risk of bias
All outcomes	188	2,121	0.96	0.88, 1.03	139	0.06	0.01, 0.15	0.07	0.02, 0.16
Mortality	35	358	0.92	0.74, 1.13	27	0.11	0.03, 0.29	0.15	0.03, 0.42
Other objective/semiobjective outcome	49	524	0.94	0.81, 1.08	40	0.07	0.01, 0.18	0.09	0.02, 0.25
Subjective outcome/mixture	104	1,239	0.95	0.86, 1.07	72	0.10	0.02, 0.23	0.08	0.02, 0.20
Blinding: high/unclear risk of bias vs. low risk of bias
All outcomes	144	1,678	0.88	0.81, 0.94	105	0.10	0.02, 0.22	0.12	0.03, 0.23
Mortality	31	327	0.87	0.73, 1.03	25	0.10	0.02, 0.26	0.10	0.02, 0.28
Other objective/semiobjective outcome	32	334	0.95	0.79, 1.12	24	0.09	0.02, 0.24	0.10	0.02, 0.34
Subjective outcome/mixture	81	1,017	0.84	0.75, 0.95	56	0.17	0.04, 0.33	0.19	0.05, 0.35
Incomplete outcome data: high/unclear risk of bias vs. low risk of bias
All outcomes	167	1,956	1.01	0.94, 1.09	112	0.07	0.01, 0.16	0.07	0.02, 0.16
Mortality	29	303	0.99	0.82, 1.18	19	0.11	0.02, 0.31	0.10	0.02, 0.30
Other objective/semiobjective outcome	43	471	1.04	0.90, 1.21	28	0.11	0.02, 0.30	0.09	0.02, 0.26
Subjective outcome/mixture	95	1,182	1.00	0.90, 1.12	65	0.07	0.01, 0.17	0.11	0.03, 0.27

Abbreviations: CrI, credible interval; MA, meta-analysis; ROBES, Risk of Bias in Evidence Synthesis; ROR, ratio of odds ratios; RT, randomized trial.

^a For a graphical representation of these results, see Web Figure 4.

^b “Mixture” refers to meta-analyses in which some trials had subjective measures and some had objective measures of the same outcome (e.g., self-reports and laboratory measures of smoking cessation).