Table 6

Parameter estimates from multivariable meta-regression model using different estimation techniques for between-study heterogeneity (Bayesian priors: Embedded Image, Embedded Image)

MethodVariableEstimate
(SE/MC error)
95% Cl
REMLEmbedded Image(Intercept)42.9 (5.53)*31.8 to 54.0
Embedded Image(Blinded induction of ischaemia: true)−11.5 (7.03)−25.5 to 2.60
Embedded Image(Route of delivery: intracerebroventricular)10.1 (6.80)−3.55 to 23.7
Embedded Image(Route of delivery: intravenous)−2.85 (6.10)−15.1 to 9.37
Embedded Image(Route of delivery: other)−24.4 (6.66)*−37.8 to −11.1
Embedded Image(Time to outcome measure)−0.03 (0.02)−0.07 to 0.01
Embedded Image(Dose)0.01 (0.02)−0.03 to 0.05
DLEmbedded Image(Intercept)43.6 (5.47)*32.6 to 54.5
Embedded Image(Blinded induction of ischaemia: true)−10.9 (6.92)−24.7 to 2.97
Embedded Image(Route of delivery: intracerebroventricular)9.28 (6.69)−4.12 to 22.7
Embedded Image(Route of delivery: intravenous)−3.71 (6.04)−15.8 to 8.39
Embedded Image(Route of delivery: other)−24.7 (6.84)*−38.4 to −11.0
Embedded Image(Time to outcome measure)−0.03 (0.02)−0.07 to 0.01
Embedded Image(Dose)0.01 (0.02)−0.03 to 0.04
BayesianEmbedded Image(Intercept)39.0 (0.22)*27.4 to 50.2
Embedded Image(Blinded induction of ischaemia: true)−13.7 (0.34)−31.1 to 3.83
Embedded Image(Route of delivery: intracerebroventricular)12.6 (0.23)−1.67 to 27.2
Embedded Image(Route of delivery: intravenous)0.05 (0.22)−13.1 to 12.9
Embedded Image(Route of delivery: other)−20.9 (0.21)*−35.3 to −6.25
Embedded Image(Time to outcome measure)−0.003 (0.001)−0.05 to 0.05
Embedded Image(Dose)0.01 (0.001)−0.03 to 0.06
  • *Represents significant effect of a predictor at an alpha level of 0.05.

  • DL, DerSimonian and Laird; MC, Monte Carlo; REML, restricted maximum likelihood.