Skip to main content

Table 6 Results of four network meta-analyses: based on only aggregate data; combining IPD and aggregate data with no covariate adjustments; combining IPD and aggregate data with covariate adjustments for individual patient AGE, baseline STATUS and baseline PVR at within-study level; using the covariate adjusted IPD and aggregate data model with the observational studies down-weighted by inflating their standard errors by a factor of 10; using the covariate adjusted IPD and aggregate data model with constructed control arms for observational studies with an assumed deterioration of 25 meters in 6MWD

From: Network meta-analysis combining individual patient and aggregate data from a mixture of study designs with an application to pulmonary arterial hypertension

Model

M1

M2

M3

S1

S2

Model description

Aggregate data only

IPD and aggregate with no covariate adjustments

IPD and aggregate with covariate adjustments

Down-weighted observational studies

Constructed control arms

α i , Jacobs 2009

4.77 (-18.60,29.12)

4.55 (-19.07, 29.45)

4.39 (-18.77, 28.59)

4.07 (-17.16, 26.84)

−1.10 (-22.54, 17.77)

α i , (imatinib to ERA + PDE5i)

4.27 (-14.13, 22.59)

3.63 (-10.57, 17.78)

3.50 (-10.59, 17.69)

3.55 (-9.78, 17.12)

0.93 (-11.93, 14.71)

α (mean of α i s)

4.78 (-4.77, 14.60)

4.39 (-5.14, 14.80)

4.30 (-4.73, 14.17)

3.99 (-4.87, 13.68)

0.02 (-8.74, 8.78)

Ï€ 1 (AGE)

NA

NA

−0.90 (-1.52, -0.27)

−0.90 (-1.52, -0.27)

−0.89 (-1.53, -0.26)

Ï€ 3 (STATUS)

NA

NA

−11.89 (-28.20, 4.44)

−11.92 (-28.29, 4.36)

−12.02 (-28.64, 4.60)

Ï€ 5 (PVR)

NA

NA

−0.05 (-0.70, -0.02)

−0.05 (-0.07, -0.02)

−0.05 (-0.07, -0.02)

β 6 (Pr to ERA + PDE5i)

35.98 (-44.15, 114.70)

35.16 (-43.81, 113.80)

34.65 (-42.18, 113.20)

22.64 (-168.30, 217.00)

40.65 (-37.09, 117.80)

β 8 (imatinib to ERA + PDE5i)

39.09 (2.72, 75.80)

39.91 (11.31, 68.36)

40.06 (13.38, 67.67)

39.40 (13.83, 66.03)

42.62 (15.18, 69.79)

(β 8  − β 6 ) imatinib v Pr to ERA + PDE5i

3.11 (-83.70, 89.27)

4.75 (-77.88, 87.10)

5.41 (-76.65, 85.24)

16.76 (-179.26, 209.80)

1.97 (-78.62, 83.53)

  1. Results sampled from 250000 iterations following a burn-in of 100000.