Skip to main content

Table 3 Application: results under the partial pooling random-effects model with different priors

From: Comparing Bayesian hierarchical meta-regression methods and evaluating the influence of priors for evaluations of surrogate endpoints on heterogeneous collections of clinical trials

Parameter (Subgroup, N Studies)

Diffuse Priors

Constrained Priors Set 1

Constrained Priors Set 2

Analysis: Cardiovascular Studies Left-Out (3 Subgroups in Model Fitting)

     Mean \(\beta\)

-0.30(-0.85,0.2)

-0.30(-0.68,0.02)

-0.30(-0.61,-0.06)

     Between-Subgroup SD of \(\beta\)

0.15(0.01,1.53)

0.13(0.01,0.77)

0.12(0.01,0.55)

     \(\beta _1\) (CKD, 28)

-0.25(-0.39,-0.13)

-0.25(-0.38,-0.13)

-0.25(-0.37,-0.13)

     \(\beta _2\) (Diabetes, 21)

-0.30(-0.48,-0.13)

-0.30(-0.48,-0.14)

-0.30(-0.48,-0.15)

     \(\beta _3\) (Glomerular Diseases, 10)

-0.35(-0.66,-0.16)

-0.35(-0.66,-0.17)

-0.35(-0.66,-0.18)

Analysis: Small Intervention Subgroups Left-Out (7 Subgroups in Model Fitting)

     Mean \(\beta\)

-0.41(-0.75,-0.13)

-0.41(-0.74,-0.13)

-0.4(-0.68,-0.18)

     Between-Subgroup SD of \(\beta\)

0.19(0.02,0.73)

0.18(0.02,0.70)

0.15(0.01,0.47)

     \(\beta _1\) (Antiplatelets, 3)

-0.39(-1.03,0.34)

-0.39(-1.01,0.33)

-0.39(-0.88,0.14)

     \(\beta _2\) (DPP-4, 3)

-0.40(-1.04,0.16)

-0.40(-1.01,0.14)

-0.39(-0.89,0.04)

     \(\beta _3\) (Immunosuppressants, 9)

-0.47(-0.92,-0.23)

-0.47(-0.94,-0.24)

-0.46(-0.87,-0.24)

     \(\beta _4\) (Modify Blood Pressure, 7)

-0.45(-0.84,-0.18)

-0.44(-0.83,-0.17)

-0.43(-0.78,-0.18)

     \(\beta _5\) (RASB vs CCB, 4)

-0.41(-0.81,-0.06)

-0.41(-0.79,-0.06)

-0.41(-0.75,-0.14)

     \(\beta _6\) (RASB vs Control, 21)

-0.50(-0.82,-0.25)

-0.49(-0.82,-0.24)

-0.47(-0.75,-0.23)

     \(\beta _7\) (SGLT-2, 4)

-0.25(-0.49,0.01)

-0.25(-0.49,0.00)

-0.27(-0.48,-0.07)

  1. Intervention subgroup names: DPP-4 Dipeptidyl peptidase 4 inhibitor, RASB Renin-angiotensin system blockers, CCB Calcium channel blockers, SGLT-2 Sodium-glucose Cotransporter-2 inhibitors, SD Standard deviation