Skip to main content
. Author manuscript; available in PMC: 2018 Jul 23.
Published in final edited form as: Stat Methods Med Res. 2016 Sep 30;26(1):414–436. doi: 10.1177/0962280214548748

Table 3.

SBWR results, for the fixed effects and top tumour markers associated breast cancer survival in SEARCH.

HR 95% CrIa MPPIb Imputed
Fixed parameters
    Intercept −7.32 (−8.07, −6.63)   –
    log(beta) Hyperprior SD (σβ)   0.24 (0.12, 0.71)   –
    Weibull scale   1.74 (1.54, 1.96)   –
    Number positive nodesc
    (0, 1, 2–4, 5–9, 10+)
  1.61 (1.45, 1.79)   8.4%
    Tumour size, mmc
    (<10, 10–19, 20–29, 30–49, 50+)
  1.26 (1.09, 1.45)   3.8%
    Tumour gradec
    (Low, Intermediate, High)
  1.47 (1.14, 1.89) 10.5%
    Morphology – Ductular   0%
    Morphology – Lobular   1.55 (1.10, 2.16)   –
    Morphology – Other   1.06 (0.64, 1.68)   –
    HER2   1.47 (0.97, 2.18) 10.8%
    Detection by screening   0.79 (0.55, 1.11)   6.1%
    Hormone therapy   2.20 (1.38, 3.86) <0.01%
    Study entry delay, years   0.88 (0.79, 0.98)   0%
‘Top’ tumour markers
    PGRP   0.86 (0.80, 0.93) 0.92   5.2%
    PDCD4O2   0.75 (0.62, 0.89) 0.84 43.2%
    AURKAP   1.30 (1.11, 1.51) 0.68 31.0%
    CD8P   0.92 (0.85, 0.98) 0.30 32.9%
    GATA3I   0.80 (0.68, 0.94) 0.30 41.6%
    BCL2P   0.94 (0.90, 0.98) 0.26   9.0%
a

Credible Intervals (CrI): For the tumour markers, these were calculated conditional on inclusion in the model.

b

Marginal Posterior Probabilitiy of Inclusion in the model – may be interpreted as the posterior probability an association exists with survival, adjusted for all other covariates in the model.

c

Modelled as ordinal continuous.