Skip to main content

Table 4 The method of spatial analysis: effect of the socioeconomic status on the spatial structure of cancer incidence [regression coefficient with 95 % confidence intervals for Poisson models or credible intervals for HBSM]

From: Impact of socioeconomic inequalities on geographic disparities in cancer incidence: comparison of methods for spatial disease mapping

 

SpODT

SATSCAN

HBSM

α + βχι

α + Vi + βχι

α + Ui + βχι

α + Ui + Vi + βχι

Lung

---

---

0.059 [0.050; 0.069]

0.067 [0.054; 0.080]a

0.065 [0.053; 0.077]

0.068 [0.055; 0.080]

Prostate

5 clusters

---

-0.017 [-0.023; -0.011]

-0.010 [-0.021; -0.0005]

-0.012 [-0.022; -0.001]a

-0.011 [-0.021; -0.0009]

Colon rectum

---

---

-0.001 [-0.011; 0.009]a

0.0003 [-0.011; 0.012]

0.0004 [-0.010; 0.011]

0.0007 [-0.011; 0.013]

Bladder

1 cluster

---

0.023 [0.007; 0.039]

0.023 [0.004; 0.043]

0.022 [0.003; 0.041]a

0.024 [0.003; 0.045]

  1. aBest Bayesian model in term of DIC, βχ ι additional covariate - \( \mathbf{\mathcal{U}} \): Autocorrelation - \( \mathbf{\mathcal{V}} \): heterogeneity \( \mathbf{\mathcal{X}} \) Townsend index components