Table 4

Determinants of all-cause mortality: results from multilevel parametric frequentist versus Bayesian Weibull proportional hazard regression models (HRs with 95% CI and credible interval, respectively)

Model 3.1Model 3.2
Frequentist Bayesian
HR95% CIHR95% CI
Individual level
 Generalised trust (group-mean centred) 0.92 0.88 to 0.97 0.94 0.91 to 0.98
 Age 1.07 1.07 to 1.08 1.07 1.07 to 1.08
Race
 Black (ref: white) 1.24 1.13 to 1.36 1.20 1.13 to 1.28
 Other race1.03 0.91 to 1.16 1.05 0.97 to 1.13
Sex
 Female (ref: male) 0.69 0.65 to 0.73 0.69 0.67 to 0.72
Degree
 Less than high school
 High school (ref:)
1.21 1.13 to 1.29 1.21 1.17 to 1.26
 Junior college 0.84 0.73 to 0.97 0.89 0.81 to 0.97
 Bachelor 0.84 0.77 to 0.91 0.89 0.85 to 0.92
 Graduate 0.87 0.78 to 0.97 0.82 0.77 to 0.88
Marital status
 Married (ref.)
  Widowed 1.22 1.14 to 1.31 1.23 1.15 to 1.32
  Divorced0.99 0.92 to 1.08 0.99 0.95 to 1.04
  Separated1.05 0.97 to 1.26 1.11 1.00 to 1.10
  Never married 1.32 1.21 to 1.44 1.26 1.21 to 1.33
Income
 Household income (group-mean centred) 0.94 0.93 to 0.96 0.94 0.93 to 0.95
Contextual level
 Generalised trust (z-standardised) 0.96 0.93 to 0.99* 0.96 0.94 to 0.98
 Income inequality (GINI-coefficient, z-standardised)0.99 0.96 to 1.03 1.00 0.98 to 1.02
 Mean income (z-standardised)0.99 0.95 to 1.02 0.99 0.96 to 1.01
 Ln(p)0.512
 Residual variance (level 1)0.593
 Residual variance (level 2)0.004
 Level 2 variance explained73%
  • Notes: statistically significant coefficients (p<0.05) boldfaced. Residual variance, level 1=0.921, residual variance, level 2= 0.015. Data source: GSS–NDI: n=25 270; number of failures=6424. Data weighted with WTSSALL (model 3.1).

  • *Full CI interval: 0.9252366- 0.999961.

  • NDI, National Death Index; GSS, General Social Survey.