Skip to main content

Table 3 Summary of the four PEDIS risk models. To support a more straightforward interpretation of the logistic regression model coefficients, we present them as odds ratios (OR) in this table; the logistic regression coefficients can be interpreted as odds ratios when exponentiated. The OR is the exponentiated median of the posterior coefficient distribution. Furthermore, the 95% HDI of the OR is presented. HDI is derived from the posterior distribution. Any value within the interval has a higher density than the values outside the HDI. The total mass of values inside the HDI is 95%

From: Predicting the amputation risk for patients with diabetic foot ulceration – a Bayesian decision support tool

Risk Factors (Predictors) Any Amputation Major Amputation
Non-Informed Model Informed Model Non-Informed Model Informed Model
Beta-Coefficients Odds Ratios Beta-Coefficients Odds Ratios Beta-Coefficients Odds Ratios Beta-Coefficients Odds Ratios
Perfusion 0.688 [0.264–1.152] 1.990 [1.302–3.164] 0.703 [0.352–1.116] 2.020 [1.422–3.052] 0.471 [−0.017–1.069] 1.601 [0.983–2.913] 0.586 [0.170–1.070] 1.798 [1.185–2.914]
Extend 1.484 [0.617–2.376] 4.411 [1.853–10.762] 1.283 [0.562–2.233] 3.609 [1.754–9.326] 0.985 [−0.079–2.665] 2.678 0.924–14.364] 0.799 [0.061–2.272] 2.222 [1.063–9.702]
Depth 0.665 [0.050–1.418] 1.945 [1.051–4.128] 0.656 [0.168–1.302] 1.927 [1.183–3.677] 0.932 [− 0.088–2.275] 2.540 [0.916–9.726] 0.694 [0.069–1.893] 2.001 [1.071–6.642]
Infection − 0.112 [− 0.483–0.266] 0.894 [0.617–1.305] −0.021 [− 0.439–0.368] 0.979 [0.644–1.446] 0.155 [− 0.296–0.635] 1.167 [0.744–1.888] 0.368 [− 0.108–0.757] 1.445 [0.898–2.131]
Sensation 0.037 [− 0.726–1.077] 1.037 [0.484–2.937] 0.516 [− 0.303–1.223] 1.675 [0.738–3.397] 0.076 [−0.861–1.782] 1.079 [0.423–5.943] 0.553 [− 0.404–1.639] 1.738 [0.668–5.149]
AUC Value 0.793 0.790 0.765 0.790
AUC HDI [0.778–0.801] [0.774–0.802] [0.725–0.779] [0.774–0.802]