Skip to main content

Table 6 Calculation of the linear predictor and probabilitya

From: The Amsterdam wrist rules: the multicenter prospective derivation and external validation of a clinical decision rule for the use of radiography in acute wrist trauma

Linear predictor ALL WRIST FRACTURES

 0.0309* age + 0.5862 + (if male) + 1.1486 * (if swelling wrist present) + 0.5757 (if swelling anatomical snuff box is present) + 1.7123 *(if visible deformation present) + 0.7029 * (if distal radius tender to palpation) + 0.4963 *(if pain on radial deviation) + -0.1793 * (if on axial compression thumb) - 3.616

Linear predictor DISTAL RADIUS FRACTURES

 0.0341* age + 1.7298 * (if swelling of wrist present) + 1.6462 *(if visible deformation present) + 1.8117 * (if distal radius tender to palpation) + 0.4228 *(if pain on palmar flexion) + 0.6567 * (if pain on supination) – 0.2941 (if pain on ulnar deviation) + 0.5949 * (if pain during radioulnar ballottement test) - 6.0202

Formula to calculate probability of a fracture based on final model

 1/ (1 + EXP(-Linear Predictor))

  1. * signifies a multiplication sign
  2. All individual parameters add to the probability of a fracture
  3. aCoefficients were derived from a fit of the model on both cohort combined (N = 882) and corrected for optimism by bootstrapping (N = 500 replications)