Skip to main content

Table 3 Predictive models for improvement

From: Predicting the evolution of neck pain episodes in routine clinical practice

a. Predictive model for the improvement of neck pain (n = 2372).a,b
VariablesOdds ratiopFrequency in bootstrapping validationc (100)
(I.C. 95%)
Being treated with neuro-reflexotherapy9.90 (6.81; 14.38)< 0.001100
Pain being chronic (≥ 90 days)0.53 (0.41; 0.70)< 0.00198
Baseline intensity of arm pain (VAS)d0.93 (0.90; 0.97)< 0.00195
Employment status (ref. working)
 Non worker0.87 (0.69; 1.10)0.25994
 Receiving financial compensation for neck pain0.48 (0.33; 0.69)< 0.00194
Signs of disc degeneration on imaging0.77 (0.62; 0.95)0.01768
Clinical diagnosis (ref. nonspecific pain)
 Spinal stenosis0.78 (0.39; 1.56)0.48263
 Disc herniation/protusion0.63 (0.49; 0.81)< 0.00163
Female0.77 (0.61; 0.97)0.03059
Baseline intensity of neck pain (VAS)d1.06 (1.00; 1.13)0.04138
Constant0.84 (0.60; 1.18)0.325
b. Predictive model for the improvement of pain referred down into the arm (n = 1938).e,f
VariablesOdds ratiopFrequency in bootstrapping validationg (100)
(I.C. 95%)
Being treated with neuro-reflexotherapy16.96 (10.90; 26.37)< 0.001100
Baseline intensity of arm pain (VAS) h1.17 (1.10; 1.24)< 0.00196
Pain being chronic (≥ 90 days)0.57 (0.43; 0.75)< 0.00184
Signs of disc degeneration on imaging0.68 (0.54; 0.85)0.00183
Baseline intensity of neck pain (VAS)d0.91 (0.85; 0.98)0.01073
Clinical diagnosis (ref. nonspecific pain)
 Spinal stenosis0.57 (0.52; 0.86)0.08856
 Disc herniation/protrusion0.67 (0.52; 0.86)0.00256
Constant0.31 (0.19; 1.24)< 0.001
c. Predictive model for the improvement for disability (n = 983).i,j 
VariablesOdds ratioPFrequency in bootstrapping validationk (100)
(I.C. 95%)
Baseline intensity of arm pain (VAS)l0.89 (0.85; 0.93)< 0.00199
Being treated with neuro-reflexotherapy2.92 (1.90; 4.49)< 0.00197
Employment status (ref. working)
 Non worker0.69 (0.49; 0.97)0.03190
 Receiving financial compensation for neck pain0.45 (0.28; 0.73)0.00190
Baseline disability (NDI)m1.02 (1.01; 1.02)0.00284
Signs of facet joint degeneration on imaging0.60 (0.39; 0.93)0.02373
Pain being chronic (≥ 90 days)0.65 (0.46; 0.91)0.01256
Constant0.89 (0.59; 1.34)0.589
  1. aThe number of patients who reported some degree of pain referred down to the arm (AP) at baseline (VAS > 0), was 2961,6 had baseline scores below the cut-off for considering potential improvements as “clinically relevant”, 583 had missing data at the baseline or the follow-up assessment, and 2372 were included in the model
  2. bAUC = 0.718 (95%CI; 0.694–0.742). Hosmer-Lemeshow: chi2 = 0.383
  3. cOverfitting = 0.020. Shrinkage factor = 0.906
  4. dVAS: Visual Analog Scale (range from better to worse; 0–10)
  5. eThe number of patients who reported some degree of neck pain (VAS > 0) at baseline, was 2961, 18 had baseline scores below the cut-off for considering potential improvements as “clinically relevant”, 238 had missing data at the baseline or the follow-up assessment, and 2372 were included in the model
  6. fAUC = 0.717 (95%CI; 0.691–0.742). Hosmer-Lemeshow: chi2 = 0.369
  7. gOverfitting = 0.030. Shrinkage factor = 0.882
  8. hVAS: Visual Analog Scale (range from better to worse; 0–10)
  9. iThe number of patients who reported some degree of disability at baseline (NDI > 0), was 1500,49 had baseline scores below the cut-off for considering potential improvements as “clinically relevant”, 468 had missing data at the baseline or the follow-up assessment, and 983 were included in the model
  10. jAUC = 0.677 (95%CI; 0.644–0.711). Hosmer-Lemeshow: chi2 = 0.128
  11. kOverfitting = 0.037. Shrinkage factor = 0.787
  12. lVAS: Visual Analog Scale (range from better to worse; 0–10)
  13. mScore on the Neck Disability Index (range from better to worse, 0–100)