PMC full text: | Published online 2018 Apr 5. doi: 10.1371/journal.pone.0191852
|
Table 3
Item variable | Model (i) | Model (ii) | |||
---|---|---|---|---|---|
Beta | OR (95% CI) | Beta | OR (95% CI) | ||
Subjective sensory changes | 0.98 | 2.66 (1.20, 5.90) | NS | NS | |
Below knee pain | 1.83 | 6.25 (2.80, 13.94) | 0.76 | 2.13 (1.19, 3.83) | |
Leg pain worse than back pain | 1.52 | 4.55 (1.89, 10.99) | 1.08 | 2.94 (1.77, 4.89) | |
Leg pain intensity | NS | NS | 0.14 | 1.15 (1.03, 1.29) | |
Positive cough / sneeze | NS | NS | 0.92 | 2.50 (1.34, 4.65) | |
Neural Tension tests | 3.07 | 21.63 (9.00,51.97) | 0.56 | 1.76 (1.03, 3.00) | |
Neurological deficit | 2.14 | 8.50 (3.80,19.01) | 1.04 | 2.81 (1.69, 4.69) | |
Intercept | -3.25 | -2.98 | |||
AUC | 0.95 (0.93, 0.98) | 0.82 (0.78, 0.86) |
Model (i): Confidence ≥80% sciatica clinical diagnosis
Model (ii): Confidence ≥80% sciatica clinical diagnosis plus confirmatory magnetic resonance imaging (MRI) findings
The predicted probability of sciatica can be calculated using the following formulae
Model (i): Probability (sciatica+) = 1/ 1+exp -[-3.25 + (subjective sensory changes x 0.98) + (below knee pain x 1.83) + (leg pain worse than back pain x 1.52) + (neural tension x 3.07) + (neurological deficit x 2.14)].
Model (ii): Probability (sciatica+) = 1/ 1+exp -[-2.98 + (below knee pain x0.76) + (leg pain worse than back pain x 1.08) + (intensity leg pain x 0.14) + (positive cough/sneeze x 0.92) + (neural tension x 0.56) + (neurological deficit x 1.04)]