Reduced model A: the variables at first consultation predicting surgery (multiple logistic regression analysis with eventual surgery as dependent variable)
| Baseline predictors of surgery | Crude odds ratio | 95% CI | Adjusted odds ratio | 95% CI |
| History/questionnaire: | ||||
| Mentally demanding job | 3.8 | 1.7–8.6 | 4.0 | 1.5–10.3 |
| Sudden onset | 0.28 | 0.1–0.7 | 0.22 | 0.07–0.68 |
| More pain on coughing/sneezing/straining | 5.3 | 2.0–13.6 | 4.3 | 1.5–12.4 |
| Difficulty putting on socks/stockings | 1.6 | 0.7–1.3 | 2.3 | 0.86–6.0 |
| VAS of pain intensity in the leg | 1.03 | 1.0–1.05 | 1.033 | 1.004–1.062 |
| Physical examination: | ||||
| Reversed straight leg raising | 2.5 | 1.1–5.8 | 3.2 | 1.2–8.9 |
| Exponential value of the intercept | 3.6*10−3 | 2.9*10−4–4.4*10−2 | ||
VAS = visual analogue scale; crude odds ratio=the association between two variables calculated as the odds of a finding being present in the case of eventual surgery divided by the odds of a finding being absent in the absence of eventual surgery (with odds being defined as the probability that the event will occur divided by the probability that the event will not occur). In diagnostic research this odds ratio is often referred to as the likelihood ratio. Logistic regression analysis is an alternative method of calculating odds ratios; adjusted odds ratio=the association between two findings adjusted for the confounding effect of other findings. Logistic regression has the additional benefit that in a multivariate analysis the presence or absence of a finding can be related to more than one factor. The following paradigma holds for the calculation of the probability of surgery in the next 6 months:
Probability (X)= 1/ (1 + e-Y)
Y=α+ln(4)*X1+ln(0.22)*X2+ln(4.3)*X3+ln(2.3)*X4 + ln(1.033)*X5+ln(3.2)*X6
α=intercept and X1 through X6 are the six variables in the order of the table.