Final multivariate logistic model. Deviance values and p values are for sequential tests, adding each variable to a model containing all those above it. Odds ratios <1 are associated with a decreasing probability of survival. Odds ratios for continuous variables are for a difference of 5 units* or 1 unit† as indicated. Age is fitted as a function with two linear parts; there is no effect on the odds of survival until the age of 50, but after this point the odds of survival decrease by a factor of 0.545 for each 5 year increase in age
| Variable | p Value | Categories | Odds ratio | (95% CI) |
|---|---|---|---|---|
| Age | <0.0001 | Piecewise linear | 0.545* | (0.43–0.69) |
| GCS total | <0.0001 | Linear | 1.31† | (1.12–1.53) |
| ISS score | <0.0001 | Linear | 0.737* | (0.60–0.91) |
| Pupils | 0.004 | Both reacting | 1 | — |
| Only one reacting | 0.599 | (0.18–2.00) | ||
| Neither reacting | 0.168 | (0.06–0.50) | ||
| CT | 0.004 | Visible haematoma | 1 | — |
| No visible haematoma | 3.53 | (1.43–8.73) |