TABLE 1.
Comparison between Laslett M et al51 and van der Wurff et al20 studies of the validity of multiples of positive pain provocation SIJ tests.
| Number of positive provocation SIJ tests | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 1 or more | 2 or more | 3 or more | 4 or more | 5 or more | ||||||
| Diagnostic accuracy statistic | ML | PvW | ML | PvW | ML | PvW | ML | PvW | ML | PvW |
| Sensitivity % | 100 | 100 | 93 | 93 | 91 | 85 | 60 | 26 | 27 | 0 |
| Specificity % | 44 | 42 | 66 | 58 | 78 | 79 | 81 | 82 | 88 | 100 |
| Positive LR | 1.8 | 1.7 | 2.7 | 2.2 | 4.3 | 4.0 | 3.2 | 1.4 | 2.1 | 0 |
| Negative LR | 0.0 | 0.0 | 0.10 | 0.13 | 0.08 | 0.19 | 0.49 | 0.91 | 0.84 | 1.00 |
- LR = likelihood ratio, ML = Laslett M et al 2005, PvW = van der Wurf et al 2006
- The shaded cells represent the optimal number of positive SIJ provocation tests producing the highest positive likelihood ratio, i.e., 3 or more.
- The tests included in this study are distraction, compression, thigh thrust, Gaenslen's test, sacral thrust, and Patrick's FABER test.