Table 2.

Significant Independent Predictors (Adjusted) of the Relative Rate Ratio of RTW during the six month study

Main effects model only
Factor	P	Relative Rate Ratio (adjusted)	95% C.I
Should not work with current pain: agree	0.007	0.54	0.35-0.84
Co-morbid conditions: one or more	0.027	0.59	0.37-0.94
Blue collar worker	0.016	0.56	0.35-0.89
Orthopaedic injury: more than one region	0.030	0.49	0.26-0.93
Social functioning: high	0.009	1.89	1.17-3.07
Injury severity: ISS >9	0.036	0.36	0.39-0.97
Age (cont)	0.017	0.17	0.96-0.99
Self employed	0.015	2.31	1.18-4.53
Main effects plus interactions
Factor	P	Relative Rate Ratio (adjusted)	95% C.I
Should not work with current pain: agree	0.002	0.49	0.31-0.77
High Initial pain intensity	0.008	0.47	0.27-0.82
Co-morbid conditions: one or more	0.016	0.56	0.35-0.89
Blue collar worker	0.008	0.52	0.32-0.84
Initial need for surgery	0.034	0.61	0.39-0.96
Orthopaedic injury: more than one region	0.007	0.41	0.21-0.78
Injury severity: ISS>9	0.045	0.63	0.39-0.99
Age (cont)	0.037	0.98	0.96-0.99
Education*high initial pain level	0.006	4.50	1.55-13.09
Relative rate ratio for interaction	0.56 × 0.47 × 4.50 = 1.18
Social functioning* receipt of compensation	0.010	3.58	1.35-9.47
Relative rate ratio for interaction	1.22 × 0.59 × 3.58 = 2.58

Factors included in the final main effects model. Shouldn't work, age, education, injury severity, compensation, blue collar worker, co-morbidities, orthopaedic injury to more than one region, high initial pain, recovery beliefs, self employment. Confounding Factors tested but did not remain in the final model: gender, prior pain