Table 4

 Change in relational justice between Phase 1 and Phase 2 as a predictor of GHQ caseness by Phase 2 or by Phase 3

Change between Phase 1 and Phase 2GHQ caseness by Phase 2GHQ caseness by Phase 3
n (cases)Unadjusted OR (95% CI)Adjusted* OR (95% CI)n (cases)Unadjusted OR (95% CI)Adjusted* OR (95% CI)
*Adjusted for age, grade, and physical illness indicator (diabetes, diagnosed heart trouble, hypertension, ECG abnormalities, and/or respiratory illness) and relational justice at Phase 1.
Women 1684 (404)1464 (491)
Relational justice
    Favourable change383 (82)0.94 (0.71 to 1.26)0.74 (0.55 to 1.01)331 (114)1.18 (0.90 to 1.55)0.94 (0.70 to 1.26)
    No change964 (216)1.001.00844 (260)1.001.00
    Adverse change337 (106)1.59 (1.21 to 2.09)1.74 (1.31 to 2.30)289 (117)1.53 (1.16 to 2.01)1.68 (1.26 to 2.24)
Test for linear trendp < 0.0001p < 0.0001p = 0.02p < 0.0001
Men 4016 (785)3622 (1008)
Relational justice
    Favourable change808 (144)0.98 (0.79 to 1.20)0.75 (0.60 to 0.94)728 (199)1.06 (0.87 to 1.27)0.86 (0.70 to 1.05)
    No change2471 (449)1.001.002246 (590)1.001.00
    Adverse change737 (192)1.59 (1.31 to 1.93)1.81 (1.48 to 2.21)648 (219)1.43 (1.19 to 1.73)1.59 (1.31 to 1.93)
Test for linear trendp < 0.0001p < 0.0001p = 0.0003p < 0.0001