Skip to main content
. 2020 Sep 2;370:m2984. doi: 10.1136/bmj.m2984

Table 3.

Results from Cox regression analyses on workplace sexual harassment stratified by occupational position, presented as hazard ratios (HR) and 95% confidence intervals with and without adjustment for covariates

Supervisor* Subordinate*
No with valid data No of cases HR (95% CI) No with valid data No of cases HR (95% CI)
Suicide
Model 0 25 544 35 1.60 (0.38 to 6.70) 59 375 90 2.46 (1.23 to 4.94)
Model 1 25 388 35 1.94 (0.45 to 8.42) 58 564 89 3.18 (1.56 to 6.47)
Model 2§ 25 089 33 1.04 (0.14 to 7.84) 57 682 88 3.06 (1.50 to 6.25)
Model 3 25 035 33 0.95 (0.12 to 7.30) 57 548 88 3.05 (1.47 to 6.33)
Suicide attempts
Model 0 25 365 221 1.77 (1.08 to 2.88) 58 908 593 1.43 (1.04 to 1.96)
Model 1 25 210 219 1.98 (1.20 to 3.28) 58 107 578 1.42 (1.03 to 1.97)
Model 2§ 24 912 217 1.97 (1.19 to 3.26) 57 233 567 1.38 (0.99 to 1.92)
Model 3 24 859 217 1.98 (1.19 to 3.31) 57 100 566 1.38 (0.99 to 1.92)
*

Participants with supervisory duties were regarded as supervisors, those reporting no supervisory duties were regarded as subordinates.

Unadjusted analyses.

Adjusted for sex, birth country, family situation, education, and income at baseline.

§

Adjusted for sex, birth country, family situation, education, income, and poor mental health at baseline

Adjusted for sex, birth country, family situation, education, income, demands, control, social support at work, workplace bullying, and poor mental health at baseline.