To investigate whether self-reported exposure to negative acts in the workplace (bullying and threats of violence) predicted turnover in three occupational groups (human service and sales workers, office workers and manual workers).
Survey data on 2766 respondents were combined with data from a national labour force register to assess turnover. Mixed effects logistic regression analysis was used to examine the association between self-reported exposure to negative acts at baseline and risk of turnover after a 1-year follow-up.
We found no significant associations between exposure to negative acts (bullying and threats of violence) and risk of turnover. When participants were stratified by occupational group and analyses were adjusted for age, gender, tenure and psychosocial working conditions, we found that exposure to bullying predicted risk of turnover in office workers (OR 2.03, 95 % CI 1.05-3.90), but neither in human service and sales workers, nor in manual workers. The association in office workers lost statistical significance when additionally adjusted for depressive symptoms (OR 1.77, 95 % CI 0.90-3.49). However, in a sensitivity analysis in which we used a 2-year (instead of a 1-year) follow-up period the association between bullying and turnover remained statistically significant in office workers even after adjusting for depressive symptoms (OR 2.10, 95 % CI 1.17-3.76). We found no statistically significant associations between threats of violence and risk of turnover in the stratified analyses.
Exposure to bullying predicted risk of turnover among office workers but not among human service and sales workers and among manual workers. Threats of violence were not associated with turnover in any occupational group.