Background: Previous studies of organisational justice and mental health have mostly examined women and have not examined the effect of change in justice. Aim: To examine effects of change in the treatment of employees by supervisors (the relational component of organisational justice) on minor psychiatric morbidity, using a cohort with a large proportion of men. Methods: Data are from the Whitehall II study, a prospective cohort of 10 308 white-collar British civil servants (3143 women and 6895 men, aged 35-55 at baseline) (Phase 1, 1985-88). Employment grade, relational justice, job demands, job control, social support at work, effort-reward imbalance, physical illness, and psychiatric morbidity were measured at baseline. Relational justice was assessed again at Phase 2 (1989-90). The outcome was cases of psychiatric morbidity by Phases 2 and 3 (1991-93) among participants case-free at baseline. Results: In analyses adjusted for age, grade, and baseline physical illness, women and men exposed to low relational justice at Phase 1 were at higher risk of psychiatric morbidity by Phases 2 and 3. Adjustment for other psychosocial work characteristics, particularly social support and effort-reward imbalance, partially attenuated these associations. A favourable change in justice between Phase 1 and Phase 2 reduced the immediate risk (Phase 2) of psychiatric morbidity, while an adverse change increased the immediate and longer term risk (Phase 3). Conclusion: This study shows that unfair treatment by supervisors increases risk of poor mental health. It appears that the employers' duty to ensure that employees are treated fairly at work also has benefits for health.