Intended for healthcare professionals


Sexual harassment and suicide

BMJ 2020; 370 doi: (Published 02 September 2020) Cite this as: BMJ 2020;370:m3330

Linked Research

Work related sexual harassment and risk of suicide and suicide attempts

  1. Urmimala Sarkar, professor of medicine1,
  2. Shirin Hemmat, chief resident—internal medicine1,
  3. Eleni Linos, professor of dermatology and epidemiology2
  1. 1University of California, San Francisco, USA
  2. 2Stanford University, Stanford, CA, USA
  1. Correspondence to: E Linos linos{at}

Sexual harassment is a serious public health problem and workplace hazard

The high prevalence and significant adverse effects of workplace sexual harassment have become increasingly clear in the wake of the #MeToo movement. In the linked prospective study, Magnusson Hanson and colleagues convincingly show excess rates of suicide attempts and deaths among people reporting workplace sexual harassment (doi:10.1136/bmj.m2984).1 These results persist after extensive adjustment, including sex, country of birth, family type, education, income, baseline mental health, and working conditions. Importantly, neither the incidence nor the association with suicidality are limited to women.

Although we agree with the authors in exercising caution in interpreting their findings, this large prospective, population based cohort design represents the most rigorous approach to examine this question, and we are unlikely to generate more definitive evidence of a causal relation between workplace sexual harassment and suicide.

The authors’ findings are consistent with previous studies of sexual harassment. Although estimates vary, sexual harassment is highly prevalent.2 Other studies have identified serious physical and mental health sequelae of sexual harassment.34 Sexual harassment also leads many to leave their employment, leading to substantial financial stress. This financial stress resulting from sexual harassment is comparable with financial stress from serious injury or illness.5 Based on these findings, experts have deemed workplace sexual harassment an occupational health risk.6 This new study adds weight to this argument. Indeed, the observed hazard ratio for suicide death of 2.82 (95% confidence interval 1.49 to 5.34) is similar to the risk of death from cancer attributable to occupational asbestos exposure.7 In the context of existing evidence, the study underscores the urgent need to consider workplace sexual harassment both an occupational hazard and a substantial public health problem.

Inadequate response

The current occupational response to sexual harassment is sobering. One recent large scale study examined 805 companies over 32 years to understand the effect of sexual harassment grievance procedures and training. They found that sexual harassment grievance procedures were more likely to cause retaliation against the person being harassed without satisfying the complaint against the harasser and were also associated with fewer women in management.8 Disappointingly, sexual harassment training, in which those receiving training are depicted as potential harassers, can actually lead to more hostility towards victims of sexual harassment and women in the workplace more broadly.9 Previous research also showed that reporting sexual harassment did not improve, and sometimes worsened, job, psychological, and health outcomes.10

Given that the most widespread approaches to prevention (sexual harassment training) and to mitigation (reporting or grievance procedures) have been shown to do more harm than good, new ways to prevent and deal with workplace sexual harassment are urgently needed.

One evidence based approach to prevention is bystander training, pioneered in the US. Bystander intervention programmes focus on the assumption that all employees are allies in tackling harassment rather than casting participants as potential perpetrators. The programmes have been successful, with participants more likely to report intervention in real life situations than those taking traditional sexual harassment training programmes. The US Centers for Disease Control and Prevention also supports the use of bystander training to prevent sexual violence.11

Experts recommend setting up an ombudsman office to deal with incidents of workplace sexual harassment to provide an alternative to a formal legal process. This would allow sexual harassment complaints to be handled confidentially and investigated before the accused even knows about the complaint, which can help reduce retaliation. The person experiencing harassment is able to share their story with the ombudsman and learn their options before going forward with legal measures.12

Sexual harassment is a serious public health concern and workplace hazard. Although it predominantly affects women, it should not be cast as a women’s issue. If 5% of the workforce experienced asbestos exposure, there would be public outcry and immediate reforms. We believe no workplace can be considered safe unless it is free of harassment, and this problem cannot be sidelined any longer. Promising, evidence based solutions exist and should be widely implemented and evaluated. Victims of sexual harassment should receive mental health screening and treatment to mitigate risks for subsequent mental health concerns and suicidality.


  • Competing interests: We have read and understood BMJ policy on declaration of interests and declare that US serves as a scientific adviser for non-profit organisations HealthTech 4 Medicaid and for HopeLab. She has been a clinical adviser for Omada Health and an advisory panel member for Doximity. She is funded by Agency for Healthcare Research and Quality, California Health Care Foundation, Center For Care Innovation, National Cancer Institute, US Food and Drug Administration and the Commonwealth Fund. She is also supported by an unrestricted gift from the Doctors Company Foundation and has previously received funding from the Gordon and Betty Moore Foundation and the Blue Shield of California Foundation. She holds contract funding from Appliedvr, Inquisithealth, and Somnology.

  • Provenance and peer review: Commissioned; not externally peer reviewed.