Sexual and gender harassment and use of psychotropic medication among Swedish workers: a prospective cohort study

Abstract

ObjectiveTo estimate the prospective association between the exposure to three types of gender-based violence and harassment (GBVH) and psychotropic medication.MethodsInformation on three measures of workplace GBVH—sexual harassment (1) from superiors or colleagues, (2) from others (eg, clients) and (3) gender harassment from superiors or colleagues—were retrieved from the biannual Swedish Work Environment Survey 2007–2013 (N=23 449), a representative sample of working 16–64 years old registered in Sweden. The survey answers were merged with data on antidepressants, hypnotics/sedatives and anxiolytics from the Swedish Prescribed Drug Register. Cox proportional hazards analyses with days to purchase as time scale and first instance of medicine purchase as failure event were fitted, adjusted for demographic and workplace factors.ResultsWorkers who reported exposure to gender harassment only (HR 1.2, 95% CI 1.07 to 1.36), to sexual but not gender harassment (HR 1.21, 95% CI 1.04 to 1.40), or to gender and sexual harassment (HR 1.31, 95% CI 1.08 to 1.60) had an excess risk of psychotropics use in comparison to workers who reported neither of the exposures in the past 12 months. We found no interaction between the exposures and gender in the association with psychotropics use.ConclusionsExposure to sexual or gender harassment at the workplace may contribute to the development of mental disorders.

Sexual and non-sexual forms of gender harassment: a persistent problem in the world of work

“…The conceptual distinction between non-sexual and sexual gender harassment is important, and future studies should consider both the frequency and severity of different types of harassment to adequately assess levels of ‘exposure’ and their impact on different health outcomes. Furthermore, as exemplified by the Swedish study,5 data on the status and sex of harassers are often not presented and are difficult to analyse due to few cases 3. Therefore, large-scale studies or pooled data analysis, together with qualitative studies,1 are necessary to better understand whether harassment by superiors, colleagues or customer/clients has a more damaging effect on health or, more specifically, how the difference in status and sex of the perpetrator and the target of abuse interact to affect the experience and impact of different forms of harassment 2.…”

“…In this issue of Occupational and Environmental Medicine , Blindow and coworkers5 provide a methodologically sound study that addresses some of the mentioned shortcomings in the literature. Recognising the larger issue of gender harassment at work, the study defines gender-based sexual harassment (ie, unwanted advances and offensive remarks of a sexual nature) and non-sexual harassment (ie, non-sexualising sexist offences and expressions of disrespect that the affected perceived as based on their gender) as two conceptually different stressors that can have a bearing on the risk of developing a mental disorder among both women and men.…”

“…Importantly, from an occupational health perspective, the available evidence to support the inference of a possible causal association between different aspects of gender harassment and mental health is limited; with few exceptions, studies are predominantly cross-sectional and are based solely on self-reported measures of both exposure and outcomes 1 2. Blindow et al 5 have provided one of the most compelling studies in this area so far. They followed a representative sample of the Swedish working population over a six year period in the ‘national prescribed drug registry’ and observed an excess risk of ‘psychotropic use’, as an independent measure of a mental disorder, in workers who had experienced sexual or non-sexual gender harassment.…”

“…Lastly, the stereotypical harassment scenario is still that of a male boss and a female subordinate,1 2 7 and although customers or clients are often more likely to be the source of sexual harassment, for example, within the healthcare sector or the hospitality industry,7 this source of harassment has received relatively little attention in the literature 2. According to the Swedish study,5 customers/clients are more likely to be the source of sexual harassment even in the general working population. Furthermore, the results also indicated that the negative effect of sexual harassment by customers/clients on the risk of subsequent ‘psychotropics use’ was similar in strength to sexual harassment by superiors/colleagues.…”

“…In the literature, researchers commonly use two different measurement methods: the self-labelling method, which allows the respondent to define whether they feel harassed, and the behavioural method, which provides the respondent with a list of experiences defined by the researcher as harassing. The self-labelling approach, used by Blindow et al ,5 is common within the sociopsychological tradition, which emphasises that subjective experience is important in predicting victim responses and health outcomes. However, because people’s perceptions of what constitutes different types of harassment are likely to vary between men and women,1 in cultural contexts and change over time,9 the self-labelling approach can introduce bias 2.…”