The present study showed associations between experience of physical violence during the past year and self rated psychological health in women, but not men, aged 18–34 years. This is in accordance with results from some of the earlier population-based studies on psychological health and violence that included both men and women. A Danish study showed more than doubled odds of symptoms of anxiety and depression in women, but not men, who had been exposed to physical violence during the past year . An Italian study of university students showed more than doubled odds of psychological distress in women, but not men, who had been exposed to a high degree of intimate partner violence (IPV) . Some population studies have shown associations between violence and psychological distress in both men and women, but to a higher degree in women. A Swedish study of 17-year old students showed increased odds of psychological distress in girls and boys who had experienced physical violence during the past 12 months, OR: 2.68 for girls and OR: 1.90 for boys . Another Swedish study on adults showed much higher odds of anxiety in women than men who had been exposed to physical violence or threats of violence during the past 12 months . A Finnish study of university students showed that exposure to violence (life-time prevalence) was strongly associated with poor mental health in both men and women, with higher symptom levels in female victims . Furthermore, population-based studies excluding men have shown significant associations between psychological distress and experience of violence in women [7, 12–14, 24].
In consistency with earlier research we found strong associations between poor self rated psychological health and socioeconomic status (especially not being part of the workforce) and psychosocial factors among both men and women . Furthermore, men of non-European origin reported poor psychological health more often than men born in Sweden. A similar but not significant pattern was seen among women. Earlier studies have shown that mental ill-health is more common among foreign-born compared to native-born Swedes, mainly due to poorer socio-economic living conditions .
In our study alcohol risk consumption was associated with poorer self rated psychological health among women but not among men. Studies have shown bi-directional relationships between high alcohol consumption and both anxiety disorders  and depression  among men and women. The paralleled increases of alcohol consumption and poor mental health in young people over the last 20 years, with a poorer development among women in both respects , may point to a connection.
Those who are socially and/or economically disadvantaged are much more likely to experience violence [3, 10]. This might in part be due to the fact that they are often restricted to live in neighborhoods with higher crime rates  and that problems with financial resources are linked to a range of negative outcomes including violence [3, 35]. The fact that socioeconomic status is associated with both violence and psychological health has been taken into account by adjusting for socioeconomic status as a confounder in the analysis. Alcohol is a risk factor for violence [15, 36], but neither a necessary nor sufficient cause [37, 38]. It plays a larger part in situational violence than in controlling violence in intimate relationships . Being violated may cause serious damage to basic trust. In our study a higher proportion of men and women who had been violated reported low trust compared to men and women who had not been exposed to violence.
Our results showed that men were most often violated in public places including streets, venues, buses, trains and subways. Women were most often violated at home, although younger women (18–24 years) were also most often violated in public places. This could be a reflection of different life styles in the two age groups. In a Swedish study, foreign-born women 18–64 years reported twice as much exposure to physical violence in the home compared to Swedish-born women .
The severity of symptoms may be influenced by the victim-offender relationship [2, 6]. Violence against women often occurs in a private, isolated context including an intimate relationship to the perpetrator, while the perpetrator of men often is unknown. Physical abuse among women is often combined with sexual and/or emotional abuse, whereas physical abuse among men often occurs in isolation . Women are less able to protect/defend themselves against perpetrators  and often have concerns on how to protect their children . Furthermore, social and economic inequalities make it harder for women to leave an abusive partner . A Swedish study showed that 22% of women aged 18–24 years had experienced some type of violence (physical, sexual or threats of violence) during the past year, and 85% worried about becoming victims of violence . There might be some connection between the concurrent high prevalence of worrying about violence and the high prevalence of poor psychological health among young women in Sweden, as worrying is negatively related to psychological health .
The impact of violence on men’s health needs to be further explored. Abuse against men is highly prevalent in Sweden. A population-based study showed that 68% of Swedish men had experienced threats of violence and/or violence at some point during their lifetime and 14% during the past 12 months. The most common forms of violence were threatening or aggressive language and physical assaults, and many men had been victimized several times . Victimization of men has been shown to be associated with health issues such as alcohol use problems [6, 23, 44]. It is possible that other measures than GHQ-12 might better capture psychological consequences of violence among men.
Strengths and limitations
The current study is subject to some limitations. Firstly, the study is cross-sectional. A cross-sectional design makes it formally hard to infer causality, although such studies may well form at least part of causal inferences. Secondly, we had only one question on physical violence and one on location, but none on frequency or relationship to the perpetrator. Thirdly, in our study 9.7% of the men and 5.0% of the women 18–34 years reported experience of physical violence during the past year. This is in line with 12% of the men and 6% of the women 16–24 years reported in The Swedish National Public Health Surveys statistics 2006–2008 , which used the same single question on physical violence. However, this is probably an underestimation. Studies with several detailed questions on physical violence have reported considerably higher figures; 28% of the men and 11% of the women 16–24 years in a Danish national health interview survey , and 25% of the boys and 15% of the girls in a Swedish study of 17 year old high school students . Earlier studies have also stated that violence against women is heavily underreported [1, 8]. For example, the Swedish National Council for Crime Prevention has estimated that 75–80% of the cases of domestic violence go unreported .
We have explored the association between poor psychological health and violence, but of course there could be other factors contributing to emotional distress that we lack information on in this study (e.g., relationship problems, illness in the family, the demise of loved ones). The item we have used to measure generalized trust in other people is self rated and thus might be difficult to validate, but it has been used in many previous investigations . Strengths of this study are the large population sample, the use of the well-validated GHQ-12 measure to assess psychological health and the use of a questionnaire to assess exposure to violence . Although there are more complex GHQ-12 instruments (with for example 28 and 60 items) to measure psychological health, there is little difference in validity [26, 28]. Furthermore, the GHQ-12 measure, as well as the question used to assess experience of physical violence, has been validated by the National Institute of Public Health and by Statistics Sweden .