Violence and mental suffering among men in primary health care

OBJECTIVE: To analyze the association between male mental health problems and violence experienced. METHODS: Cross sectional study with 477 males aged between 18 and 60, users of two primary healthcare centers in São Paulo, Southeastern Brazil. The selection for the sample was based on a sequentiality criterion, according to the order of arrival of the users. Sociodemographic and health characteristics and reports of having experienced violence at any time and/or having witnessed violence in childhood were collected. Information was also collected on the use of mental health services and/or psychological complaints/diagnoses during consultation at medical clinics by reading medical records, to categorize the dependent variable “mental suffering”. The variables were described as absolute and relative frequencies. The association was tested using a confirmatory Poisson model with robust variance adjusted for age, marital status, education, violence witnessed in childhood and psychoactive substance use. RESULTS: The prevalence of mental suffering was 29.4%. Mental suffering was associated with experiencing repeated physical and/or sexual violence (RP 1.75, 95%CI 1.13;2.72). The association with a single episode of violence lost significance after the inclusion of psychoactive substance use in the model. Analysis of the fraction attributable to repetitive physical and/or sexual violence for the mental suffering of the men, verified it as 30.4%. CONCLUSIONS: The relationship between violence and mental suffering, already highlighted in studies with women, is also relevant to men’s health, drawing attention to the similar need of identification, in the health services, of situations of violence experienced by the male population. For men, this relationship was shown to be influenced by the presence of psychoactive substance use; a situation which must be dealt with, more and in a better way, by the health care service. DESCRIPTORS: Men’s Health. Mental Health. Violence. Primary Health Care. Cross-Sectional Studies. 2 Violence and mental suffering among men Albuquerque FP et al Various studies have approached the triad of men, violence and health, in the sense of recognizing health problems associated with involvement in violent situations.7,17,a The same would apply to mental health. Violent acts are regarded as elements in the male socialization process and in the exercise of masculinity, affecting how men look after their health and their bodies.17 Violent episodes are problems closely related to the health care needs of the male population.4,7 They have, however, been neglected. Studies on the impact of experiencing violence on the male population’s health problems show that episodes of aggression are associated with symptoms of mental suffering of various forms and intensities for common and more serious mental illness.11,14 In terms of assaults, episodes of domestic and urban violence are dealt with.3,11 Rhodes et al16 show that men involved in situations of intimate partner violence (IPV), seen in hospitals in the USA, presented higher rates of psychiatric problems; 18.4% showed symptoms of severe or moderate depression, while 3.3% of those not involved showed the same symptoms (p = 0.001). Those involved had a 10.3% prevalence of post-traumatic stress disorder, 1.1% for those not involved (p < 0.001). Findings by Coker et al3 corroborate the theory that experiencing IPV has negative consequences for mental health. Men who suffer IPV are more prone to show symptoms of depression, to take psychiatric medication and to develop some form of chronic mental illness, as well as perceiving their own health as fragile. Acierno et al,1 in a revision of health problems triggered by various forms of violence, stated that psychiatric problems are identified as a risk factor for experienced physical aggression. The greater part of the literature studied approached exposure to IPV. However, recent studies such as that by Kaminer et al11 also show a link with urban violence. These authors indicated that 42.9% of men in the study population, in South Africa, reported having experienced at least one form of domestic, criminal or political violence, whether in the form of moral, physical or sexual violence. For men who had experienced any of these types of violence, there was a prevalence rate of 2.2 (95%CI 1.1;4.57) diagnoses of posttraumatic stress disorder. Physical abuse during childhood, criminal violence and multiple forms of violence together were the most pathogenic forms of trauma. INTRODUCTION Alcohol abuse is a common trait in men involved in violent situations.5,20 Freitas et al8 highlighted that 57.1% of victims of physical aggression seen at an emergency room in Sao Paulo had consumed alcohol and 93.3% of these victims were male. Little is known about mental health problems in Brazilian men, and less about the link between them and experience of violence. This study aims to analyze the link between male mental illness and experiencing episodes of violence.

a The same would apply to mental health.
Violent acts are regarded as elements in the male socialization process and in the exercise of masculinity, affecting how men look after their health and their bodies. 17Violent episodes are problems closely related to the health care needs of the male population. 4,7They have, however, been neglected.
Studies on the impact of experiencing violence on the male population's health problems show that episodes of aggression are associated with symptoms of mental suffering of various forms and intensities for common and more serious mental illness. 11,14In terms of assaults, episodes of domestic and urban violence are dealt with. 3,11Rhodes et al 16 show that men involved in situations of intimate partner violence (IPV), seen in hospitals in the USA, presented higher rates of psychiatric problems; 18.4% showed symptoms of severe or moderate depression, while 3.3% of those not involved showed the same symptoms (p = 0.001).Those involved had a 10.3% prevalence of post-traumatic stress disorder, 1.1% for those not involved (p < 0.001).
Findings by Coker et al 3 corroborate the theory that experiencing IPV has negative consequences for mental health.Men who suffer IPV are more prone to show symptoms of depression, to take psychiatric medication and to develop some form of chronic mental illness, as well as perceiving their own health as fragile.
Acierno et al, 1 in a revision of health problems triggered by various forms of violence, stated that psychiatric problems are identified as a risk factor for experienced physical aggression.
The greater part of the literature studied approached exposure to IPV.However, recent studies such as that by Kaminer et al 11 also show a link with urban violence.These authors indicated that 42.9% of men in the study population, in South Africa, reported having experienced at least one form of domestic, criminal or political violence, whether in the form of moral, physical or sexual violence.For men who had experienced any of these types of violence, there was a prevalence rate of 2.2 (95%CI 1.1;4.57)diagnoses of post-traumatic stress disorder.Physical abuse during childhood, criminal violence and multiple forms of violence together were the most pathogenic forms of trauma.

INTRODUCTION
Alcohol abuse is a common trait in men involved in violent situations. 5,20Freitas et al 8 highlighted that 57.1% of victims of physical aggression seen at an emergency room in Sao Paulo had consumed alcohol and 93.3% of these victims were male.
Little is known about mental health problems in Brazilian men, and less about the link between them and experience of violence.This study aims to analyze the link between male mental illness and experiencing episodes of violence.

METHODS
This was a transversal study using data from a larger study b on intimate partner violence against women.There were 789 men interviewed, users of two primary health care services; each man to arrive at the service was asked to take part until the stipulated sample size was reached.The two health care units were selected as they had significant demand from male users, they had mental health care services and the perception, on the part of the management or administration, of the problem of violence as a health care need.These services were located in the center-west region of the city of Sao Paulo, in urbanized neighborhoods, and did not constitute a context of high levels of urban violence.That larger study aimed to estimate the prevalence of intimate partner violence against women from the male perspective, but it also produced data which could be used in other studies of the male population.A descriptive study on the prevalence and type of violence experienced by men, as well as details of the methodology used in the abovementioned research, were published by Schraiber et al. 19 There were 477 records of data collected using three instruments: a questionnaire carried out by a male interviewer and two standardized forms for gathering information from medical records.
The questionnaire included questions on sociodemographics; questions on general health; alcohol and/or illegal drug use; frequency of occurrence and reoccurrence of violent episodes experienced by the men, specifying the aggressor and the age at which the episode occurred.
In order to collect data on violence experiences, questions adapted from a validated questionnaire, 18 used with women, were used, and this showed a god level of internal consistency for male respondents. 19he questions were: for psychological violence: "Has anyone ever (in the street, in a bar, at work or at home) insulted or belittled you or made you feel worthless?";for physical violence: "Has anyone ever (in the street, in a bar, at work or at home) physical assaulted you (slaps, pushes, punches, kicks etc.)?If yes, how many times?(Once, a few times, many times)"; for sexual violence: "Has anyone ever forced you to have sexual intercourse against your will?If yes, how many times?(Once, a few times, many times)".The men were also asked if they had witnessed domestic violence during childhood (seen fights between parents and/or seen the mother assaulted by the father or partner).
Using the two types of standardized forms, the medical records of the interviewees who had been registered as users for more than one month preceding the data of the interview were examined.One reading characterized the life of the service user in terms of how often they were seen and the profile of the type of care used.In this study data on mental health appointments, available in 461 database records are used.A second reading collected complaints/diagnoses noted in the medical records of men who had used the service in the preceding 12 months, discriminating according to the kind of care provided.Psychological complaints/diagnoses available in 410 database records were used.
The 477 records were selected bearing in mind wither the use of at least one of the types of care on offer in the mental health service (individual or group psychotherapy, or psychiatric consultation), or the presence of at least one of the following complaints/diagnoses on the form: depression, anxiety/nervousness, insomnia, anger/irritability, suicide attempt or confusion/memory problems.This set of conditions was considered positive case for mental suffering, the outcome variable.
This variable was used due to the abovementioned larger piece of research not having data on common mental health disorders (CMHD), which include: somatoform disorders, depression and anxiety and symptoms of insomnia, fatigue, irritability, forgetfulness, difficulty concentrating and complaints such as headaches, tremors or poor digestion, 9 this being a category often used in studies of psychiatric epidemiology.The definition of outcome is based on Dejours et al, 6 to whom mental suffering is a general sense of malaise expressed as nervousness, worry, anxiety, tension and/ or apathy.It is not necessarily a mental illness in the strictest sense of medical nosography and may be viewed more as a psychic imbalance.Mental suffering is a subjective experience which leads to the individual having complaints and requiring health care services, such as outlined in this study.Mental suffering, therefore, includes CMHDs.
The following were considered to be independent variables; socio-demographic characteristics, age group, marital status, schooling and socioeconomic status (level A and level B were grouped together, due to the lack of subjects in these levels); psychoactive substance use (alcohol and/or illicit drug use at any point in their life); witnessing domestic violence during childhood and reporting having experienced domestic violence at any point on their life ("no report of violence"; "reported only psychological violence"; "reported physical and/or sexual violence on the same occasion" (it only occurred once); "reported repeated physical and/or sexual violence" (occurred a few or many times).Experiencing sexual violence was analyzed together with physical violence as there was no case of exclusively sexual violence.
The variable "psychoactive substance use" was characterized as "no psychoactive substance", "only alcohol" and "alcohol and illegal drug consumption" (the use of an illegal drug at any point in their life).Those who only took illegal drugs were included in the "alcohol and illegal drug consumption" category as they represented only 0.7% of the sample.
The variables were described using proportions and Pearson's Chi-square test was used to identify the differences between them.In order to better understand the link between mental suffering and violence, the various types of mental suffering in relation to each type of violence were examined.Poisson's model, with robust confirmatory variance, was used to test the link between them. 10The multiple analysis was adjusted for those variables which presented p < 0.20 in the uni-variate analysis (Chi-squared), as well as those of theoretical significance.Attributed risk was examined in the data analysis as being an indicator if interest to public policy intervention programs.
Variables related to sociodemographic questions (age group, marital status and schooling) were included in the first multiple model.The variable referring to witnessing violence during childhood was added to the second multiple model.The inclusion of psychoactive substance use in the final model was justified by the literature consulted. 2The analyses were carried out using the Stata 10.0 program.This study was approved by the Ethical Committee of the Faculty of Medicine of the Universidade de São Paulo, Process 348/10, 12/15/2010, as well as following ethical guidelines for studies about violence. 19

RESULTS
Of the 477 subjects, 29.4% were cases of mental suffering.The prevalence was higher among single men and those who used psychoactive substances (alcohol and/ or illegal drugs) (Table 1).
A strong link was found between the presence of any kind of psychological complaint/diagnosis and experiencing violence (p < 0.001); 87.9% of the medical records with this complaint were men who reported having experienced some form of violence, primarily recurring physical and/or sexual violence (45.2%) (Table 2).
The types of complaints/diagnoses associated with experiencing violence were: depression, nervousness/ anxiety, anger/being highly irritable and insomnia/ sleep disturbances.The majority of the men with records reporting depression and anger/irritability had experienced at least two of the three types of violence in question; 89.6% of the men with the most prevalent recorded (anxiety/nervousness) reported having experienced violent situations, especially of the repeated physical and/or sexual kind (Table 2).
Of the 103 medical records with a record of some kind of mental health care, 13.5% belonged to men who reported having never experienced an episode of violence.Men who used mental health care services had higher rates of prevalence of experiencing violence and almost half of them had experienced repeated physical and/or sexual violence.The most commonly used type of care was an interview or appointment with a doctor and was that which showed the strongest link with having experienced violence (p = 0.001); 6% of the men who had individual psychotherapy reported never having experienced any form of violence (Table 2).
Almost half of the men who presented mental suffering (psychology complaint/diagnosis and/or use of mental health care services) (45.7%) had experienced situations of physical and/or sexual violence once, followed by 25.7% who had experienced it on only one occasion (Table 2).Violence which was exclusively psychological did not show any link with mental suffering (Raw PR) in the uni-variate analysis (Table 3).
Except in the case of exclusively psychological violence, there was an increase in the strength of the link for the other types of violence after adjusting for sociodemographic characteristics (age group, marital status and schooling) in model 1 (Table 3).
In model 2, there was an association of the same types of violence as in the previous model, with the inclusion of the variable "witnessing violence in childhood", as well as adjusting the magnitude of the prevalence ratio (Table 3).
Repeated physical and/or sexual violence was linked to mental suffering after adjusting for psychoactive substance use, with the prevalence ratio diminishing in the final model.Use of psychoactive substances means the link with mental suffering becomes borderline when this type of violence is experienced on only one occasion (Table 3).
The fraction of mental suffering in men attributable to repeated physical and/or sexual violence was 30.4% in this model.

DISCUSSION
The prevalence of mental suffering in males in this study (29.4%) may represent a higher level than that found in other studies on mental health in men using health care services because of the wider formulation of the variable of mental suffering not being limited to CMHDs.Moreover, it may be a higher rate than that found in population studies as it is a study of health care service users.However, the rate was lower than that found in the study by Fortes et al, c in which 714 patients aged 18 to 65 were interviewed in Family Health units and a prevalence of 39.2% was found in men.That study is important for the purposes of comparison with the results of this study as it has a similar population, the same age range, it is the same type of study and the sample is of health care service users.The lower proportion found may be due to the fact that, in this study, the data were from medical records and not from a questionnaire being directly applied to the sample subjects.Three different situations may have affected the data: 1) service users may have had difficulty in spontaneously presenting this type of assault as a clinical complaint, as indicated in the literature, 4,7 whereas the interview actively sought cases; 2) the service users may do this, but it may not have been included in the records with the spontaneous complaint; 3) difficulties or less attention paid to diagnosing mental health problems on the part of the health care professionals. 7The indicator of mental health constructed here, compared to national and international studies was within the range of observations observed in those studies.
Experiencing repeated physical and/or sexual violence is associated with mental suffering, which is not the case when the violence experienced is exclusively psychological.The lack of a link between purely psychological abuse may be explained by the low prevalence of any of the types of violence studied (psychological, physical and/or sexual) occurring in isolation).Schraiber et al 19 indicate that the types of violence occur simultaneously in men, with high proportions of overlap, especially of psychological and physical types of violence.
International studies have found consistent links between experiencing some kind of violence and mental health problems. 1,11,16The majority of them focus on one particular type of violence: IPV (intimate partner violence).There is little scientific output on urban and community violence.Therefore, this study includes violence at the hands of a partner, by a family member within the home and that experienced from friends, acquaintances and strangers in public spaces.
Kaminer et al, 11 in South Africa, found a prevalence ratio of 2.2 (95%CI 1.1;4.57)for the diagnosis of post-traumatic stress disorder at some point in their life for men who had experienced at least one type of violence, be it domestic, criminal or political.Experiencing multiple types of violence proved to be more pathogenic.The cumulative effect was also highlighted: the more often violent episodes are experienced the greater the probability of developing a CMHD.
The factor of the cumulative effect of violent episodes on mental health is highlighted by Lima et al, 13 which states that there is a higher prevalence of mental disorders (around 60%) in those who have experienced stressful life events, the increasing number of these events proportionally increases the probability of metal disorders.According to Onocko-Campos & Gama, 15 "[...] the accumulation of many negative factors occurring simultaneously contributes more to the appearance of mental problems than isolated stress factors, irrespective of their magnitude".
In spite of no studies dealing only with men being found, we can use the vast output on the repercussions of violence on women's health as a counterpoint.These studies show that repeatedly experiencing IPV is linked to psycho-emotional diagnoses and/or complaints, and this effect is cumulative. 18This study reiterates that, as in women, experiencing physical and/or sexual violence on one occasion is less related to mental suffering than experiencing physical and/or sexual violence more than once.
Use of psychoactive substances has been shown to be an important adjustment variable.Our results reinforce the findings of the bibliography which indicate an interaction between mental suffering and the use of psychoactive substances.Abuse of alcohol or other drugs would be a comorbidity associated with some kind of psychiatric disorder at some point in the lives of at least 56.8% of the men in the population study in the USA, 12 which would explain the lessened strength of the link in the statistical model developed.
Repetition of violent episodes overrides the effect caused by the use of psychoactive substances on mental health and remains linked to mental health problems in men as a cause or as a consequence, as in this study there is no information related to the temporality of the events.Even so, 30.4% of cases of mental suffering were attributed to repeated physical and/or sexual violence, according to the calculation of the attributable fraction.
These statements gain further importance when we reflect on the symbolic significance of acts of violence in the social constitution of masculinity.Committing violence is reinforced in men's socialization process.
It may be considered as a fundamental element in forming their identity 17 and it requires daily affirmation throughout their lives.Fights in the street, as well as various forms of domination over partners constitute situations which legitimize masculinity.Aggressive behavior patterns in men are naturalized, often justified by biological or physiological reasons.
The naturalization of these acts makes it difficult for their repercussions to be viewed as health care needs, especially when these acts are valued by the socio-cultural demands made on masculinity, which turn them into generators of "[...] behavior which is harmful to health, encouraging the emergence of significant risk factors for falling ill", 17 especially in relation to mental health.The same process takes place with the use of psychoactive substances, considered an adjustment variable in this study and encouraging mental suffering in the male population, and valued as an element of affirmation of masculine hegemony.
This study had several limitations due to being designed primarily to assess violence perpetrated by men against their partner.Data on mental suffering and violence experienced were complementary data not mentioned in the larger study.Therefore, the data on the principal outcome were limited to data from medical records, as they were not collected through medical interviews.The category of mental suffering included any sign and/ or symptom related to mental health as positive, there being no diagnostic scale.
The links shown, although significant, may represent reverse associations as the design of the study (transversal) makes it difficult to infer causality due to the lack of control over the temporal sequence of the events.Experiencing violence could just as well be a cause of mental suffering 14 as a consequence of it. 5wever, the findings are valid, making this study relevant to understanding the relationship between mental health problems in males and experience of violent episodes.
Thus, it can be concluded that the link between experiencing violence and the situation of mental suffering, so well highlighted in studies with women, also proves to be relevant to men's health, and should be a target of future research to deepen understanding.On the other hand, the existence of this link indicates the need for health care services to identify situations of violence experienced by the male population.Moreover, for men, this relationship between mental suffering and experiencing violent situations has been shown to be influenced by the presence of psychoactive substances, which should also receive further attention from the health care services.

Table 1 .
Frequencies of the variables related to mental suffering recorded in medical records, by socio-demographic variables, associated violence and use of psychoactive substances.Sao Paulo, Southeastern Brazil, 2003.
a The variable socioeconomic status had one missing.

Table 2 .
Frequencies of variables related to mental distress recorded in medical records, according to forms of violence experienced.Sao Paulo, Southeastern Brazil, 2003.

Table 3 .
Confirmatory Poisson statistical regression model with robust variance calculation for mental distress.Sao Paulo, Southeastern Brazil, 2003.
Model adjusted for age, marital status, schooling, witnessing violence during childhood, psychoactive substance use.
The South African study converges with our results in relation to the magnitude of PR found (PR of repeated physical and/or sexual violence in subjects with mental suffering was 1.75; 95%CI 1.13;2.72).Moreover, the strength of the link between the recurring nature of the episodes of any type of violence and mental health problems was reinforced.