Factors associated with suicidal ideation among university students 1

ABSTRACT Objective: to analyze the factors associated with suicidal ideation in a representative sample of university students. Methods: cross-sectional study, carried out with 637 students of the Federal University of Mato Grosso. The presence of suicidal ideation, demographic and socioeconomic variables, use of alcohol through the Alcohol, Smoking and Substance Involvement Screening Test, and depressive symptoms (Major Depression Inventory) were investigated. Bivariate analysis was performed with the Chi-square test and multivariate analysis using the Poisson regression model. Results: it was found that 9.9% of the students had suicidal thoughts in the previous 30 days and, in the bivariate analysis, the variables economic class, sexual orientation, religious practice, suicide attempts in the family and among friends, alcohol consumption and depressive symptoms were associated with suicidal ideation. In the multivariate analysis sexual orientation, suicide attempts in the family and the presence of depressive symptoms remained as associated factors. Conclusion: these findings constitute a situational diagnosis that enables the formulation of academic policies and preventive actions to confront this situation on the university campus.

in the family and among friends, alcohol consumption and depressive symptoms have also been shown in the literature to be relevant (6)(7)(8)(11)(12)(13) .
Therefore, among university students, the different possible factors associated with suicidal ideation may be present at an unusual time of life when many changes are taking place, which include challenges of the personal, social and academic development process that requires maturity and autonomy to take decisions considering the strict determinations of the academic environment (11,(14)(15) .
Thus, identifying the factors that are associated with suicidal ideation in university students can be an important tool for the planning of prevention and protection activities, both by university managers, as well as the health teams who assist these students on and off campus. The international literature has produced some information about suicidal ideation directed toward this population (1,6,11,(14)(15) , however, there is a lack of national studies on this subject in the university setting, a situation that reinforces the need for studies with this population.
Based on this, the aim of this study was to analyze the association of demographic and socioeconomic factors, suicidal behavior in the family and among friends, alcohol consumption and depressive symptoms with suicidal ideation among university students.

Method
This cross-sectional study was conducted with students of the Federal University of Mato Grosso-UFMT, Brazil. The sample selection method was random cluster (classes) and stratified (large areas) sampling, in which all groups were equally likely to be drawn. To determine the size of the sample, a confidence level of 95%, a ratio of 50% and an estimation error of 3.5% were considered, totaling 714 students eligible for the study.
The inclusion criterion established was for the student to be 18 years or over and, of 714 academics who responded to the instruments, 77 were excluded due to inconsistencies or blank responses, totaling 637 valid questionnaires.
To obtain the data a closed instrument was constructed, aiming to assess the demographic and socioeconomic conditions and, regarding the presence of suicidal ideation, the question "In the last 30 days have you thought about killing yourself?" was introduced. The construction of this question was directed by studies that address the theme (10)(11) . To determine the economic class the Brazil Economic

Classification Criterion of the Brazilian Association of
Market Research Companies (16) was used. A second instrument was used to identify alcohol consumption,

ASSIST (Alcohol, Smoking and Substance Involvement
Screening Test), which aims to detect the risk level of the use of tobacco, alcohol, marijuana, cocaine,

The final instrument used was the Major Depression
Inventory constructed from the DSM-IV and ICD-10 and used to identify the presence of depressive symptoms.
In the Brazilian validated version, the instrument investigates how the person has felt in the previous two weeks, with 10 multiple-choice questions featuring items 8 and 10 with two options. The alternatives range from 0 to 5 (never to all the time). The cutoff equal to or greater than 16 indicates the presence of depressive symptoms (18) .
To allow the collection of data, after authorization of access to the students by the Dean of Undergraduate Studies/UFMT, a schedule was designed for the application of the questionnaires in the class. On the appointed day, a trained applicator explained the aims of the study to the students and delivered the consent form for signature and the questionnaires, which were

Results
The prevalence of suicidal ideation was obtained through the following question: "In the last 30 days have you thought of killing yourself?" A percentage of 9.9% of the students answered yes and 90.1% indicated the no option with CI: 7.68 ; 12.47 and 87.52 ; 92.31, respectively. Table 1  Among those without a religious practice, the prevalence rate of suicidal ideation was higher compared to those that reported having a religion (p<0.001). Table 2 shows that the university students who reported cases of attempted suicide in the family and among friends were more likely to have suicidal ideation than those who did not report the event (PR=3.15, 95% CI: 1.99 ; 4.99 and PR=1.92, 95% CI: 1.20 ; 3.07, respectively).
In Table 3, it can be observed that the alcohol consumption and depressive symptoms variables showed statistically significant associations with suicidal ideation, with p=0.002 and CI: 1.31 ; 3.34 and p<0.001, CI: 5.75 ; 29.9, respectively.     (1) . Lower prevalences were found in a comparative study conducted with medical students in Austria (n=320) and Turkey (n = 326), where 5% of the Austrian university students and 3.7% of the Turkish students had thought about suicide in the previous weeks (12) . The  Similarly, these categories remained in the model after the regression (2) .
The condition of heterosexuality, socially, is configured as a major reference regarding the desires, ideals, principles and values emerging, thus there is the feeling of superiority in relation to all the other various expressions of sexuality, causing those that do not follow this reference to feel excluded and different (13) .
This condition, as a choice of a sexual orientation other than that socially expected, can lead to several consequences among university students who define themselves as homosexual and bisexual, as being the target of prejudice can arouse immense suffering and intense emotional fragility, leading to the production of suicidal ideation (20) .
Regarding the religious practice variable, the association between the not practicing a religion and the presence of suicidal ideation evidenced here was also found in a study with students from a University of Florida-USA (3) . This association, although it did not remain in the adjusted model may suggest that having a religious practice contributes to the spiritual welfare of the student inhibiting the emergence of suicidal ideation.
The exercise of religious practices, such as prayer, meditation and other manifestations of belief, contribute to the balance of emotions and feelings (21) . Thus, having a religious practice is configured as a protective factor for the individual and for the emergence of suicidal ideation.
Through this context, which involves cultural aspects and subjective values , it is important for the scientific literature to investigate religious practice (belief/habits) and its possible relationship with suicidal ideation among university students (3) , with future studies focused on this still little explored theme. Rev. Latino-Am. Enfermagem 2017;25:e2878.
Regarding suicide attempts in the family and among friends, these variables were associated with suicidal ideation. A study with a public in an age group slightly lower than in the sample of the present study (students aged 15 to 19 years) showed that the young people that had a friend who had attempted suicide were twice as likely to present suicidal ideation compared to those who did not know anyone who had attempted suicide.
Furthermore, according to the study, of the 188 young people who had suicidal ideation, 24% had someone in the family who had attempted suicide (9) .
The suicide attempt in the family variable was found to be strongly associated with suicidal ideation when the final model was applied (p<0.001). Interpersonal relationships can exert a strong influence on the behavior of an individual. Therefore, being involved with someone who has already made a suicide attempt can lead to behavior of the reproduction of the act, becoming a behavior learned as a way of resolving conflicts, thereby increasing cases of suicide (8) .
Regarding the use of alcohol, students who showed high/moderate risk for this use were twice as likely to present suicidal ideation compared to the students of the low-risk category. In a study of 1,100 University students in the USA, using a screening test for alcohol use (AUDIT), the authors demonstrated a significant association between heavy alcohol consumption and suicidal ideation (p=0.001) (22) . A study that tracked alcohol consumption in students, performed in a public higher education institution, found that of 32 university students, 28 presented risk use of this substance (19) .
Upon entering academia, the university student is faced with a new environment that brings the possibility of socialization through parties, relaxation of parental control and accountability for alcohol use, which may promote increased use in this population. It should be mentioned that alcohol consumption among university students has been associated with the presence of suicidal ideation and attempts (23) . In the present study this variable, although not remaining in the final model, indicates the need for a thorough look at this population, as in this stage they are more vulnerable to the initiation and constant consumption of alcohol, as well as its highest incidence (24) .
The association between the presence of depressive symptoms and suicidal ideation is highlighted, because among the students who had these symptoms (n=267),  (9) . Thus, suicide emerges as the only existing way out faced with a current confrontational moment and negative expectations for the future (14) .
The negative aspects that arise when a person has depressive symptoms can lead to a lack of meaning in life and a sense of powerlessness, with the emergence of these feelings strengthened in this context, predisposing the student to suicidal ideation (6) . It should be noted that even though depressive symptoms have been described as a factor associated with suicidal ideation, there can be no affirmation of a possible cause and effect relationship between depressive symptoms and suicidal ideation, as many people with depressive symptoms do not necessarily wish to end their lives (15) .
The factors associated with suicidal ideation among university students, investigated in this study, have also been evidenced in the general population (4,20) . The consequences of suicide may involve social, economic and emotional harm to family and friends, especially in younger age groups, where the impact may be more pronounced (20) . Given this context, the need is highlighted for nurses to hone their skills, from the assessment to the implementation of care in the different areas in which they operate, significantly contributing to addressing this growing public health problem.
The study presents, like any research, advantages and some limitations that need to be highlighted. The advantage referred to is that this subject is still little

Conclusion
It was found that that the variables associated with suicidal ideation were economic class, sexual orientation, religious practice, suicide attempts in the family and among friends, moderate and high risk alcohol consumption and depressive symptoms.
However, in the multivariate analysis, sexual orientation, suicide attempts in the family and depressive symptoms remained in the adjusted model.
These findings constitute a situational diagnosis that higher can assist education institutions in the promotion of prevention and coping actions regarding these questions, as well as help health professionals, who work within the campus or attend students off site, to understand the importance of measures that identify and minimize the situation.