Trends in suicidal behavior at a general hospital emergency department in southern Brazil

Abstract Objective To assess the prevalence of and factors associated with suicidal behavior in patients seen at the emergency department (ED) of a general hospital in southern Brazil. Method Descriptive, observational, cross-sectional study. The records of all patients who had an emergency psychiatric consultation at the ED conducted by the emergency psychiatric consultation service at Hospital São Lucas da Pontifícia Universidade Católica do Rio Grande do Sul in 2016 and 2017 were analyzed and stratified by sex and by age groups (15-29 years, 30-49 years, 50-69 years, and 70 years and older). Suicidal behavior was characterized by factors such as thoughts of death, suicidal thoughts, and suicidal risk. Suicidal behavior was compared by sex and between age groups with chi-square tests. Multivariate analysis of suicidal behavior and gender, age, and specific diagnoses were compared with Poisson regression. Results A total of 1,172 records from January 2016 to December 2017 were examined. There were more ED visits by females (63.1%) than males. Younger patients (15-29 years) had a higher severe risk of suicide than elderly (≥ 70 years) patients (54.1 vs. 19%; p < 0.01). Indicators of suicide behavior stratified by sex and by age group revealed marked differences between age groups for all variables among female patients. Overall, age group patterns for males were very similar in terms of suicidal behavior variables. Conclusions A high prevalence of suicidal behavior was observed in this sample, particularly among young adults and especially associated with female gender and diagnoses of depression and personality disorders.


Introduction
According to the World Health Organization (WHO), more than 800,000 people die by suicide each year. 1 For every suicide, there are many more people who attempt suicide every year and there are indications that for each adult who died of suicide there may have been more than 20 others attempting suicide; significantly, a prior suicide attempt is the single most important risk factor for suicide in the general population. 1 Suicidal behavior ranges from ideation of killing oneself to forming a plan and obtaining the means to execute the act. Among young people 15-29 years of age, suicide is the second leading cause of death globally, 1 and the fourth leading cause of death in Brazil. 2 The global suicide rate is 10.6 per 100,000 people. 3 The mortality rate from suicide is rising in Brazil, from 5.3 per 100,000 in 2011 to 5.7 per 100,000 in 2015. 2 These rates are higher in the southern states of Brazil. In 2016, Rio Grande do Sul, the state with the highest incidence of suicide, had a rate that was double the Brazilian average. 4 Regardless of the large variability in average suicide rates among Brazilian states, there was a stable or increasing trend in suicide rates from 1997 to 2015 in Brazil. 5 Suicidal ideation and fatal and nonfatal suicide attempts often begin at age 12-13 and increase during adolescence, 6 and may establish a pattern of suicidal behavior in young adults. 7 Suicidal behavior among adolescents and young adults is related to this stage of development, which is characterized by presence of risk factors for suicide such as substance abuse, depression, thrill-seeking, and high-risk behaviors. 8 The hospital ED is often the gateway to care for unrecognized or untreated cases of mental disorders, including suicide attempts and patients at risk of suicide. 9,10 The risk of suicide or other causes of death following ideation has not been well established. A recent meta-analysis found that suicidal ideation was associated with more than a three-fold increase in the risk of suicide, 11 but the authors also noted considerable between-study heterogeneity. A significant number of individuals who complete suicide have had contact with the health system in the month prior to death. 12 The first contact with the health system can offer an excellent opportunity to identify patients at risk and to intervene in potentially fatal cases. In a multicenter study of emergency department patients with elevated suicide risk, a combination of brief interventions administered both during and after the emergency department visit decreased suicidal behavior after discharge. 13 Despite the importance of and recent increase in suicidal behavior, there is a dearth of studies investigating the prevalence of suicidal behavior in emergency care settings in Brazil. This study aims to describe the prevalence of suicidal thoughts and behaviors in people presenting to the ED at a general hospital in southern Brazil and investigate associations with clinical and sociodemographic variables. and there is no specialist pediatric mental health service at the HSL. In total, 1,172 records were analyzed.

Setting
All patients presenting at the HSL ED are initially screened by a general practitioner. The psychiatric emergency consultation service is called when an urgent need for psychiatric care is identified, such as suicidal behavior, psychomotor agitation, or risk of aggression.
The service is staffed by on-call physicians from the HSL residency and fellowship programs in psychiatry.
All consultations are entered into the medical record system and the management plan is discussed with the attending physician.

Instrument
The psychiatric emergency assessment record is routinely used by the HSL psychiatric emergency consultation service as an interview script and on-call log.

Statistical analysis
All analyses were performed using the Statistical

Results
A total of 1,172 records from consultations between January 2016 and December 2017 were examined.
Mean (SD) age was 38.6 (±18.1) years. There were more ED visits by females than males: 63.1% of the sample. The most frequent primary complaints in the sample were suicidal ideation (20.0%) and suicide attempt (17.7%). The most common psychiatric diagnosis was unipolar depression (34.7%), followed by bipolar disorder (16.7%). Table 1 presents the sociodemographic characteristics for the entire sample and stratified by age ranges.
When the total sample was stratified by age group, statistically significant differences were found in almost all items indicative of suicidal behavior, with the exception of family history of suicide. In general, younger patients (15-29 years) had a higher risk of suicidal behavior than elderly patients. Considering the age groups analyzed, 61.0% of people aged 15-29 had suicidal thoughts,  Table 3.
Univariate analyses were also conducted of associations between variables related to suicide risk    thinking about a way to kill oneself and current suicide attempt, which were not associated with gender. Once more, there was no association between family history of suicide and suicidal behavior (Table 4).

Discussion
The objective of this study was to analyze clinical and sociodemographic differences associated with suicidal behavior in patients seen at the ED by an emergency  30 One study has shown that although suicidal ideation is frequent in adolescents seen in emergency departments, it goes unnoticed by physicians and patients are not referred to mental health services. 31 Screening youth for suicidal behavior is an important way of preventing suicide. 32 Use of standard screening tools in primary care increases detection of suicidal ideation, enabling referral to mental health services before a serious or fatal suicide attempt occurs. 33 Effective intervention for suicide prevention includes involvement of family members and caregivers to provide young people with the necessary support and ensure commitment to followup after ED discharge. 34 It is equally important to continue investigating suicidal behavior during such follow-up. 35 This study has some limitations. The psychiatric emergency assessment record is completed by an oncall physician, and its items focus on delivery of urgent care; other information that is potentially useful from a research standpoint is not collected, and some records are incomplete. Since data collection occurred in a naturalistic environment, information such as the diagnostic hypothesis was obtained through a clinical interview (i.e., without the aid of psychometric instruments) and at a single point in time, and is thus susceptible to recall and information bias. Furthermore, data collection took place in a nonpsychiatric department of a general teaching hospital that mostly serves patients with private health insurance and the findings are not therefore generalizable.
The hospital does not have a specialized pediatric psychiatry service and all potential psychiatric patients (including children and adolescents) are seen by professionals trained in general psychiatry. Furthermore, as a matter of course, the population sample was limited to patients who sought care at HSL.

Conclusions
We found a high prevalence of suicidal behavior in this sample of patients seeking emergency psychiatric consultation at a general hospital, particularly among young women. Suicidality was associated mainly with young age and with diagnoses of depression and personality disorders. These findings highlight the need for robust assessment in general emergency departments, training of emergency department staff (including in the use of standardized tools to screen for suicidal behavior), and referral to specialized mental health services when needed.

Disclosure
No conflicts of interest declared concerning the publication of this article.