Data from the World Health Organization (WHO) indicate that in the global population, one in eight people live with mental disorders, regardless of social class, gender, ethnicity, culture, or country. Approximately one billion people suffer from psychological disorders, and after the pandemic, there was a 25% increase in the prevalence of depression and anxiety cases (1). Studies (2, 3) indicate that psychiatric emergency services are struggling with increased patient demand, leading to overcrowding in emergency units, lack of qualified professionals, and inadequate physical structures.

Psychiatric emergencies can include psychoses, agitation, anxiety, substance use, mood disorders, suicidal or homicidal behavior, as well as overlapping clinical conditions such as delirium and dementia (3). Emergency situations represent complications of mental disorders and may involve social exposure, restraints and more aggressive approaches (4).

Brazilian national studies (5–9) show that mental disorders are a crucial part of emergency care in general hospitals. The literature (5–12) indicates that the most prevalent mental disorders include schizophrenia, anxiety disorders, disorders due to the use of psychoactive substances, depressive and mood disorders, neurodevelopmental disorders, stress-related and somatoform disorders, and schizotypal and delusional disorders. Regarding suicide ideation or attempt, connecting individuals in need of psychiatric treatment with adequate medical services has proven to be an effective suicide prevention strategy in Japan (13).

Evaluating suicidal ideation in emergencies can alert professionals to subsequent suicidal acts and offer insights into the patient’s degree of psychological distress (14). However, despite attending to patients with mental disorders and suicide risk in emergencies, clinicians feel emotional distress and discomfort asking about suicidal ideation due to fear of feeling responsible, contributing to avoidant anxious behavior (15). Currently, with the increase in patients with psychiatric complaints arriving at emergencies, professionals need to have knowledge about conducting psychiatric interviews (16).

A retrospective cohort study conducted in England examining patient characteristics and healthcare resource utilization revealed that individuals with major depressive disorder (MDD) and moderate to high suicidal intent are generally younger, predominantly male, and present with higher rates of comorbid anxiety, substance abuse, and healthcare utilization compared to those without suicidal intent (17). These findings underscore the urgent need for more effective medical interventions to alleviate symptoms, reduce suicide-related mortality, and mitigate the burden on healthcare systems.

Suicide prevention strategies should also incorporate the assessment of physical aggression and impulsive behavior, which may act as proximal facilitators of suicidal acts. Notably, impulsivity scores among females were significantly higher in the low-lethality suicide attempt group than in suicide ideators, whereas levels of hostility and aggression were significantly elevated among females with high-lethality suicide attempts compared to ideators (18). A study (19) highlights that an initial presentation to the emergency department (ED) for self-harm is a significant risk factor for subsequent self-harm and suicide, showing that approximately one in four individuals later re-present to the ED with recurrent self-harm episodes.

A systematic review identified that self-harm and aggression are driven by their association with separate self-harm or aggressive behaviors. These findings suggest that dual harm does not constitute a distinct clinical entity. Rather, complex interactions between risk factors associated with self-harm and aggression, as well as their multiplicative effects, may lead to dual harm (20). In a general adult psychiatric setting, patients at risk for violent behavior, especially dual-harm patients, should be identified and monitored as part of the risk assessment (21).

People with mental disorders have the right to receive immediate and qualified assistance during psychiatric crises (3). Suicide attempts and hetero aggression are key clinical outcomes in emergency mental health care, as they represent critical risk predictors for adverse events and potential need for hospitalization. However, there is a lack of mental health assessment instruments in emergencies in Brazil that address mental state, social support, self-care ability, and risk of self-harm and hetero aggression.

In this context, the interRAI Emergency Screener for Psychiatry (ESP) (22) includes specific domains designed to identify these risks, particularly through the Severity of Self-Harm, Risk of Harm to Others, and Self-Care Index scales, which support early risk recognition and clinical decision-making across emergency settings. The inclusion of the Life Stress Events Scale (23) aimed to assess whether patients had experienced potentially stressful life events during the previous year. Such events may trigger crises, mental disorders, and increased risks for the patient.

Given these considerations, the study aims to describe the main characteristics and clinical symptoms of the sample of mental health patients across main diagnoses, and to identify the risk factors of suicide, as well as of hetero aggression.

A total of 182 patients were interviewed, with ten interviews excluded from the sample due to missing data on the patient’s age or other socio-demographic information. Due to missingness in diagnostic information, 16 patients were excluded in the second step of the analysis.

Table 1 shows data from 172 patients attended at the UPA and HU within the first 24 hours of symptom onset. The total sample consisted of 56.98% men with a mean age of 41.5 (± 14.7) years. Most patients were admitted at the family’s request (60.1%), and 38.95% had a history of 1 to 3 previous admissions. Psychiatric hospitalization in Brazil is divide in three modalities based on the degree of patient consent and the authority responsible for the decision to hospitalize: voluntary hospitalization, conducted with the patient’s free and informed consent; involuntary hospitalization, which occurs without the patient’s consent and at the request of a third party, typically a family member or legal guardian; and hospitalization by judicial determination (compulsory hospitalization), which is ordered by a court.

A total of 77 participants were interviewed at the HU, of which 61.04% were men. The most frequent referral forms to the HU were 47.62% by the family network, 23.81% by the Psychosocial Care Center (CAPS), and 21.43% of patients requested their own admission. The UPA sample included 95 participants, of which 53.68% were men, with the most frequent referral forms by the family network (65.93%) or the patients themselves requesting admission (23.08%). No significant differences were found between locations for age, sex, and having dependents. The only significant differences between the samples were found for the form of referral.

Table 2 shows the main diagnoses in the sample and their clinical characteristics, as well as the ESP instrument scales. The most prevalent mental health disorders in the sample are addictions (32.56%) and schizophrenia (26.74%). Significant differences were found indicating a higher presence of patients with schizophrenia and anxiety disorders in the hospital, while more depressive disorders were present in the UPA. In the sample the diagnoses were clinical, made by psychiatrists, using the DSM5.

Regarding the clinical characteristics of the total sample, 75.58% of patients had a high degree of social withdrawal according to the social withdrawal scale, with similar percentages in both samples. The risk of harm to others scale showed that 62.21% presented risks of hetero aggression, and the self-harm severity scale showed that 59.88% of patients had a risk of severe self-harm, both scales without significant differences between samples.

Table 3 shows that interviewees with significantly higher rates of financial difficulties were patients with addiction (80.36%) followed by patients with anxiety diagnosis (72.73%). Addicted patients were also the most forced to move residence (50.0%), victims of aggression (39.29%), or robbery (12.5%).

The EVPE instrument analysis showed that 60.23% of interviewees had financial difficulties, 27.49% were forced to move, 22.81% had a marital breakup, 22.22% suffered aggression, 16.96% faced a health problem in the last year, 15.79% were hospitalized, 11.70% were mourning a family member’s death in the past twelve months, and 5.26% were robbed. Data analysis from the two institutions showed similar results, with the most significant being financial difficulty for both populations. At the UPA, 18.09% were affected by a family member’s death, while only 3.90% were affected at the HU.

The data showed that women (41.89%) planned to commit suicide significantly more than men (18.37%, p=0.001), and suicide attempts were significantly more frequent among women (54.05%) than men (36.73%, p=0.024). The logistic regression model in Table 4 shows the risk factors for suicide attempts. The following scales showed significant odds ratios: depression severity scale (2.69), risk of harm to others scale (3.31), and social withdrawal scale (4.27). A protective factor found in the item Family/friend or professionals’ concern about self-harm damage showed an odds ratio below 1, being 0.34, indicating that the suicide risk for these patients is significantly lower. No significant associations were found between suicide attempts and diagnoses or the patient’s admission location.

The second logistic regression model on Table 5 shows the risk factors for hetero aggression, using the harm to others scales. Only the mania scale and the history of multiple hospitalizations (4 or more) were significant factors, with the following odds ratios: 4.31 for the mania scale and 2.80 for previous hospitalizations. No other significant factors were found, and there were no significant associations between hetero-aggression and diagnoses or the patient’s admission location.

This study investigated the main characteristics and clinical symptoms of the sample of hospitalized mental health patients across main diagnoses, and to identify the risk factors of suicide, as well as of hetero-aggression. The study participants were predominantly male, with an average age of 41 years. The study population is similar to other national and international studies, where the main disorders identified in emergency and urgency are depressive, anxiety, and schizophrenia disorders (8, 9, 11, 29, 30). Predictors of hospitalization include a history of psychiatric hospitalization, suicidal thoughts, and a mood disorder or schizophrenia/schizotypal/delusional disorder diagnosis (12).

National studies (5, 31) refer to the main disorders evidenced in emergencies as mood disorders, psychotic disorders, and disorders due to psychoactive substance use. Suicide attempts also have high rates in emergency care in Brazil, as well as self-inflicted violence, behaviors more associated with females (32, 33). As already evidenced in the literature, our results corroborate previous research (32–34), where suicide attempts were more associated with depressive disorders and females.

The greater susceptibility of females to self-harm behavior is related to high rates of domestic violence, sexual violence, moral violence, financial vulnerability, and cultural aspects related to gender inequality (35). However, males have the highest mortality rates, being more common in the population aged 60 years or older (36, 37). Deaths related to males are associated with violent suicide attempts such as hanging, jumping from heights, and traffic accidents (38).

Our results are analogous to other studies (30, 39–41) that refer to depressive disorders, self-harm behavior, financial difficulties, violence, and alcohol or tobacco use associated with a higher prevalence of suicide attempts. A cohort study (42) with patients with Major Depressive Disorder (MDD) and suicidal behavior using DSM-5 identified that younger patients and those more likely to have comorbid psychiatric conditions such as personality disorders, substance use, and anxiety had a higher risk of suicide. The most important risk factors associated with suicidal behavior within 1 year after the onset of an MDD episode were a history of suicidal behavior and age, a history of substance use, sleep disorders, and the context of care in which the MDD was diagnosed using the ICD-10. Patients with depression should be assessed for self-harm risk, hetero-aggression, and the ability to care for themselves (41).

Suicide risk is also evidenced in patients with Bipolar Disorder (BD). Completed suicide is a major cause of death in bipolar disorder (BD) patients. Some factors related to completed suicide including early onset of illness, family history of suicide, previous attempted suicides, comorbid conditions and treatment (43). BD increases morbidity and mortality not only due to the disorder and associated suicide risk but also due to associated diseases and the side effects of medications used for treatment. All patients with BD need to be evaluated for suicidal ideation using a specific instrument that assesses this risk (44).

Research (45) showed the association between MDD and anxiety symptoms and identified that suicide attempts were associated with anxiety symptoms in young adult MDD patients, highlighting the importance of reducing anxiety symptoms in this population to prevent suicides. Climate changes that affect the most vulnerable populations also impact mental health and are related to anxiety disorders, eco-anxiety, depression, suicide, substance use, and behavioral problems (46).

Studies (47–49) with addicted patients showed a prevalence of males in situations of severe social vulnerability, corroborated in this research. Individuals who have experienced violence also present risk factors for psychoactive substance use (50). Alcoholism and drug use are referred to as risk factors for depressive disorders (51–53). Multiple diagnoses, such as substance abuse and mental disorders, are frequently reported in emergency patients with suicidal ideation (54).

Our research presents similar results to international studies, with men showing more hospitalizations for schizoaffective disorders compared to women (55, 56). A third of patients with schizophrenia are also addicted, with the most used substances being cannabis, followed by cocaine and alcohol (57). Another study shows that 37.1% of patients diagnosed with schizophrenia were smokers, and 17.3% had harmful alcohol use (56). Schizophrenia patients also have suicide risks, with suicidal behavior, suicide attempts, and self-mutilation also evident in this population (58).

In Brazil, the Ministry of Health instituted a regulation in 2011 that makes it mandatory to report self-inflicted injuries and suicide attempts for all health services (18). Later, in 2014, a new regulation (59) required health secretaries to report suicide attempts within 24 hours, and in 2019, a law was approved (60) establishing the National Policy for the Prevention of Self-Mutilation and Suicide. However, despite the new regulations and law, national research (2) indicates that emergency and urgent care services are overcrowded, lacking physical infrastructure and specialized professionals to provide qualified care to patients in psychiatric crises.

Patients with suicidal ideation or behavior were often referred to emergency care within a short period (18% in less than a month). Those previously referred for ideation/plans had a 66% higher risk of transitioning to a suicide attempt, with 25% making this transition within a month after the previous referral (61). National cohort research (62) showed that patients with a current suicide attempt and a previous attempt in the last 6 months have an especially high risk of dying by suicide shortly after going to the emergency room.

Suicidal behavior causes functional losses, representing up to a third of the total years lived with disabilities, with high mortality rates among young people (43, 63). Suicidal ideation is closely associated with both suicide attempts and deaths, serving as a significant risk factor for future suicide attempts (14). Evidence suggests that a third of high-risk patients make new suicide attempts a month after being seen in emergencies (64).

From this perspective, patients discharged need follow-up in external mental health services (65). Therefore, risk screening and classification in emergencies for patients exhibiting suicidal or self-harm behavior should be conducted, along with the treatment of all individuals (3, 65). Thus, identifying suicidal ideation and early intervention reduce suicide rates (14, 66).

Our study showed the importance of family involvement in the prevention of self-harm. This finding is supported by recent literature (67) showing that family members play an important role in suicide prevention efforts, which highlights the need of strategies to empower families and to keep them involved in the care for the patient. The study also emphasized the need for prioritizing emotional and expressive support in prevention and treatment, such as therapy, peer networks, and community resources should focus here. Families also played a key role in keeping patients safe by limiting access to means of self-harm and providing valuable contextual information to healthcare professionals. To enhance their involvement, better communication and dissemination of safety and care plans, shared learning opportunities and tailored support for carers may be essential strategies (68).

This study evidenced, using the ESP, that people with anxiety and depressive disorders present high rates of social withdrawal, risk factors for self-harm, suicidal ideation, and suicide attempts. Patients with schizophrenia had higher rates of rehospitalization, aggressive behavior, self-aggression, and hetero aggression. Schizophrenia patients also exhibit more positive symptoms and cognitive impairment. The mania and depression scales also indicated an increased risk of hetero and self-aggression, respectively. The self-harm scale showed that people with depression and schizophrenia have a higher suicide risk, while suicide plans and attempts are associated with depression. A protective factor for self-harm risk is the presence of family, friends, or professionals. The EVPE instrument demonstrated that patients with addictions report greater social vulnerability, are forced to move, and suffer violence.

The ESP proved to be a feasible instrument for use in emergency services and general hospitals, UPAs, and specialized services such as CAPS. At the end of the evaluation, the software generates a risk summary in four items: risk to oneself, risk to others, inability to care for oneself and dependents. These risks come from the scales Severity of Self-Harm (SoS), Risk of Harm to Others (RHO), and Self-Care Index (SCI). Identifying these risks can qualify care aimed at developing a personalized care plan. Additionally, a new instrument is being presented for use by different professional categories and in various contexts, such as emergency settings, mental health units in general hospitals, UPAs, and CAPS. The interRAI ESP is also useful for professional training at both undergraduate and postgraduate levels.

This study showed that the interRAI ESP is a feasible tool for psychiatric assessment in emergency settings, supporting decision-making where specialists in psychiatry are scarce. The tool can be a first help to guide care and can inform follow-up in integrated health systems. Limitations include missing sociodemographic data and a cross-sectional design, limiting causal conclusions. Despite the use of a non-representative sample, the findings provide valuable insights into a population that remains under-studied in emergency care settings. The minimum required sample size for the analysis was not achieved, and 16 patients were excluded due to missing diagnostic information; however, their inclusion is unlikely to have affected the results. A limitation of the research concerns working with the patient’s primary diagnosis. We suggest the use of secondary diagnoses in future research. A potential bias relates to the fact that all study participants were receiving care through public health services, which generally means they have lower income levels.