Background. Suicide accounted for 1.3% of deaths worldwide in 2019, and is a serious public health problem with enormous variability in its characteristics according to the age, sex and region of the people who commit suicide. People with mental health problems are a vulnerable population, with a prevalence of mental health problems accounting for 56% of suicide deaths in low- and middle-income countries, reaching 90% in high-income countries. In addition, it has been shown that certain diagnoses, such as depression or alcohol use disorder, are more closely related to the presence of suicidal behaviour. The aim of this work was to identify psychiatric profiles related to suicidal behaviour and to analyse the relationships of these profiles with sociodemographic and clinical factors and the characteristics of suicidal behaviour itself. Method. A study was conducted with 683 adults (71.30% female; M= 40.85 years, SD= 15.48) from the SURVIVE study. All participants were recruited after a suicide attempt in the last 15 days in the psychiatric emergency room of seven public, general and university hospitals in Spain. An ad hoc interview was conducted to collect sociodemographic data and the International Neuropsychiatric Interview (MINI) was used to determine diagnosis. The Patient Health Questionnaire (PHQ-9) was used to assess symptoms of depression and symptoms of worry and anxiety were assessed with the General Anxiety Disorder Scale-7 (GAD-7). The Columbia Suicide Rating Scale (C-SSRS) was usedfor the assessment of suicidal ideation and suicidal behaviour and the Acquired Capacity for Suicide-Fearfulness of Death Scale (ACSS-FAD) was conducted. The Barratt Impulsivity Scale (BIS-11) was also applied. Psychiatric profiles were identified by latent class analysis (LCA), using fit indices: log-likelihood value (LliK) of model convergence, Akaike information criterion (AIC) and Bayesian information criterion (BIC) and R 2 entropy. Binary logistic regression was used to study the relationship between comorbidity profile membership and sociodemographic and clinical factors. The odds ratio (OR) was used as an estimate of the risk effect. Results. 82.73% of the sample had some diagnosis of mental disorder. Of these patients, 20.20% had more than three diagnoses. The most common mental disorder in our sample was MDD (60.47%) followed by GAD (34.85%), PT (21.52%) and AUD (16.25%). Two classes of comorbidity of mental disorders were found among the participants. Class I participants (n = 503; 73.64%), had mainly emotional disorders. The number of diagnosed conditions in Class I members was m = 1.24 (SD = 0.98), presenting low comorbidity. Class II (n = 180; 26.35%) was characterised by patients with a high likelihood of a wide variety of disorders (m = 4.49 diagnoses, SD = 1.56). Results indicated that Class II (compared to Class I) were more likely to be female (OR = 0.97, p < 0.01), younger (OR = 0.52, p < 0.05) and with more depressive symptoms (OR = 1.10, p < 0.001) and greater impulsivity (OR = 1.03, p < 0.05). In addition, participants in this class had more suicidal ideation during the interview (?2 = 8.24, df = 1, p < .01, Cohen's W = 0.11), with more reported suicidal behaviour(?2 = 26.33, df = 1, p < .001, Cohen's W = 0.2) and number of previous attempts(t = -6.18, df = 318.01, p < .001, Cohen's d = 0.53). Conclusions. The main results of our study show the identification of two comorbidity profiles, evidencing the importance of considering different subtypes of suicide attempts, according to psychiatric diagnosis. The characterisation of these groups is important to determine the associated factors and, in turn, to develop new personalised therapeutic approaches that provide valuable information about the uniqueness of the patient and clinical management in real-life settings.