Risk Factors and Prevalence of Suicide in Chilean University Students

Abstract

Mental health among higher education students is a growing public health concern in Chile, where 58 universities host a diverse student population facing significant academic and emotional challenges. This study aimed to determine the prevalence of suicidal risk, ideation, and attempts, as well as associated risk factors in Chilean university students. A cross-sectional study was conducted with 1511 participants (72.3% women, 27.7% men; mean age = 25.7 ± 7.82 years), using a digital self-administered questionnaire that included the Okasha’s Suicidality Scale (OSS), Depression, the Anxiety, and Stress Scale—21 items (DASS-21), the Emotional Exhaustion Scale (ECE), and sociodemographic variables. Logistic regression identified key factors associated with suicide attempts, such as being female (OR = 1.418, 95% CI [1.037, 1.939]), belonging to sexual minorities (OR = 2.539, 95% CI [1.899, 3.396]), being aged 26–30 (OR = 1.952, 95% CI [1.344, 2.836]), and being in the third year of university (OR = 1.483, 95% CI [1.097, 2.005]). Depression (OR = 7.065, 95% CI [5.307, 9.407]) and anxiety (OR = 1.895, 95% CI [1.400, 2.565]) were the strongest predictors, while substance use, including marijuana (OR = 2.107, 95% CI [1.620, 2.740]), cocaine (OR = 1.575, 95% CI [1.193, 2.078]), and non-prescribed antidepressants (OR = 6.383, 95% CI [1.524, 26.733]), significantly increased risk. These findings highlight the urgent need for targeted mental health interventions and policy actions in Chilean higher education to address post-pandemic increases in suicide-related behaviors.

1. Introduction

Suicide is the intentional act of ending one’s own life, often linked to underlying factors such as depression or other mental health conditions [1]. It is one of the most serious public health problems worldwide, affecting millions of people every year. According to the World Health Organization (WHO), more than 700,000 people take their own lives annually, ranking among the leading causes of death in the world [2]. This phenomenon represents an individual tragedy and significantly impacts family, social, and community environments. Recent years have shown growing concern about mental health disorders underlying suicide, such as depression and anxiety, as their prevalence has increased alarmingly following events such as COVID-19 [3,4,5]. In this context, suicide is presented as a critical priority for public health prevention policies.

University students represent a population particularly vulnerable to suicide and associated mental disorders. During adolescence to adulthood, students face numerous challenges, such as identity consolidation, increased independence, and academic pressures [6]. These demands can increase the risk of developing anxiety, depression, and other mental health problems that, if not addressed promptly, may lead to suicidal ideation or suicide attempts [7,8,9]. According to the literature, 12% of college students in the United States reported suicidal ideation in 2019 [10], while recent studies in Latin America have reported a prevalence of up to 18% in 2021 [11].

In Chile, the duration of higher education programs varies significantly depending on the type of institution and the specific field of study. Bachelor’s degree programs typically last between 4 and 6 years; however, the overall landscape of higher education includes programs ranging from 2 to 7 years, with some leading to academic degrees and others to professional or technical qualifications [12,13]. That means that students typically enter higher education at the age of 18.

In Chile, the mental health of higher education students has begun to emerge as a critical public health issue. Recent research shows that this population faces high rates of depression, anxiety, and academic stress [14], in addition to a 40.6% increase in cases of depression diagnosed during the pandemic [15]. However, current knowledge about the prevalence of suicide attempts and suicidal ideation in this group is limited, and the risk factors related to these behaviors have not yet been thoroughly studied in the national context. Furthermore, aspects such as the stigma surrounding mental disorders and the lack of adequate resources in educational institutions hinder the implementation of effective psychological support strategies [16].

This gap in research poses a significant challenge to understanding the interaction of psychological, sociodemographic, and academic factors that influence suicide risk among Chilean university students. The current literature has identified elements such as depression, anxiety, substance use, and loneliness as key risk factors, but also highlights the importance of integrating protective factors, such as social support and emotional coping resources [9]. Without a comprehensive understanding of these aspects, it is challenging to develop effective preventive interventions that respond to the specific needs of this vulnerable population.

In this context, the objective of the present study was to determine the prevalence of suicidal risk, suicidal ideation, and suicide attempts, as well as to identify the associated risk factors in Chilean university students.

2. Materials and Methods

The following study was developed following the guidelines and recommendations of Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) [17,18].

2.1. Study Design and Data Source

This study used a cross-sectional design based on self-administered surveys collected during 2023 in various higher education institutions in Chile. The data were obtained through convenience sampling using a snowball method, which allowed access to university students from multiple majors and academic levels. A digital platform was used for data collection, guaranteeing the anonymity and confidentiality of the participants. The resulting database included sociodemographic, emotional, and behavioral information collected through internationally validated instruments and adapted to the Chilean context. To ensure the representativeness of the sample and determine the appropriate sample size, statistical software (G*Power, v3.1.9.7, Heinrich-Heine-Universität, Düsseldorf, Germany) was used, which allowed obtaining the minimum size necessary for the study [19]. For this purpose, a possibly eligible population of 1,385,828 university students in Chile [20] was considered, with a confidence level of 95% and a margin of error of 5%. According to the calculations, the required sample size was 384 participants. However, a final and effective sample of 1511 students was obtained (see Figure 1). Therefore, this study far exceeds the minimum required, reinforcing the statistical validity of the results obtained.

2.2. Participants

The sample consisted of 1511 university students enrolled in various higher education institutions in Chile during 2023. The inclusion criteria considered students over 18 years of age pursuing undergraduate programs, having a device that provided an internet connection to complete the online questionnaire, and voluntarily agreeing to participate in the study after signing a digital informed consent. At the same time, the exclusion criteria were participants who did not complete all the questionnaire items and students in postgraduate or continuing education programs, since they could alter the homogeneity of the target group in terms of academic and sociodemographic characteristics. No participants were excluded due to gender, sexual orientation, or educational level to guarantee the sample’s diversity.

2.3. Ethical Considerations

This study was developed under the Declaration of Helsinki and was approved by the Central Bioethics Committee of the Universidad Andrés Bello, Approval Act 024/2022. All participants gave their informed consent expressly in the questionnaire.

2.4. Data Collection

Data were collected between November 2023 using a self-administered 40-question questionnaire in digital format, designed in Google Forms, which could be accessed through any device with an internet connection. The questionnaire was distributed through institutional emails and social networks (Facebook, Instagram, and WhatsApp), using a snowball method to effectively represent university students in Chile. The questionnaire included sociodemographic, emotional, and behavioral questions, along with internationally validated scales, such as the DASS-21 to assess depression, anxiety, and stress, and the Okasha Suicidality Scale to identify suicidal risk, suicidal ideation, and previous attempts, all in Spanish. The average duration to complete the questionnaire was approximately 20 min. The anonymity and confidentiality of the responses were guaranteed, complying with the ethical standards of mental health research.

2.5. Measurement

Okasha’s Suicidality Scale [21]: This four-item instrument is recognized for its good predictive power, ease of administration, and short completion time, and has been validated in Chile [22]. It is ideal for use in primary health care or community contexts, where early detection of risk and appropriate referral are essential, especially in populations where such risk has often been invisible and not addressed until the occurrence of the suicidal act. For the present study, Cronbach’s alpha was 0.898.

DASS-21 (Depression, Anxiety, and Stress Scale—21 items) [23]: This instrument aims to assess the presence of adverse effects of depression, anxiety, and stress, achieving maximum discrimination between these conditions. It consists of 21 questions answered on a Likert scale, where 1 = totally disagree, 3 = neutral, and 5 = totally agree. The items are grouped into three subscales. For each scale (D, A, and S), the scores of the identified items are added. In Chile, this scale is validated and has excellent psychometric properties. For the present study, the reliability of the scale was 0.955.

The Emotional Exhaustion Scale (ECE): This 10-item tool is designed to measure emotional exhaustion, which is considered one of the main components of burnout [24]. This scale assesses emotional fatigue from work overload or prolonged stress, especially in work or academic contexts. The Chilean version of the ECE [25] has excellent psychometric properties. It is recognized for its ability to capture the emotional impact of chronic stress on people, making it useful in mental health studies in various populations. In the present study, the reliability of this scale was 0.925.

Sociodemographic Variables: This questionnaire incorporated five sociodemographic questions: age, gender, people with whom one lives, academic program, and substance use. Two authors developed and subsequently validated it with three external expert peers who had no relationship with the study.

2.6. Statistical Analysis

Data were analyzed using SPSS version 29 software [26]. Initial exploratory analyses were conducted to describe the sample and assess the distribution of variables using descriptive statistics (frequencies, percentages, means, and standard deviations). The normality of continuous variables (depression, anxiety, and age) was assessed using the Shapiro–Wilk test, given its robustness for small and large sample sizes. Outliers were examined through boxplots, particularly for psychological variables, where extreme values are expected. Additionally, a correlation matrix was computed to evaluate relationships between key variables and assess patterns in the data.

Subsequently, bivariate analyses were performed using chi-square tests and Student’s t-analysis to explore initial associations between sociodemographic factors and suicidal risk. Criteria for interpreting the strength of the r coefficients were as follows: <0.1 = trivial, 0.1–0.3 = small, 0.3–0.5 = moderate, 0.5–0.7 = strong, 0.7–0.9 = very strong, or >0.9 = practically perfect.

A stepwise binary logistic regression method was used to determine the most significant predictors of suicide risk, suicidal ideation, and suicide attempts. This method allowed for the progressive selection of variables, ensuring that only the most relevant predictors remained in the final model.

Binary logistic regression was used to identify significant predictors of suicide risk, suicidal ideation, and suicide attempts. Odds ratios (ORs) with 95% confidence intervals were calculated for each independent variable. In addition, multicollinearity between independent variables was verified using the tolerance index and the variance inflation factor (VIF), ensuring the validity of the models. Finally, the overall fit of the predictive model was assessed using the Hosmer–Lemeshow goodness-of-fit test (χ² = 3.77, p = 0.926), confirming that the models provided an adequate fit to the data, and Nagelkerke’s pseudo-R² was calculated as an indicator of explained variance. Significance values (p < 0.05) were considered statistically significant.

3. Results

The main descriptive statistics of the present study are presented in

Table 1. Descriptive statistics.

Characteristics	Categories	n (%)	Suicide Risk		Suicidal Ideation		Suicide Attempt
		M ± SD	M ± SD	t/F	M ± SD	t/F	M ± SD	t/F
Gender	Female	1092 (72.3)	4.37 ± 3.68	5.39 **	3.85 ± 3.08	506 *	0.51 ± 0.87	4.82 **
	Male	419 (27.7)	3.30 ± 3.35		3.00 ± 3.35		0.30 ± 0.70
Ages	Age 18–20	321 (21.2)	4.71 ± 3.67	17.57 **	4.18 ± 3.13	18.01 **	0.52 ± 0.86	7.60 **
	Age 21–25	694 (45.9)	4.48 ± 3.66		3.98 ± 3.05		0.50 ± 0.88
	Age 26–30	198 (13.1)	4.11 ± 3.76		3.57 ± 3.13		0.54 ± 0.87
	Age 31–35	117 (7.7)	2.45 ± 2.92		2.22 ± 2.51		0.23 ± 0.59
	Age 36–40	79 (5.2)	2.05 ± 2.65		1.96 ± 2.42		0.08 ± 0.42
	Age 41 or More	102 (6.8)	2.62 ± 2.84		2.38 ± 2.40		0.24 ± 0.63
Sexual Tendency	Heterosexual	1188 (78.6)	3.42 ± 3.32	−13.5 **	3.08 ± 2.85	−13.5 **	0.33 ± 0.72	−9.04 **
	Sexual Minority	323 (21.4)	6.48 ± 3.68		5.59 ± 2.96		0.89 ± 1.04
Children	Yes	294 (19.5)	2.67 ± 3.11	−8.28 **	2.35 ± 2.58	−9.00 **	0.32 ± 0.72	−3.35 **
	No	1217 (80.5)	4.41 ± 3.66		3.92 ± 3.08		0.48 ± 0.85
Number of children	One	163 (10.8)	2.87 ± 3.26	14.6 **	2.47 ± 2.64	16.9 **	0.39 ± 0.79	3.17 **
	Two	83 (5.5)	2.18 ± 2.78		1.95 ± 2.35		0.22 ± 0.59
	Three	35 (2.3)	2.97 ± 3.37		2.68 ± 2.88		0.28 ± 0.71
	Four or more	13 (0.9)	2.61 ± 2.36		2.53 ± 2.33		0.07 ± 0.27
Couple Relationship	Yes	187 (12.4)	3.31 ± 3.59	−3.07	2.87 ± 2.90	’−3.71 **	0.43 ± 0.89	−0.28
	No	1324 (87.6)	4.18 ± 3.61		3.72 ± 3.06		0.45 ± 0.82
Chilean	Yes	1490 (98.3)	4.09 ± 3.62	1.91	3.63 ± 3.05	1.87	0.45 ± 0.83	2.01
	No	21 (1.4)	2.57 ± 3.38		2.38 ± 2.94		0.19 ± 0.60
Major Program	Professional	1319 (87.3)	4.08 ± 3.61	0.31	3.63 ± 3.05	0.56	0.44 ± 0.82	’−0.71
	Technical	192 (12.7)	4.00 ± 3.73		3.50 ± 3.05		0.49 ± 0.88
Year Level	First Year	337 (22.3)	3.74 ± 3.69	2.51 **	3.32 ± 3.14	2.54 **	0.41 ± 0.80	1.56
	Second Year	305 (20.2)	4.33 ± 3.66		3.84 ± 3.06		0.49 ± 0.87
	Third Year	335 (22.2)	4.26 ± 3.67		3.73 ± 3.04		0.52 ± 0.87
	Fourth Year	251 (16.6)	4.44 ± 3.64		3.96 ± 3.07		0.47 ± 0.86
	Fifth Year	157 (10.4)	3.80 ± 3.42		3.44 ± 2.97		0.35 ± 0.75
	Graduating	126 (8.3)	3.45 ± 3.28		3.09 ± 2.78		0.35 ± 0.73
Study Area	Education	205 (13.6)	4.14 ± 3.69	3.91 **	3.66 ± 3.14	4.45 **	0.47 ± 0.84	1.19
	Social Sciences	506 (33.5)	4.00 ± 3.64		3.52 ± 3.04		0.48 ± 0.86
	Engineering and Administration	312 (20.6)	3.66 ± 3.53		3.26 ± 2.93		0.40 ± 0.84
	Health	21 (21.2)	3.98 ± 3.48		3.58 ± 2.98		0.39 ± 0.76
	Basic Sciences	105 (6.9)	5.30 ± 3.76		4.73 ± 3.13		0.57 ± 0.90
	Arts	62 (4.1)	4.87 ± 3.78		4.33 ± 3.23		0.53 ± 0.83
Qualification Tranche	Under 40	13 (0.9)	3.84 ± 3.67	4.42 **	3.53 ± 3.12	5.15 **	0.30 ± 0.85	0.871
	Among 41–45	26 (1.7)	5.15 ± 3.12		4.61 ± 2.78		0.53 ± 0.76
	Among 46–50	227 (15.0)	4.88 ± 3.62		4.33 ± 3.03		0.55 ± 0.88
	Among 51–55	272 (17.9)	4.32 ± 3.55		3.87 ± 3.06		0.45 ± 0.78
	Among 56–60	608 (40.1)	3.89 ± 3.60		3.45 ± 3.03		0.43 ± 0.83
	About 60	365 (24.1)	3.63 ± 3.66		3.20 ± 3.03		0.42 ± 0.84
Time distribution	Full-time Student	830 (54.9)	4.51 ± 3.64	5.27 **	4.03 ± 3.09	5.83 **	0.48 ± 0.85	1.72
	Study and Work	681 (45.1)	3.53 ± 3.52		3.12 ± 2.93		0.41 ± 0.80
Program Entry	Pre-Pandemic	286 (18.9)	3.64 ± 3.36	3.6	3.28 ± 2.88	3.17	0.35 ± 0.74	3.01
	In Pandemic	587 (38.7)	4.33 ± 3.65		3.83 ± 3.05		0.50 ± 0.86
	Post-Pandemic	643 (42.4)	4.02 ± 3.69		3.57 ± 3.11		0.45 ± 0.84
Country Location	North	74 (4.9)	74 ± 3.52	0.28	74 ± 2.97	0.33	74 ± 0.82	0.15
	Center	901 (59.4)	901 ± 3.66		901 ± 3.09		901 ± 0.83
	South	536 (35.4)	536 ± 3.57		536 ± 2.99		536 ± 0.83
Consumption of Alcohol	No	350 (23.2)	3.36 ± 3.17	4.56 **	3.07 ± 2.79	4.08 **	0.29 ± 0.64	4.74 **
	Yes	1161 (76.8)	4.28 ± 3.72		3.78 ± 3.11		0.50 ± 0.87
Consumption of Marijuana	No	1009 (66.8)	3.33 ± 3.29	11.1 **	3.02 ± 2.85	10.8 **	0.31 ± 0.68	8.60 **
	Yes	502 (33.2)	5.55 ± 3.80		4.81 ± 3.09		0.74 ± 1.01
Consumption of Cocaine	No	1165 (77.1)	3.60 ± 3.49	9.46	3.22 ± 2.96	9.52	0.38 ± 0.77	5.40 **
	Yes	346 (22.9)	5.65 ± 3.63		4.95 ± 2.96		0.69 ± 0.98
Solvent Consumption	No	1501 (99.3)	4.06 ± 3.61	1.5	3.61 ± 3.05	1.01	0.44 ± 0.82	1.69 **
	Yes	10 (0.7)	5.80 ± 4.73		4.60 ± 3.65		1.20 ± 1.39
Consumption of hallucinogens	No	1452 (96.1)	4.01 ± 3.60	3.25	3.57 ± 3.04	2.97	0.44 ± 0.82	2.54 **
	Yes	59 (3.9)	5.57 ± 3.80		4.77 ± 3.04		0.79 ± 1.06
Consumption of antidepressants without a prescription	No	1366 (90.4)	3.89 ± 3.57	5.96	3.47 ± 3.02	5.8	0.42 ± 0.80	3.70 **
	Yes	145 (9.6)	5.76 ± 3.71		5.00 ± 2.96		0.75 ± 1.06
Consumption of antidepressants prescribed by a doctor	No	1306 (86.4)	3.88 ± 3.57	5.21	3.47 ± 3.02	4.89	0.41 ± 0.79	3.95 **
	Yes	205 (13.6)	5.29 ± 3.75		4.58 ± 3.06		0.70 ± 1.01

M, mean; SD, standard deviation; t = values of t-test; F = values of ANOVA; * p < 0.05; ** p < 0.00.

.

In the correlation matrix (

Table 2. Correlation matrix between suicide risk, suicidal ideation, and suicide attempt in university students.

Variables	1	2	3	4	5	6	7
1. Suicide Risk	-
2. Suicidal Ideation	0.983 **	-
3. Suicide Attempt	0.744 **	0.610 **	-
4. Emotional Exhaustion	0.370 **	0.384 **	0.202 **	-
5. Depression	0.637 **	0.655 **	0.368 **	0.594 **	-
6. Anxiety	0.478 **	0.481 **	0.316 **	0.591 **	0.730 **	-
7. Stress	0.485 **	0.491 **	0.306 **	0.664 **	0.778 **	0.833 **	-

**. Correlation is significant at the 0.01 level (2-tailed).

), significant correlations can be observed between all the variables studied. The variables suicidal risk (0.637) and ideation (0.655) have a strong correlation with depression. The same variables have a moderate correlation with anxiety.

The current study determined the frequency of suicidal thoughts, suicide risk, and suicide attempts in Chilean university students, as represented in

Table 3. Prevalence of suicidal ideation, suicidal risk, and suicidal attempt in Chilean university students.

Categories		n	%	CI 95%
Suicidal Ideation	Without Ideation	363	24.0	[21.8, 26.2]
	Mild Ideation	438	29.0	[26.7, 31.3]
	Moderate Ideation	403	26.7	[24.5, 28.9]
	Severe Ideation	307	20.3	[18.3, 22.3]
Suicide Risk	No Suicide Risk	359	23.8	[21.6, 26.0]
	Low Risk	533	35.3	[32.9, 37.7]
	Moderate Risk	389	25.7	[23.5, 27.9]
	High Suicide Risk	230	15.2	[13.4, 17.0]
Suicide Attempt	No Attempts	1100	72.8	[70.7, 74.9]
	One attempt	199	13.2	[11.6, 14.8]
	Two attempts	148	9.8	[8.4, 11.2]
	Three or more attempts	64	4.2	[3.3, 5.1]

.

Exploratory and Confirmatory Data Analysis Results

The Shapiro–Wilk test indicated that depression (W = 0.89, p < 0.001), anxiety (W = 0.91, p < 0.001), and age (W = 0.95, p < 0.001) deviated from normality. Given the nature of psychological scales, these results were expected. Boxplots revealed the presence of some extreme values in depression and anxiety, reflecting the natural variability in mental health scores. These values were not removed, as they represent valid responses from participants.

The correlation matrix confirmed significant relationships between depression, anxiety, and suicide attempts, supporting their inclusion in the regression models. To further assess multicollinearity, the variance inflation factor (VIF) was computed, with all predictor variables showing values below 5, confirming the absence of collinearity.

Finally, the Hosmer–Lemeshow goodness-of-fit test yielded χ² = 3.77, p = 0.926, indicating that the logistic regression models fit the data well, with no significant deviations between observed and predicted values.

Table 4. Logistic regression for suicidal risk in Chilean university students.

Variables	B	S.E.	Wald	df	Sig.	Exp(B)	95% C.I. for EXP(B)
							Lower	Upper
Sexual Minorities	0.932	0.161	33.523	1	0.000	2.540	1.853	3.483
Depression	1.955	0.146	179.239	1	0.000	7.065	5.307	9.407
Anxiety	0.515	0.145	12.541	1	0.000	1.673	1.259	2.225
Marijuana Consumption	0.773	0.137	31.991	1	0.000	2.165	1.657	2.830
Cocaine Consumption	0.442	0.188	5.510	1	0.019	1.556	1.076	2.251
Non-prescribed Antidepressant Consumption	1.854	0.731	6.435	1	0.011	6.383	1.524	26.733
Prescribed Antidepressant Consumption	1.018	0.271	14.124	1	0.000	2.767	1.627	4.704
Constant	−2.315	0.132	307.710	1	0.000	0.099

B = Unstandardized beta coefficient; S.E. = Standard error; Wald = Wald statistic for the significance of the coefficient; df = Degrees of freedom; Sig. = Level of significance (p-value); Exp(B) = Odds ratio; 95% C.I. for Exp(B) = 95% confidence interval for the odds ratio (Lower limit, Upper limit).

presents the results of the logistic regression analysis performed to identify factors associated with suicide risk in higher education students in Chile. The results show that sexual minorities are at higher risk for suicide (B = 0.932, p < 0.001, OR = 2.540, IC 95% [1.853, 3.483]), indicating that this group is more than twice as likely to present suicidal risk compared to those who do not belong to such minorities. Among the emotional factors, depression stands out as the strongest predictor (B = 1.955, p < 0.001, OR = 7.065, IC 95% [5.307, 9.407]), suggesting that students with depressive symptoms are more than seven times more likely to present suicidal risk. Anxiety also showed a significant association (B = 0.515, p < 0.001, OR = 1.673, IC 95% [1.259, 2.225]), increasing the risk by 67.3%. Regarding substance use, marijuana use (B = 0.773, p < 0.001, OR = 2.165, IC 95% [1.657, 2.830]) more than doubles the probability of suicidal risk, and cocaine use (B = 0.442, p = 0.019, OR = 1.556, IC 95% [1.076, 2.251]) also shows a significant increase in risk. In addition, the use of antidepressants presents a relevant pattern: Non-prescribed use (B = 1.854, p = 0.011, OR = 6.383, IC 95% [1.524, 26.733]) increases the risk more than sixfold. In contrast, students who are using prescribed antidepressants (B = 1.018, p < 0.001, OR = 2.767, IC 95% [1.627, 4.704]) are a population at risk but face half the risk compared to those who consume antidepressants unprescribed. The constant value of the model (B = −2.315, p < 0.001, OR = 0.099, IC 95% [0.099, 0.099]) indicates that, in the absence of the factors considered, the base probability of presenting suicidal risk is low.

Table 5. Logistic regression for suicidal ideation in Chilean university students.

Variables	B	S.E.	Wald	df	Sig.	Exp(B)	95% C.I. for EXP(B)
							Lower	Upper
Sexual Minorities	0.931	0.160	33.696	1	0.000	2.537	1.853	3.475
Emotional Exhaustion	0.429	0.145	8.816	1	0.003	1.536	1.157	2.039
Depression	1.868	0.147	160.501	1	0.000	6.473	4.849	8.642
Anxiety	0.468	0.152	9.514	1	0.002	1.597	1.186	2.150
Marijuana Consumption	0.655	0.138	22.575	1	0.000	1.926	1.470	2.524
Cocaine Consumption	0.436	0.193	5.101	1	0.024	1.546	1.059	2.256
Non-prescribed Antidepressant Consumption	0.441	0.214	4.240	1	0.039	1.554	1.021	2.364
Constant	−2.440	0.137	315.667	1	0.000	0.087

B = Unstandardized beta coefficient; S.E. = Standard error; Wald = Wald statistic for the significance of the coefficient; df = Degrees of freedom; Sig. = Level of significance (p-value); Exp(B) = Odds ratio; 95% C.I. for Exp(B) = 95% confidence interval for the odds ratio (Lower limit, Upper limit).

presents the results of the logistic regression analysis to identify factors associated with suicidal ideation in Chilean higher education students. The findings indicate that belonging to sexual minorities significantly increases the probability of suicidal ideation (B = 0.931, p < 0.001, OR = 2.537); that is, sexual minority students are more than twice as likely to present suicidal ideation compared to those who do not belong to these minorities. Among the emotional factors, depression is the most relevant predictor (B = 1.868, p < 0.001, OR = 6.473), which implies that students with depressive symptoms are more than six times more likely to experience suicidal ideation. Furthermore, anxiety (B = 0.468, p = 0.002, OR = 1.597) and emotional exhaustion (B = 0.429, p = 0.003, OR = 1.536) also showed significant associations, increasing the probability of suicidal ideation by 59.7% and 53.6%, respectively. Substance use stood out as a relevant factor. Marijuana use (B = 0.655, p < 0.001, OR = 1.926) almost doubled the probability of suicidal ideation, while cocaine use (B = 0.436, p = 0.024, OR = 1.546) also showed a significant association. On the other hand, the use of antidepressants without a prescription (B = 0.441, p = 0.039, OR = 1.554) increases the probability of suicidal ideation by 55.4%, suggesting that these students present additional vulnerabilities in terms of mental health. The constant value of the model (B = −2.440, p < 0.001, OR = 0.087) indicates that, without the presence of the analyzed factors, the base probability of suicidal ideation is low.

Table 6. Logistic regression for suicide attempt in Chilean university students.

Variables	B	S.E.	Wald	df	Sig.	OR	95% CI OR
							Lower	Upper
Woman	0.349	0.160	4.793	1	0.029	1.418	1.037	1.939
Sexual Minorities	0.932	0.148	39.464	1	0.000	2.539	1.899	3.396
Age between 26 to 30 years	0.669	0.191	12.325	1	0.000	1.952	1.344	2.836
Third Year	0.394	0.154	6.569	1	0.010	1.483	1.097	2.005
Depression	0.985	0.148	44.573	1	0.000	2.678	2.005	3.575
Anxiety	0.639	0.154	17.106	1	0.000	1.895	1.400	2.565
Marijuana Consumption	0.745	0.134	30.933	1	0.000	2.107	1.620	2.740
Cocaine use	0.454	0.141	10.301	1	0.001	1.575	1.193	2.078
Prescribed Antidepressant Use	0.580	0.181	10.231	1	0.001	1.786	1.252	2.547
Constant	−3.054	0.190	257.547	1	0.000	0.047

B = Unstandardized beta coefficient; S.E. = Standard error; Wald = Wald statistic for the significance of the coefficient; df = Degrees of freedom; Sig. = Level of significance (p-value); Exp(B) = Odds ratio; 95% C.I. for Exp(B) = 95% confidence interval for the odds ratio (Lower limit, Upper limit).

shows the results of the logistic regression analysis performed to identify factors associated with suicide attempts in Chilean university students. The results show that women (B = 0.349, p = 0.029, OR = 1.418) and individuals from sexual minorities (B = 0.932, p < 0.001, OR = 2.539) are significantly more likely to attempt suicide, with sexual minorities being more than twice as likely. Among the sociodemographic factors, students in the 26 to 30 age range have a significantly higher risk of suicide attempts (B = 0.669, p < 0.001, OR = 1.952) compared to students outside this age group (younger than 26 or older than 30 years). Likewise, being in the third year of university is also associated with a higher probability of suicide attempts (B = 0.394, p = 0.010, OR = 1.483). Emotional factors emerge as important determinants. Depression more than doubles the likelihood of a suicide attempt (B = 0.985, p < 0.001, OR = 2.678), while anxiety also showed a significant association (B = 0.639, p < 0.001, OR = 1.895), increasing the risk by 89.5%. Substance use is significantly associated with suicide attempts. Marijuana consumption more than doubles the risk (B = 0.745, p < 0.001, OR = 2.107), while cocaine use also shows a significant effect (B = 0.454, p = 0.001, OR = 1.575). Finally, patients who are under-prescribed antidepressant treatment are at risk of suicide by 78.6% (B = 0.580, p = 0.001, OR = 1.786). The constant value of the model (B = −3.054, p < 0.001, OR = 0.047) indicates that, in the absence of these factors, the probability of a suicide attempt is low.

4. Discussion

At the end of this study, which aimed to determine the prevalence of suicidal risk, suicidal ideation, and suicide attempts and their risk factors in Chilean university students, it was determined that these prevalences were high. Specifically, 47% of the sample presented suicidal ideation from moderate to severe, 40.9% of the sample presented a moderate to high suicidal risk, and 27.2% had attempted suicide at least once. This highlights the high prevalence of these phenomena in the mental and emotional health of these students. This may be because university students are currently presenting degrees of depression, anxiety, and post-pandemic stress [5], which may be closely linked to the appearance of suicidal risk, suicidal ideation, and suicide attempts. In addition, academic overload, emotional exhaustion, and lack of psychosocial support can also impact these high suicide indicators [27]. Specifically, the disruption of social and academic life and widespread uncertainty during the pandemic negatively affected mental health, exacerbating pre-existing disorders such as depression and anxiety [28,29].

4.1. Symptoms of Depression and Anxiety as Risk Factors

The results of the logistic regressions (see Table 4, Table 5 and Table 6) indicate that both depression and anxiety are significantly associated with the different manifestations of suicidal behavior. In the case of suicidal risk, depression (OR = 7.06, p < 0.001) and anxiety (OR = 1.67, p < 0.001) were significant predictors. Suicidal ideation was also associated with depression (OR = 6.473, p < 0.001) and anxiety (OR = 1.597, p = 0.002). Finally, in the suicide attempt model, significant associations were observed with depression (OR = 2.678, p < 0.001) and anxiety (OR = 1.895, p < 0.001). These findings suggest that depression is a particularly important risk factor for suicidal behavior, whereas anxiety also shows a consistent association, albeit with a smaller effect size. Depression is associated with intense feelings of hopelessness, worthlessness, and persistent sadness [30]. These emotional states increase the desire to escape suffering, which can translate into suicidal ideation [31]. Individuals with depression have an impaired ability to manage difficult emotions, increasing vulnerability to making impulsive decisions, such as attempting suicide [32]. Specifically, patients who report suicide attempts have been found to endorse higher levels of general impulsivity, and those with recent attempts report more difficulties with suicidal impulses [33]. These findings highlight the complex interplay between depression, hopelessness, and impulsivity in suicide risk assessment and intervention. On the other hand, anxiety disorders generate constant physiological activation, such as increased stress and threat perception, which can amplify psychological discomfort and the feeling of being trapped in an unsustainable situation [34]. Anxiety often coexists with depression in college students, creating a cycle in which constant worry aggravates depressive symptoms and vice versa; this cycle increases the risk of suicidal ideation [35]. In this sense, it has been shown that people with anxiety tend to avoid experiences that cause them discomfort. Still, when they constantly prevent them, they can develop a feeling of failure or inability to deal with problems, which can lead to suicidal thoughts [36]. Likewise, university life has been considered a stressful stage with high academic demands and pressure, which can lead to the symptoms of anxiety and depression being exacerbated [37]. This is how recent studies have shown that those who suffer from anxiety and depression present a decrease in social participation [38]. The latter reduces emotional support networks and increases loneliness, which may impact the appearance of suicidal thoughts [39].

4.2. Vulnerability of Sexual Minorities and Other Specific Groups

The results of the logistic regressions (see Table 4, Table 5 and Table 6) indicate that belonging to sexual minorities is significantly associated with the different manifestations of suicidal behavior. In the case of suicidal risk, being part of sexual minorities was a significant predictor (OR = 2.540, p < 0.001). Suicidal ideation was also associated with belonging to sexual minorities (OR = 2.537, p < 0.001). Finally, in the suicide attempt model, a significant association was observed with being part of sexual minorities (OR = 2.539, p < 0.001). These findings suggest that belonging to sexual minorities constitutes an important risk factor for suicidal behavior in college students, which could be due to experiences of discrimination, stigma, and lack of social support. University students who belong to sexual minorities often face a series of complex situations loaded with prejudice, rejection, and discrimination [40]. These stigmas are reflected through peer rejection, generating social exclusion and isolation [41], as well as situations of physical and psychological violence, such as bullying or cyberbullying [42], increasing the risk of mental health problems, impacting the appearance of suicidal ideation and suicide attempts [43,44]. Besides these external factors, sexual minorities are often associated with body image concerns [45,46]. However, protective factors such as social connection and identity affirmation can moderate the negative impacts of victimization on mental health [47]. Cultural and social factors, including family support and access to healthcare, also play a role in the mental health outcomes of sexual minorities [48]. These findings highlight the need for targeted interventions to support the well-being of sexual minority college students.

Within this study, female higher education students were more likely to attempt suicide. This may be because this gender has higher prevalence rates of depression, anxiety, and other affective disorders compared to men [49]. This condition was exacerbated post-pandemic, where the differences with the male gender increased even more [50]. Women often take on multiple responsibilities in different areas (such as work, studies, and family care), which can lead to emotional exhaustion and chronic stress [51]. The expectation of fulfilling these roles can create a sense of inadequacy and exhaustion, increasing vulnerability to mental health problems [52]. Women are also more likely to experience physical, psychological, or sexual violence, which increases the risk of developing post-traumatic stress disorder (PTSD) and affective disorders. The accumulation of these traumatic experiences increases the likelihood of suicide attempts [53,54].

4.3. Consumption of Drugs and Other Illicit Substances

The results of the logistic regressions (see Table 4, Table 5 and Table 6) indicate that substance use is also significantly associated with the different manifestations of suicidal behavior. In the case of suicidal risk, marijuana (OR = 2.16, p < 0.001), cocaine (OR = 1.56, p = 0.019), and non-prescription antidepressant (OR = 6.38, p = 0.011) use were significant predictors. Suicidal ideation was associated with marijuana (OR = 1.926, p < 0.001), cocaine (OR = 1.546, p = 0.024), and non-prescription antidepressant (OR = 1.554, p = 0.039) use. Finally, in the suicide attempt model, significant associations were observed with marijuana (OR = 2.107, p < 0.001), cocaine (OR = 1.575, p = 0.001), and prescription antidepressant (OR = 1.786, p = 0.001) use. These findings suggest that substance use, especially marijuana and non-prescription antidepressants, constitutes a relevant risk factor for suicidal behavior in college students. The use of drugs and illicit substances is also considered a determining factor in the appearance and development of risk, ideation, and suicide attempts. This can be explained, for example, by the chronic use of marijuana, which can affect executive functioning, making rational decision-making difficult, and increasing impulsivity, factors that can contribute to suicide attempts [55]. Similarly, a meta-analysis study [56] and a longitudinal study conducted in the United States [57] have demonstrated a strong association between frequent marijuana use and psychosocial disorders, such as depression and anxiety—both of which are well-established risk factors for suicide. While the latter is a study conducted in the U.S., its findings align with global research on the psychological effects of marijuana use, supporting the broader applicability of these associations, including within the Chilean university student population. While some people turn to marijuana as a form of self-medication to relieve stress or anxiety, its prolonged use can worsen these symptoms, perpetuating a cycle of psychological discomfort [58]. Therefore, frequent consumption can lead to a reduction in social interactions and emotional support, increasing the feeling of anxiety, which is a factor associated with suicidal risk [59].

On the other hand, cocaine increases dopaminergic activity, which can generate episodes of euphoria followed by severe depression [60]. During depressive states, impulsivity and despair significantly increase the risk of suicide [61]. Chronic cocaine use can cause mood disorders such as major depressive episodes, which exacerbate suicidal ideation [62]. Specifically, under the effects of cocaine, people are more likely to act impulsively without considering the consequences, including suicide attempts [63]. This is because cocaine’s effects are influenced by the dopamine β-hydroxylase genotype, with low-activity genotypes showing increased sensitivity to drug-induced impulsivity and reduced corticostriatal connectivity [64]. Furthermore, cocaine users often face increased social stigma, which may limit their access to emotional and professional support [65]. Taking antidepressants without medical supervision can result in inappropriate dosages, drug interactions, or adverse effects such as initial worsening of depressive symptoms or anxiety [66]. Non-prescribed antidepressant use often reflects attempts to self-medicate for severe emotional problems, such as depression, that have not been adequately treated [67]. Without medical supervision, possible signs of worsening mental status, such as increased anxiety, insomnia, or suicidal thoughts, are not monitored [68].

While marijuana, cocaine, and antidepressants are often used as tools to avoid or suppress negative emotions [69], this can prevent students from facing and resolving underlying mental health issues [70]. The use of these substances can lead to family conflict, social isolation, and a reduction in support networks, which increases emotional vulnerability, thus aggravating pre-existing conditions such as depression, anxiety, or impulsivity, intensifying suicidal risk [71]. Therefore, it is necessary to develop early identification screening, psychoeducation, and interventions in university settings to support students at risk of presenting coexisting substance use and mental health problems.

Risk of Students on Antidepressant Treatment, Correlation, and Causality

The results of the logistic regressions (see Table 4, Table 5 and Table 6) indicate that antidepressant use is significantly associated with the different manifestations of suicidal behavior. In the case of suicidal risk, consumption of antidepressants without prescription (OR = 6.383, p = 0.011) and with prescription (OR = 2.763, p < 0.001) were significant predictors. Suicidal ideation was associated with non-prescription antidepressant use (OR = 1.554, p = 0.039). Finally, in the suicide attempt model, a significant association was observed with prescription antidepressant use (OR = 1.786, p = 0.001). Notably, the association between antidepressant use and suicidal behavior is complex and may be influenced by factors such as the severity of depression, the presence of other mental disorders, and treatment adherence. Further research is required to understand this relationship fully. Suicide risk in students on antidepressant treatment may be related to several complex factors beyond the medication itself [27]. It is essential to consider that many of these students are already at increased risk for suicide because of the severity of the depressive symptoms that prompted treatment in the first place [39]. In addition, in the first weeks of treatment with some antidepressants, an increase in energy may occur before a significant improvement in mood is observed, which could facilitate the execution of previously present suicidal plans [66]. However, it is essential to note that the risk of suicidal ideation and/or suicide attempt in students on pharmacological treatment responds essentially to the condition that motivated the initiation of treatment, and the medication is not the trigger. On the contrary, without treatment, the symptomatology could worsen [72]. When analyzing this relationship through logistic regression, it is essential to emphasize that the results indicate a statistical association, not a causal relationship.

4.4. Other Risk Variables

The results of the logistic regressions (see Table 4, Table 5 and Table 6) indicate that age and year in college are also significantly associated with the different manifestations of suicidal behavior. In the case of suicidal risk, being between 26 and 30 years old (OR = 1.952, p < 0.001) and being in the third year of college (OR = 1.483, p = 0.003) were significant predictors. Suicidal ideation was associated with being between 26 and 30 years old (OR = 1.631, p = 0.003) and being in the third year of college (OR = 1.419, p = 0.011). Finally, in the suicide attempt model, significant associations were observed with being between 26 and 30 years old (OR = 1.952, p < 0.001) and being in the third year of college (OR = 1.483, p = 0.003). These findings suggest that being older than 25 years and being in the third year of college are relevant risk factors for suicidal behavior in college students. One of the risk variables for suicide attempts in this study was being between 26 and 30 years old, which can be explained by the fact that older students often face the perception that they are behind compared to their younger peers. This can lead to feelings of inadequacy and personal failure [73]. In emerging adulthood, there is increasing pressure to meet social goals, such as having economic stability, building a family, or advancing professionally. Students in this age range may feel disadvantaged, which increases the risk of thoughts of hopelessness. Many students between 26 and 30 years of age combine their studies with work or family responsibilities, which increases the stress load and reduces their time for self-care. This age range marks a critical period in development, where people evaluate their achievements against their expectations. Perceived failure can trigger depressive and anxiety symptoms. This age range marks a crucial period in development, where people assess their achievements against their expectations. Perceived failure can trigger depressive symptoms and anxiety [74].

Finally, third-year college students are at higher risk for suicide attempts. This can be explained by the fact that the third year of college is usually a stage in which students face the most complex subjects of their programs, significantly increasing academic stress. Therefore, they feel more significant pressure to demonstrate advanced skills since they are closer to the end of their degree, and personal and social expectations are higher [75]. This period may include essential choices, such as specializations, internships, or preparation for final projects, which increases anxiety about the future. During the third year, some students question whether their college academic program is right for them [76]. These doubts can lead to hopelessness and feelings of having wasted time or resources. The proximity to graduation generates concern about the job market, the need to gain experience, and the lack of clear plans, which can increase the feeling of uncertainty.

4.5. Limitations

Although this study presents a representative sample, we must mention that the use of a cross-sectional design can be complex since it limits the ability to establish causal relationships between the identified risk factors and suicide attempts. This implies that it cannot be determined whether the factors precede or are a consequence of suicidal risk. In addition, the present study sample was obtained through convenience sampling using a snowball method, which could have introduced selection bias and limited the representativeness of the results for the entire population of university students in Chile. The data were collected through self-administered questionnaires, which may have led to a social desirability bias or errors in self-reporting sensitive information such as substance use and suicide attempts.

4.6. Future Studies

Future studies should address this study’s limitations, particularly its cross-sectional design, which does not allow for causal inference. Future studies should adopt longitudinal approaches to assess the temporality and causality between risk factors and suicide attempts, providing a deeper understanding of the observed relationships. Additionally, expanding the sample to include a more extensive and diverse population representing different regions and socioeconomic contexts of Chile would enhance the generalizability of the findings to the entire university population and various educational contexts. It would also be beneficial to analyze the faculties or disciplines students belong to and evaluate their associated suicide risks.

5. Conclusions

In Chilean university students, the prevalence of suicidal risk (moderate to high) is 40.9%, suicidal ideation (moderate to severe) 47%, and past suicide attempts 27.2%. The associated risk factors for suicidal risk include being part of sexual minorities, depression and anxiety symptomatology, marijuana and cocaine consumption, and consumption of antidepressants. Risk factors for suicidal ideation include being part of sexual minorities, being emotionally exhausted, depression and anxiety symptomatology, marijuana and cocaine consumption, and consumption of antidepressants without prescription. For suicide attempts, the risk factors include being part of a sexual minority, being in the 26–30 y age group, being in the third year of university studies, having depression and anxiety symptomatology, marijuana and cocaine consumption, and receiving treatment with antidepressants.

Practical Implications

These results emphasize the pressing necessity for focused mental health interventions in Chilean higher education institutions. Psychoeducation, early identification and screening, and culturally sensitive mental health programs that address emotional distress, substance misuse, and the specific needs of high-risk groups, including sexual minorities, should be included in comprehensive strategies. Furthermore, policies that prioritize stigma reduction, social support, and accessible professional care should be prioritized to address the escalating post-pandemic mental health crisis among university students.

Stakeholders can more effectively address the multifaceted issue of suicide in this at-risk population by advancing understanding and devising evidence-based interventions.