Loneliness, Emotional Fatigue, Social Dysfunction, and Suicide Risk in Ecuadorian University Students: A Mediation Analysis

Moreta-Herrera, R. Pontifical Catholic University of Ecuador, Ambato, Ecuador

Gordón-Villalba, P. Technical University of Ambato, Ecuador

Rodas, J.A. Espíritu Santo University, Ecuador University College Dublin, Ireland

Cuesta-Andaluz, E. Technological University Indoamérica, Ambato, Ecuador

Revelo-Sánchez, P. Armed Forces University – ESPE, Ecuador

Introduction

Suicide is a major global public health problem. Over 700,000 people die each year from this condition, and among young people, it is the fourth leading cause of death (World Health Organization [WHO], 2021). In Ecuador, an estimated 1,000 individuals commit suicide annually, and for every completed suicide, there are at least 20 attempts (Ministerio de Salud Pública del Ecuador [MSP], 2023). Therefore, it is necessary to study this phenomenon from its early stages to understand its underlying mechanisms, assess the existing risk, and prevent the act itself in a timely manner. Suicide risk is a condition that affects various populations to varying degrees, although university students constitute a notably vulnerable group. The loss of hope and life satisfaction are already indicators of impaired academic satisfaction (Santillán-García et al., 2025). Furthermore, academic, social, and personal demands can make students susceptible to mental health problems (Castillo-Gonzáles et al., 2026; Moreno-Montero et al., 2023), including suicide risk (Lapo-Talledo et al., 2025), with reported rates of 9.7% for suicidal ideation and 0.7% for suicide attempts (Crispim et al., 2021; Oleas et al., 2025). Thus, this condition does not depend on a single factor, but on a convergence of components related to emotional distress, relational difficulties, and psychosocial vulnerability. While factors associated with mental health disorders—such as stress, anxiety, depression, sleep disturbances, and general emotional distress—have been identified (Gutiérrez et al., 2021; Ward et al., 2022), contextual and daily aspects, like financial pressure and economic difficulties, also demonstrate a significant connection (Sheldon et al., 2021). Moreover, several of these elements are part of daily academic life and can be determining factors even without presenting as clinical symptoms.

Psychosocial Factors Associated with Suicide Risk in University Students

Although factors such as life satisfaction (Moreta-Herrera et al., 2025), well-being, social support (Lascano-Arias et al., 2025), and personal strengths (Bahamón et al., 2023) are considered protective against suicidal behavior, certain factors in academic life require attention, as their unmitigated manifestation could increase suicide risk. Perceived loneliness, academic fatigue, and functional problems warrant study due to their established connections and require more in-depth analysis.

Firstly, loneliness is understood as a subjective experience of social relationships failing to meet the need for interpersonal connection or closeness (Hawkley & Cacioppo, 2010). Unlike isolation or abandonment “being alone”, loneliness involves the distress of “feeling alone” even when maintaining social relationships or support networks. This experience is prevalent among university students, often reported at high intensities (Conti et al., 2023), and is generally associated with a lack of university identity (Zhou et al., 2023). Regarding suicide-related issues, moderate positive correlations with suicide risk are reported (Jin et al., 2025), particularly with suicidal ideation (Moreta-Herrera et al., 2026; Oh et al., 2025), and it is even considered a positive predictor (Hamdan-Mansour et al., 2024).

Alongside loneliness is academic emotional fatigue, which refers to the mental exhaustion stemming from prolonged study demands, performance pressure, and academic overload (Schaufeli et al., 2002)—that is, feeling overwhelmed by academic tasks (Campos et al., 2005; Wu et al., 2023). Role ambiguity, conflicts, high pressure, and competitiveness are factors associated with this fatigue (Labrague & Ballad, 2021), which is often pronounced in university students. Its intensity can vary depending on contextual factors (Moreta-Herrera et al., 2022), negatively affecting mental health and potentially leading to self-harm and suicidal behaviors (Lin et al., 2021). In this regard, evidence shows significant positive relationships with suicide risk (Aquino-Canchari et al., 2025; Kleiman et al., 2018), and it appears to act as a positive predictor (Noyosase Igbineweka et al., 2025).

Finally, social dysfunction refers to difficulties in adequately performing everyday roles and normal daily activities. It is conceived as an indicator of the multidimensional perception of psychological distress (Goldberg, 1978). While its presence among university students is usually low (Affan et al., 2025; Yus et al., 2025), this element could be key for adequate academic performance, although available evidence is rather limited. Regarding its relationship with suicide risk, specific information on this construct is lacking; however, multidimensional studies of psychological distress show mild positive correlations (Nowakowska-Domagała et al., 2023), without estimating predictive evidence.

Summarizing the findings, loneliness, emotional fatigue, and social dysfunction are psychosocial factors present in the daily dynamics of the university population that impact psychological sensitivity. The evidence of how these interact with suicide risk provides a clear warning for decision-making, particularly concerning the reinforcement of prevention strategies and mental health support to minimize the overall risk of suicidal behavior.

The Present Study

While loneliness, emotional fatigue, and social dysfunction are variables connected to suicide risk among university students, most studies have examined them in isolation (focusing primarily on correlations). This provides a narrow and unrealistic view of their interaction. Thus, it is necessary to explore the interconnected triggering factors that operate jointly to provide a comprehensive view of suicide risk. Studies that yield explanatory models of these multiple connections would offer further insight into the mechanisms of risk through everyday, non-clinical factors present in students’ daily lives.

Proposing explanatory models helps to elucidate the mechanisms of action of various predictor variables on a criterion variable, particularly by estimating direct and indirect effects through the inclusion of mediating variables (Gunzler et al., 2013; Lange et al., 2021). Among the variables of interest, loneliness is the most prominent, showing the strongest evidence of relevant covariance and predictive ability for suicide risk in university students. Furthermore, it shares a sustained correlation with emotional fatigue (Kadanji et al., 2025; Shim & Go, 2025) and social dysfunction (Tsiridis et al., 2023). This suggests that loneliness likely exerts not only a direct effect on suicide risk, as previously established (Hamdan-Mansour et al., 2024), but also independent indirect effects via emotional fatigue and social dysfunction. There may even be an unexplored sequential indirect effect, given the positive correlation between emotional fatigue and social dysfunction (Theofilou et al., 2022).

Consequently, developing an explanatory model of suicide risk that identifies loneliness, emotional fatigue, and social dysfunction as multiple predictors—while simultaneously assessing both direct and indirect effects (see Figure 1)—would enrich existing knowledge of suicidal behavior based on daily psychosocial factors. This approach allows for a more comprehensive examination of suicide risk without requiring the presence of significant clinical indicators of mental distress, such as symptoms of anxiety, depression, or stress. University students generally face particular, everyday stressors that do not always initially manifest as mental health disorders; thus, by examining these factors from a transdiagnostic perspective, mechanisms related to mental health deterioration and even suicide risk can be identified. This can help establish a framework for interpreting suicide risk from a non-pathological or non-clinical perspective, contributing ideas, input, and resources to preventive intervention processes or to the identification of risk in subclinical stages.

Figure 1. Explanatory predictor and multiple mediation model

Therefore, based on these premises, the following research objectives are proposed: a) to examine the latent relationships between loneliness, academic emotional fatigue, social dysfunction, and suicide risk in a sample of university students; b) to determine the predictive effect of loneliness, emotional fatigue, and social dysfunction on suicide risk through a structural multiple predictor model; and c) to evaluate the latent mediating role of academic emotional fatigue and social dysfunction in the relationship between loneliness and suicide risk. Based on these objectives, it is hypothesized that loneliness will be positively associated with academic emotional fatigue, social dysfunction, and suicide risk (H₁); that loneliness, academic emotional fatigue, and social dysfunction will jointly predict suicide risk (H₂); and that academic emotional fatigue and social dysfunction will mediate the relationship between loneliness and suicide risk (H₃).

Method

Design

The study employed a descriptive, predictive, multiple mediation, and cross-sectional design (Ato et al., 2013) using structural equation modelling (SEM) techniques in a sample of Ecuadorian university students.

The study included 943 participants (64.4% women, 35.6% men) aged 18 to 47 years (M = 21.62, SD = 3.81); 89.1% are between 18 and 25 years old (typical ages for university studies), and 10.1% are over 25 years old (less common ages for undergraduate university studies). Ethnically, 89.3% identified as mestizo, while the remaining 10.7% identified as indigenous, rural residents, or Afro-Ecuadorian. Additionally, 73.3% resided in urban areas and 26.4% in rural areas. The undergraduate students came from six universities (61.1% public) in four Ecuadorian cities. Furthermore, 37.2% combined their studies with work, and 23% reported being at risk of poor academic performance. A non-probabilistic convenience sampling method was used based on the following inclusion criteria: being of legal age and providing informed consent, voluntary participation, and being officially enrolled at the participating universities.

Measures

University of California Los Angeles Loneliness Scale (UCLA-LS; Russell et al., 1978) in its 10-item Spanish version (Velarde-Mayol et al., 2016). This measure assesses subjective perceived loneliness. It consists of 10 items rated on a 5-point Likert scale ranging from 0 (never) to 4 (always). Although it lacks normative interpretative values, higher scores indicate a greater perception of loneliness. Regarding its psychometric properties, the Spanish version of the UCLA-LS (Velarde-Mayol et al., 2016) has a unidimensional structure and demonstrated high internal consistency, α = .95.

Emotional Exhaustion Scale (ECE, for its acronym in Spanish; Campos et al., 2005), adapted for Ecuadorian university students (Moreta-Herrera et al., 2022). This scale evaluates effective academic emotional fatigue arising from academic demands. It comprises 10 items rated on a 5-point Likert scale from 0 (never) to 4 (always). Higher scores denote greater emotional fatigue. The Ecuadorian version of the ECE (Moreta-Herrera et al., 2022) presents a unidimensional structure with high internal consistency, ω > .95.

The Social Dysfunction dimension of the 28-item General Health Questionnaire (GHQ-28; Goldberg, 1978), validated in Ecuadorian university students (Moreta-Herrera, Rodríguez-Lorenzana, et al., 2024), assesses the functional impact of psychological distress on an individual’s daily functioning and social environment. It includes seven items rated on a 4-point Likert scale, with higher scores reflecting greater social dysfunction. This dimension in the Ecuadorian version showed high internal consistency, ω = .98 (Moreta-Herrera, Rodríguez-Lorenzana, et al., 2024).

Paykel Suicide Scale (Paykel et al., 1974), translated into Spanish (Fonseca-Pedrero et al., 2018) and adapted for Peru (Baños-Chaparro & Ramos-Vera, 2020), analyzes the continuum of suicidal behavior, focusing primarily on suicidal ideation and planning over the past six months. It consists of five items rated on a frequency scale ranging from 0 (never) to 4 (always). Higher scores indicate a greater suicide risk. The Peruvian version (Baños-Chaparro & Ramos-Vera, 2020) used in the current study presents a unidimensional structure with correlated errors for items 4 and 5, as a portion of the variance is not explained by the latent factor. Internal consistency was acceptable, ω = .82.

Procedure

Initially, institutional permissions were obtained from the participating study centers, followed by informed consent from the potential sample. Data collection was then conducted virtually using Google Forms. Prior to the assessment, participants read the consent forms detailing the study’s objectives, their roles, and relevant ethical safeguards, before voluntarily agreeing to participate. The evaluation process lasted approximately 15 minutes and took place at the participating institutions during the first semester of 2025. Upon completion, data were cleaned and systematized in electronic databases for analysis and reporting. Regarding ethical considerations, this study complied with the Declaration of Helsinki guidelines for human research. Furthermore, the overarching project adhered to institutional ethics regulations and received approval from the Institutional Review Board of the Pontificia Universidad Católica del Ecuador, Ambato campus.

Data Analysis

Data analysis was conducted in four stages. The first involved a descriptive analysis of the variables, reporting the mean (M), standard deviation (SD), skewness (g₁), and kurtosis (g₂), along with the measurement range indicating the mean’s relative position against the test’s maximum and minimum scores. Univariate normality was assessed using g₁ and g₂ values, expecting margins of ~1.5. Multivariate normality was evaluated via Mardia’s test (1970), indicated by a lack of statistical significance (p > .05).

The second stage examined the factorial validity of the measures using Confirmatory Factor Analysis (CFA) based on a polychoric correlation matrix with Diagonally Weighted Least Squares (DWLS) estimation. This estimator was chosen due to the ordinal nature of the items and the lack of multivariate normality (Li, 2016). Factorial validity was verified using the Chi-square (χ²), Standardized Root Mean Square Residual (SRMR), Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), and Root Mean Square Error of Approximation (RMSEA). Adequate validity was determined by a non-significant χ² (p > .05), CFI and TLI > .95, SRMR and RMSEA < .08, and factor loadings > .50 (Brown, 2015; Byrne, 2008; Liu et al., 2023; Moreno-Montero et al., 2023; Wolf et al., 2013). Item internal consistency was calculated using McDonald’s coefficient (ω; McDonald, 1970), with values above .70 considered adequate.

Following this, structural models were evaluated. The first involved a latent relationship model using SEM techniques to determine the intensity of covariances, expecting at least mild latent relationships (r > .20). The second model was a structural double mediation model, comprising loneliness as the exogenous variable (X), suicide risk as the endogenous variable (Y), and emotional fatigue (M1) and social dysfunction (M2) as mediators. This model analyzed the direct effect (c′) of loneliness on suicide risk, the independent indirect effects of emotional fatigue (a1b1) and social dysfunction (a2b2), and the sequential effect of emotional fatigue and social dysfunction (adb) on this relationship. Independent and sequential mediation is supported when the indirect effects (a1b1, a2b2, and adb) are significant (p < .05). If the direct effect (c′) ceases to be significant (p>.05) after including the sequential indirect effect (adb), the mediation is considered total; otherwise, it is considered partial (Baron & Kenny, 1986; Lange et al., 2021; Moreta-Herrera et al., 2023). Finally, the mediated proportion of the model (adb/c) was calculated to determine the percentage of the total effect accounted for by mediation. Additionally, as a regression-based model, the structural coefficient of determination (R²) was calculated to ascertain the percentage of variance in suicide risk explained by loneliness, emotional fatigue, and social dysfunction. A predictability of > .20 was expected for it to be considered moderate (Ferguson, 2016).

Statistical analyses were conducted using R software version 4.2.2 (R Core Team, 2023) within the RStudio environment, utilizing the foreign, lavaan, MVN, and MBESS packages.

Results

Preliminary Analysis of the Measures

Table 1 displays the general performance of the variables of interest. Emotional fatigue is the most prominent attribute, reaching a moderate intensity among participants. This is followed by perceived loneliness at a low-to-moderate intensity, and finally, social dysfunction and suicide risk at low intensities. Since the distribution values fall within the ~1.5 range, univariate normality is assumed; however, the assumption of multivariate normality is not met due to significant results (p < .05) in Mardia’s test applied to skewness (g₁) and kurtosis (g₂).

Table 1

Preliminary Analysis of the Measures

Measures	M	Range	SD	g1	g2	Mardia g1	Mardia g2
UCLA	15.84	39.6%	8.50	0.22	-0.25	1321.24***	54.64***
ECE	21.03	52.6%	9.58	0.04	-0.64	1518.16***	65.53***
Social Dysfunction	14.31	26.1%	3.86	0.26	-0.02	747.94***	37.9***
Paykel	4.33	17.3%	4.89	1.35	1.24	2070,34***	60.39***

Notes. *** p < .001; M: mean; SD: standard deviation; g1: skewness; g2: kurtosis

Factorial Validity of the Measures

Table 2 presents the results of the CFA applied to each measure, a preliminary step before conducting structural analyses of latent relationships (general fit model) and predictions (structural regression and mediation analysis). The study’s measures for loneliness (UCLA-LS), emotional fatigue (ECE), social dysfunction (with correlated errors for items 6 and 7), and suicide risk (PSS, with correlated errors for items 4 and 5) show optimal fit indices for their theoretical models. This indicates that the measures are empirically consistent with the data and, therefore, valid for use among Ecuadorian university students. Furthermore, regarding internal consistency, all obtained values exceed the reference score (ω > .70), indicating high reliability for all measures.

Table 2

Confirmatory Factor Analysis of the Loneliness, Emotional Fatigue, Social Dysfunction, and Suicide Risk Measures

Measures	χ2	gl	CFI	TLI	RMSEA	SRMR	ω
UCLA	312.34***	35	.995	.994	.092 [.083 - .101]	.044	.942
ECE	267.31***	35	.998	.998	.084 [.075 - .094]	.035	.960
Social Dysfunction	106.31***	13	.994	.991	.087 [.072 - .103]	.045	.799
Paykel	27.56***	4	.999	.999	.079 [.053 - .108]	.021	.929

Notes. χ²: chi-square; df: degrees of freedom; CFI: comparative fit index; TLI: Tucker-Lewis index; RMSEA: root mean square error of approximation; SRMR: standardized root mean square residual. *** p < .001

Latent Relationship Analysis

Figure 2 displays the general fit model of the variables of interest for estimating their latent relationships. All variables show positive and moderately intense covariances. The strongest association is between loneliness and suicide risk, whereas the weakest is between emotional fatigue and suicide risk.

Figure 2. General fit model between suicide risk, loneliness, academic fatigue, and social dysfunction

Notes. χ²= 1980.56; p < .001; df= 456; CFI= .995; TLI= .994; RMSEA= .060 [.057 - .062]; SRMR= .050.  χ² = chi-square; df = degrees of freedom; CFI = comparative fit index; TLI = Tucker-Lewis index; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual.

Predictor and Mediation Model

Figure 3 illustrates the predictive model of suicide risk based on loneliness, emotional fatigue, and social dysfunction. In the latent structural regression model, loneliness, emotional fatigue, and social dysfunction explain 60% of the variance in suicide risk, representing a large effect. Thus, these three elements are considered strong positive predictors of suicide risk.

Regarding the multiple mediation analysis, loneliness exerts a substantial direct effect (c) on suicide risk and also presents specific indirect effects through a) emotional fatigue (M1); b) social dysfunction (M2), which has the larger effect; and c) the combination of both mediators, representing a sequential indirect effect. The total of these three indirect effects (adb) is .20, p < .001, considered moderate. Moreover, given that the direct effect (c′) is also statistically significant (p < .05) after incorporating the indirect effects (abd), this indicates that the mediation is not complete, but only partial. Furthermore, the mediated proportion of the model indicates that 35.7% of the total effect is mediated. This implies that a portion of the relationship between loneliness and suicide risk is explained by intermediate psychological mechanisms.

Figure 3. Mediation model between loneliness and suicide risk through emotional fatigue and social dysfunction

Notes. χ²= 1980.56; p < .001; df= 456; CFI= .995; TLI= .994; RMSEA= .060 [.057 - .062]; SRMR= .050.  χ² = chi-square; df = degrees of freedom; CFI = comparative fit index; TLI = Tucker-Lewis index; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual.

Regarding the fit indices (general fit model, structural regression, and mediation), the proposed models for latent relationships and mediation yield adequate fit values. This indicates that the discrepancy error between the theoretical models and the data is relatively small, making them empirically consistent.

Discussion

This study aimed to determine the relationship between loneliness, emotional fatigue, social dysfunction, and suicide risk in a sample of Ecuadorian university students. Furthermore, it sought to establish their joint predictive potential for suicide risk, along with the direct and indirect effects of loneliness on this condition.

Regarding the prevalence of the variables of interest, emotional fatigue was the most prominent condition among the students, demonstrating moderate intensity. This condition is widely reported among participants, aligning with previous research on university samples (Labrague & Ballad, 2021; Moreta-Herrera et al., 2022; Wu et al., 2023). It appears to be an inherent characteristic of their adaptation to adulthood, academic life associated with perceived overload, conflicts, and academic pressure, warranting further investigation. This was followed by perceived loneliness, reported at a slightly lower intensity than emotional fatigue. The data indicate that approximately 40% of cases experience some degree of loneliness or limited interpersonal closeness within their academic context. It is inferred that “feeling lonely” forms a significant part of the daily university experience at some stage. While the intensity reported in this study is lower than in some previous research showing elevated levels (Conti et al., 2023; Zhou et al., 2023), similar data in Ecuador are scarce, necessitating further research to expand upon these dynamics. Social dysfunction showed a low prevalence, indicating that only a small proportion of participants report difficulties in adequately and effectively performing everyday roles and daily activities. These results are consistent with previous studies reporting low levels (Affan et al., 2025; Yus et al., 2025). Although not characteristic of the university population, it requires monitoring as an independent construct.

Finally, suicide risk presented at a low intensity, although its prevalence remains notable (~17%) for this demographic. This indicates that the condition affects a significant proportion of students and elevates the probability of more severe suicidal behaviors (e.g., suicidal planning and execution). These findings align with previous reports (Crispim et al., 2021; Oleas et al., 2025; Ward et al., 2022), highlighting an issue that demands further study and intervention mechanisms, particularly given the academic training context of this population.

Regarding the latent relationships, suicide risk correlated moderately and positively with loneliness, consistent with prior national and international research (Moreta-Herrera et al., 2026; Oh et al., 2025). Similarly, it correlated with academic emotional fatigue in the same direction and magnitude, aligning with previous findings (Aquino-Canchari et al., 2025; Kleiman et al., 2018). Lastly, it correlated moderately and positively with social dysfunction—albeit measured as a dimension of a broader construct (psychological distress)—which corresponds with previous reports (Affan et al., 2025; Yus et al., 2025). In summary, loneliness (showing the strongest relational proportion), emotional fatigue, and social dysfunction are consistently connected to suicide risk at a moderate intensity. This is particularly relevant due to their shared variance.

Unlike much of the available evidence, this study examines these three variables concurrently, providing a broader perspective on their interconnectivity. This is illustrated by the latent relationship model, which demonstrates coherence between the proposed theoretical model (see Figure 1) and the empirical data.

In the multiple latent prediction model, loneliness, emotional fatigue, and social dysfunction emerged as positive latent predictors of suicide risk. Together, these attributes explain 60% of the variance in suicide risk, representing a large effect (Ferguson, 2016). Given the complexity of this explanatory model, there are no prior studies for direct comparison; however, the predictability of loneliness (Hamdan-Mansour et al., 2024) and emotional fatigue (Noyosase Igbineweka et al., 2025) aligns with previous conclusions. The inclusion of social dysfunction demonstrates a novel and relevant contribution to the multifactorial understanding of suicide risk.

Delving into the specific effects, the latent multiple mediation analysis revealed that loneliness exerts a direct effect on suicide risk, alongside indirect effects via emotional fatigue and social dysfunction. A sequential effect was also observed through emotional fatigue and social dysfunction, which share covariance (Theofilou et al., 2022). These findings highlight the complex mechanisms of loneliness, suggesting that mitigating suicide risk by targeting loneliness in isolation—without addressing intermediate variables like emotional fatigue and social dysfunction—may be challenging. Furthermore, the sequential mediation indicates a probable “cascade” effect, where loneliness contributes to emotional fatigue, which in turn impairs social functioning, culminating in increased suicide risk. The significant direct effect (c′, p < .05) indicates partial mediation. Therefore, both emotional fatigue and social dysfunction act as latent mediators in the relationship between loneliness and suicide risk.

Finally, the latent relationship and multiple latent mediation models demonstrated adequate fit indices (Brown, 2015; Byrne, 2008; Moreno-Montero et al., 2023; Wolf et al., 2013). This indicates coherence between the theoretical assumptions and the empirical data, allowing for generalization to reference samples of Ecuadorian university students. Notably, studies with these specific characteristics have not been conducted previously; thus, this pioneering work advances the understanding of psychosocial factors associated with suicide risk in the university population using everyday indicators of academic life. However, due to the lack of comparative literature, these findings should be interpreted with caution.

Regarding its implications, this study provides a novel explanatory model for suicide risk associated with loneliness, emotional fatigue, and social dysfunction. This yields valuable insights into understanding suicide risk through subclinical, everyday factors. Crucially, it demonstrates that intermediate psychological mechanisms explain a significant portion of the relationship between loneliness and suicide risk (c/adb = 35.7%), a finding of substantial explanatory relevance for this line of research. Consequently, this provides a practical foundation for formulating psychological intervention strategies targeting suicide risk, particularly for preventive action in subclinical and prodromal phases before severe suicidal manifestations consolidate. Interventions could therefore focus on cognitive strategies (Alvarado-Zurita et al., 2025), emotional regulation (Rodas et al., 2022), and social support based on these indicators to prevent suicide.

Conclusions

Loneliness, emotional fatigue, and social dysfunction are positive predictors of suicide risk among university students in Ecuador. Together, these factors explain 60% of the change in the variance of suicide risk. Furthermore, considering the effects, loneliness has both a direct effect on suicide risk and an indirect effect sequentially through emotional fatigue and social dysfunction, which suggests that the relationship between loneliness and risk is explained by intervening psychological phenomena.

Limitations

The study presents several limitations that should inform future research. Firstly, while loneliness, emotional fatigue, and social dysfunction were identified as predictors of suicide risk, the cross-sectional design precludes causal conclusions. Future research should employ longitudinal predictive and mediational designs with at least three time points, or experimental methodologies. Secondly, the sample was restricted to university students, excluding other educational levels, such as secondary school students. Consequently, the findings are limited to this specific demographic. Future studies should encompass diverse educational groups, as well as academic staff and researchers. Within this same context, it is important to note that the participants have a fairly broad age range, and 10% of them are of ages less common in undergraduate university studies. This could potentially influence the prevention of suicidal behavior. Therefore, it is recommended that future studies explore strategies to control this variable or, failing that, to segment the sample. Finally, regarding the explanation of the variance in the predictive model, although an R² of 60% does not constitute direct evidence of variance in the common method, the use of self-report measures could increase this risk. Consequently, it is recommended to evaluate and report control procedures, such as the Harman one-factor test or analysis using marker variables. Furthermore, although the model demonstrated high explanatory power, there is insufficient evidence of multicollinearity; future research could complement the analysis by evaluating potential effects (through variance inflation factors [VIF], tolerance indices, or redundancy analysis). This would examine possible overlaps between predictors.

Ethics Statement

The research was approved by the Institutional Review Board of the Pontificia Universidad Católica del Ecuador Sede Ambato, based on the ethical guidelines of the institutional regulations and the Helsinki Convention for research involving human subjects.

Author Contributions

Conceptualization: RMH and PGV; Data curation: RMH and ECA; Formal analysis: RMH and JAR; Funding acquisition: RMH; Investigation: RMH, PRS and ECA; Methodology: RMH and JAR; Project administration: RMH; Resources: RMH; Software: RMH and JAR; Supervision: PGV; Validation: PGV, ECA and PRS; Visualization: RMH and ECA; Writing – original draft: RMH and PRS; Writing – review & editing: RMH, PGV, JAR, ECA, PRS.

Conflict of Interest

The authors have declared that they have no competing or potential conflicts of interest.

Funding

This essay was funded by the Pontificia Universidad Católica del Ecuador Sede Ambato, through the project “Predictores cognitivos y emocionales de la salud mental positiva y negativa en población universitaria del Ecuador” with resolution CD-RES-098-2024.

Data availability statement

The data associated with these results are available to those interested and can be requested from the corresponding author of the study. It is important to note that the use of the same will be used solely for research and academic purposes and under no pretext for commercial purposes.

References

Affan, R., Darwich, K., Jarrouge, G., Hannouch, T., & Assaad, S. (2025). Mental health status among university students during the transition period from high school to university in Lebanon: a cross-sectional study. BMC Public Health, 25(1), 1848. https://doi.org/10.1186/s12889-025-23087-3

Alvarado-Zurita, P.D., Rodas, J.A., Aguilar, M.J., & Alvarez-Gutierrez, D.M. (2025). Impacto afectivo de entrenamientos cognitivos en personas con deterioro cognitivo [Affective impact of cognitive training in people with cognitive impairment]. Liberabit, 31(2), e1220-1. https://doi.org/10.24265/liberabit.2025.v31n2.1220

Aquino-Canchari, C.R., Florez-Soto, D.M., Lorenzo-Roque, A.I., Baquerizo-Quispe, N.S., & Cortez-Orellana, S.A. (2025). Exaustão emocional associada a pensamentos e comportamentos autolesivos e ideação suicida em estudantes de medicina de uma universidade peruana [Emotional exhaustion associated with self-harming thoughts and behaviors and suicidal ideation in medical students at a Peruvian university]. Scientia Medica, 35(1), e46519. https://doi.org/10.15448/1980-6108.2025.1.46519

Ato, M., López, J.J., & Benavente, A. (2013). Un sistema de clasificación de los diseños de investigación en psicología. [A classification system for research designs in psychology.] Anales de Psicologia, 29(3), 1038–1059. https://doi.org/10.6018/analesps.29.3.178511

Bahamón, M.J., Javela, J.J., Vinaccia, S., Matar-Khalil, S., Cabezas-Corcione, A., & Cuesta, E.E. (2023). Risk and protective factors in Ecuadorian adolescent survivors of suicide. Children, 10(3), 549. https://doi.org/10.3390/children10030549

Baños-Chaparro, J., & Ramos-Vera, C. (2020). Validez e invarianza según sexo y edad de la Escala Paykel de Ideación Suicida en adolescentes peruanos [Validity and invariance according to sex and age of the Paykel Scale of Suicidal Ideation in Peruvian adolescents]. Interacciones, 6(1), e225. https://doi.org/10.24016/2020.V6N1.225

Baron, R.M., & Kenny, D.A. (1986). The moderator-mediator variable distinction in social psychological research. conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173–1182. https://doi.org/10.1037/0022-3514.51.6.1173

Brown, T.A. (2015). Confirmatory factor analysis for applied research (2nd ed). The Guilford Press.

Byrne, B.M. (2008). Testing for multigroup equivalence of a measuring instrument: A walk through the process. Psicothema, 20(4), 872–882.

Campos, F.R., Rodriguez, D.M., & Astorga, C.M. (2005). Escala de Cansancio Emocional (ECE) para Estudiantes Universitarios: Propiedades Psicométricas y Asociación [Emotional exhaustion scale for university students: Psychometric properties and association]. Intersiquis, 1-9.

Castillo-Gonzáles, M., León-del-Barco, B., Mendo-Lázaro, S., & Terán-Andrade, E. (2026). Dating violence, emotional dependence and mental health problems in Ecuadorian college students. Healthcare, 14(7), 834. https://doi.org/10.3390/HEALTHCARE14070834

Conti, C., Lanzara, R., Rosa, I., Müller, M. M., & Porcelli, P. (2023). Psychological correlates of perceived loneliness in college students before and during the COVID-19 stay-at-home period: A longitudinal study. BMC Psychology, 11(1), 60. https://doi.org/10.1186/S40359-023-01099-1

Crispim, M.O., dos Santos, C.M.R., Frazão, I.S., de Queiroz Frazão, C.M.F., de Albuquerque, R.C.R., & Perrelli, J.G.A. (2021). Prevalence of suicidal behavior in young university students: A systematic review with meta-analysis. Revista Latino-Americana de Enfermagem, 29, e3495. https://doi.org/10.1590/1518-8345.5320.3495

Ferguson, C.J. (2016). An effect size primer: A guide for clinicians and researchers. In A. E. Kazdin (Ed.), Methodological issues and strategies in clinical research (4th ed., pp. 301–310). American Psychological Association. https://doi.org/10.1037/14805-020

Fonseca-Pedrero, E., Inchausti, F., Pérez-Gutiérrez, L., Aritio Solana, R., Ortuño-Sierra, J., Sánchez-García, M.a.Á., Lucas-Molina, B., Domínguez, C., Foncea, D., Espinosa, V., Gorría, A., Urbiola-Merina, E., Fernández, M., Merina Díaz, C., Gutiérrez, C., Aures, M., Campos, M.S., Domínguez-Garrido, E., & Pérez de Albéniz Iturriaga, A. (2018). Ideación suicida en una muestra representativa de adolescentes españoles. Revista de Psiquiatría  y Salud Mental, 11(2), 76–85. https://doi.org/10.1016/J.RPSM.2017.07.004

Goldberg, D.P. (1978). Manual of the General Health Questionnaire. NFER-NELSON Publishers.

Gunzler, D., Chen, T., Wu, P., & Zhang, H. (2013). Introduction to mediation analysis with structural equation modeling. Shanghai Archives of Psychiatry, 25(6), 390–394.

Gutiérrez, I., Antonio, J., Rico, Q., Gutiérrez Pastor, A., García, R.N., Concepción, M., & Munuera, C. (2021). Depresión, ansiedad y salud autopercibida en estudiantes de Medicina: un estudio transversal. Revista Española de Educación Médica, 2(2), 21–31. https://doi.org/10.6018/edumed.470371

Hamdan-Mansour, A.M., Hamdan-Mansour, R.A., Allaham, D.M., Alrashidi, M., Alhaiti, A., & Mansour, L.A.H. (2024). Academic procrastination, loneliness, and academic anxiety as predictors of suicidality among university students. International Journal of Mental Health Nursing, 33(6), 2054–2062. https://doi.org/10.1111/inm.13366

Hawkley, L.C., & Cacioppo, J.T. (2010). Loneliness matters: A theoretical and empirical review of consequences and mechanisms. Annals of Behavioral Medicine, 40(2), 218–227. https://doi.org/10.1007/s12160-010-9210-8

Jin, Y., Kieslich, K., Hall, A., & Pitman, A. (2025). Association between loneliness and suicidal thoughts and attempts in adolescence and young adulthood: A longitudinal analysis of a nationally representative US sample. Archives of Suicide Research, 30(1), 66–79. https://doi.org/10.1080/13811118.2025.2489159

Kadanji, S.A., Hooman, F., & Khabir, L. (2025). Predicting academic burnout based on loneliness, distress tolerance, and alexithymia in students. Educational Research in Medical Sciences, 14(2), 1. https://doi.org/10.5812/ermsj-162750

Kleiman, E.M., Turner, B.J., Chapman, A.L., & Nock, M.K. (2018). Fatigue moderates the relationship between perceived stress and suicidal ideation: Evidence from two high-resolution studies. Journal of Clinical Child and Adolescent Psychology, 47(1), 116–130. https://doi.org/10.1080/15374416.2017.1342543

Labrague, L.J., & Ballad, C.A. (2021). Lockdown fatigue among college students during the COVID-19 pandemic: Predictive role of personal resilience, coping behaviors, and health. Perspectives in Psychiatric Care, 57(4), 1905–1912. https://doi.org/10.1111/ppc.12765

Lange, T., Hansen, K.W., Sørensen, R., & Galatius, S. (2021). Applied mediation analyses: A review and tutorial. Epidemiology and Health, 39, e2017035. https://doi.org/10.4178/epih.e2017035

Lapo-Talledo, G.J., Montes-Escobar, K., Rodrigues, A.L.S., & Siteneski, A. (2025). Hospitalizations for depressive disorders in Ecuador: An eight-year nationwide analysis of trends and demographic insights. Journal of Affective Disorders, 374, 433–442. https://doi.org/10.1016/J.JAD.2025.01.074

Lascano-Arias, G., Cuesta-Andaluz, E., & Espinosa-Pinos, C.A. (2025). Suicide risk and social support in young Ecuadorian women victims of violence: A psychosocial and educational analysis. Journal of Educational and Social Research, 15(4), 22. https://doi.org/10.36941/jesr-2025-0119

Li, C.H. (2016). Confirmatory factor analysis with ordinal data: Comparing robust maximum likelihood and diagonally weighted least squares. Behavior Research Methods, 48(3), 936–949. https://doi.org/10.3758/s13428-015-0619-7

Lin, H.C., Li, M., Stevens, C., Pinder-Amaker, S., Chen, J.A., & Liu, C.H. (2021). Self-Harm and suicidality in US college students: Associations with emotional exhaustion versus multiple psychiatric symptoms. Journal of Affective Disorders, 280, 345–353. https://doi.org/10.1016/j.jad.2020.11.014

Liu, X., Lao, C.K., Shi, H., & Zhou, G. (2023). Stigma and emotional distress in Chinese mental health professionals: Moderating role of cognitive fusion. Stigma and Health, 9(2), 201–211. https://doi.org/10.1037/SAH0000479

Mardia, K.V. (1970). Measures of multivariate skewness and kurtosis with applications. Biometrika, 57(3), 519. https://doi.org/10.1093/biomet/57.3.519

McDonald, R.P. (1970). The theoretical foundations of principal factor analysis, canonical factor analysis, and alpha factor analysis. British Journal of Mathematical and Statistical Psychology, 23(1). https://doi.org/10.1111/j.2044-8317.1970.tb00432.x

Ministerio de Salud Pública del Ecuador [MSP]. (2023, July 23). MSP fortalece la Salud Mental Comunitaria e implementa estrategias para prevención del suicidio – Ministerio de Salud Pública [MSP strengthens Community Mental Health and implements strategies for suicide prevention – Ministry of Public Health.]. https://www.salud.gob.ec/msp-fortalece-la-salud-mental-comunitaria-e-implementa-estrategias-para-pre...

Moreno-Montero, E. ., Moreta-Herrera, R., & Suárez-López, A. (2023). Psychometric properties of the Cognitive Fusion Questionnaire (CFQ) from a sample of Ecuadorian college students. PSICOLOGIA, 37(2), 80–90. https://doi.org/10.17575/psicologia.1871

Moreta-Herrera, R., Gordón-Villalba, P., Rodas, J.A., Mejía, F., Mantilla, C., Ruíz, Á., Mayorga-Ortíz, W., & Mayorga-Ortíz, D. (2026). Cognitive predictors of suicidal ideation in Ecuadorian medical and nursing students: The roles of perceived loneliness and life satisfaction. Revista Colombiana de Psicología, 35(1), 47–62. https://doi.org/10.15446/RCP.V35N1.117861

Moreta-Herrera, R., Jadán-Guerrero, J., Gordón-Villalba, P., Mayorga-Lascano, M., Shugulí-Zambrano, C., Caycho-Rodríguez, T., Cuesta-Andaluz, E., & Larzabal-Fernández, A. (2025). Satisfaction with life and suicidal ideation among Ecuadorian university students: A network analysis. International Journal of Applied Positive Psychology, 10(3), 1–16. https://doi.org/10.1007/s41042-025-00242-0

Moreta-Herrera, R., Rodríguez-Lorenzana, A., Mascialino, G., Castro-Ochoa, F., Narváez-Pillco, V., Caycho-Rodríguez, T., & Mayorga-Lascano, M. (2024). Psychometric properties of the 28-item General Health Scale (GHQ-28). Analysis from the Classical Test and Item Response Theories in Ecuadorian university students. Psychology Hub, 41(3), 43–54. https://doi.org/10.13133/2724-2943/18460

Moreta-Herrera, R., Vaca-Quintana, D., Quistgaard-Álvarez, A., Merlyn-Sacoto, M.F., & Dominguez-Lara, S. (2022). Análisis psicométrico de la Escala de Cansancio Emocional en estudiantes universitarios ecuatorianos durante el brote de COVID-19 [Psychometric analysis of the Emotional Exhaustion Scale in Ecuadorian university students during the COVID-19 outbreak]. Ciencias Psicológicas, 16(1), 2755. https://doi.org/10.22235/cp.v16i1.2755

Moreta-Herrera, R., Zumba-Tello, D., de Frutos-Lucas, J., Llerena-Freire, S., Salinas-Palma, A., & Trucharte-Martínez, A. (2023). The role of negative affects as mediators in the relationship between stress and mental health in Ecuadorian adolescents. Health Psychology Report, 11(3), 241. https://doi.org/10.5114/hpr/163484

Nowakowska-Domagała, K., Podlecka, M., Sadowski, K., Pietras, T., & Mokros, Ł. (2023). The relationship between chronotype, dispositional mindfulness and suicidal ideation among medical students: mediating role of anxiety, insomnia and social dysfunction. Journal of Sleep Research, 32(4), e13823. https://doi.org/10.1111/jsr.13823

Noyosase Igbineweka, M., Edeki, O., Osariemen, &, & Aimuamwenosa, D. (2025). Emotional exhaustion and academic stress as predictors of suicide ideation among undergraduates. Journal of Contemporary Issues and Thought, 15(1), 62–69. https://doi.org/10.37134/jcit.vol15.1.7a.2025

Oh, H., Du, J., Smith, L., & Koyanagi, A. (2025). The synergy between loneliness and cannabis use on suicidal thoughts and behaviors among young adults in higher education. Journal of Substance Use, 30(3), 447–451. https://doi.org/10.1080/14659891.2024.2351011

Oleas, D., Valdevila Figueira, J.A., Valdevila Santiestevan, R., Moreta-Herrera, R., Mascialino, G., & Rodas, J.A. (2025). Psychometric evaluation of the positive and negative suicide ideation inventory in Ecuadorian population. Inquiry (United States), 62. https://doi.org/10.1177/00469580251337665

Paykel, E.S., Myers, J.K., Lindenthal, J.J., & Tanner, J. (1974). Suicidal Feelings in the General Population: A Prevalence Study. British Journal of Psychiatry, 124(582), 460–469. doi:10.1192/bjp.124.5.460

R Core Team. (2023). R: A language and environment for statistical computing (Version 4.2.2) [Computer software]. R Foundation for Statistical Computing. https://www.R-project.org/

Rodas, J.A., Jara-Rizzo, M.F., Greene, C.M., Moreta-Herrera, R., & Oleas, D. (2022). Cognitive emotion regulation strategies and psychological distress during lockdown due to COVID-19. International Journal of Psychology, 57(3), 315–324. https://doi.org/10.1002/ijop.12818

Russell, D., Peplau, L.A., & Ferguson, M.L. (1978). Developing a measure of loneliness. Journal of Personality Assessment, 42(3), 290–294. https://doi.org/10.1207/s15327752jpa4203_11

Santillán-García, N., Rueda-Espinoza, K., Orozco-Moreno, Z., Moreta-Herrera, R., & Rodas, J. A. (2025). El papel mediador de la satisfacción con la vida en la relación entre esperanza y satisfacción académica entre estudiantes universitarios ecuatorianos [The mediating role of life satisfaction in the relationship between hope and academic satisfaction among Ecuadorian university students]. Revista de Psicodidáctica, 30(1), 500154. https://doi.org/10.1016/J.PSICOD.2024.500154

Schaufeli, W.B., Martínez, I.M., Pinto, A.M., Salanova, M., & Barker, A.B. (2002). Burnout and engagement in university students. Journal of Cross-Cultural Psychology, 33(5), 464–481. https://doi.org/10.1177/0022022102033005003

Sheldon, E., Simmonds-Buckley, M., Bone, C., Mascarenhas, T., Chan, N., Wincott, M., Gleeson, H., Sow, K., Hind, D., & Barkham, M. (2021). Prevalence and risk factors for mental health problems in university undergraduate students: A systematic review with meta-analysis. Journal of Affective Disorders, 287, 282–292. https://doi.org/10.1016/j.jad.2021.03.054

Shim, T., & Go, E. (2025). The impact of school burnout on life satisfaction among university students: The mediating effects of loneliness and fear of alienation. Behavioral Sciences, 15(8), 1083. https://doi.org/10.3390/bs15081083

Theofilou, P., Iona, D., & Tzavella, F. (2022). Association between general health, fatigue and social support among health professionals: A cross-sectional study in Greece. Journal of Community Medicine & Health Care, 2(7), 1058. https://doi.org/10.26420/jcommunitymedhealthcare.2022.1055

Tsiridis, A.F., Tsapakis, E.M., Gazi, A., Tsiarea, K., & Fountoulakis, K.N. (2023). Internet use, loneliness, and mental health in Greek students. Hippokratia, 27(2), 31. https://pmc.ncbi.nlm.nih.gov/articles/PMC11268318/

Velarde-Mayol, C., Fragua-Gil, S., & García-de-Cecilia, J.M. (2016). Validación de la escala de soledad de UCLA y perfil social en la población anciana que vive sola [Validation of the UCLA loneliness scale and social profile in the elderly population living alone]. Semergen, 42(3). https://doi.org/10.1016/j.semerg.2015.05.017

Ward, C., McLafferty, M., McLaughlin, J., McHugh, R., McBride, L., Brady, J., Bjourson, A.J., Walsh, C.P., O’Neill, S.M., & Murray, E.K. (2022). Suicidal behaviours and mental health disorders among students commencing college. Psychiatry Research, 307, 114314. https://doi.org/10.1016/j.psychres.2021.114314

Wolf, E.J., Harrington, K.M., Clark, S.L., & Miller, M.W. (2013). Sample size requirements for structural equation models: An evaluation of power, bias, and solution propriety. Educational and Psychological Measurement, 73(6), 913–934. https://doi.org/10.1177/0013164413495237

World Health Organization. (2021). Suicide worldwide in 2019: Global health estimates. https://apps.who.int/iris/rest/bitstreams/1350975/retrieve

Wu, Q., Chi, P., & Zhang, Y. (2023). Association Between Pandemic Fatigue and Subjective Well-Being: The Indirect Role of Emotional Distress and Moderating Role of Self-Compassion. International Journal of Public Health, 68, 1605552. https://doi.org/10.3389/ijph.2023.1605552

Yus, T.M., Kuno, H., Silvalila, M., Zahra, Z., & Muhsin, M. (2025). The correlation of academic achievement index and mental health of medical students. EJournal Kedokteran Indonesia, 13(3), 1. https://doi.org/10.23886/ejki.13.1166.1

Zhou, H.Y., Luo, Y.H., Shi, L.J., & Gong, J. (2023). Exploring psychological and psychosocial correlates of non-suicidal self-injury and suicide in college students using network analysis. Journal of Affective Disorders, 336, 120–125. https://doi.org/10.1016/j.jad.2023.05.089