Physical Fitness and Body Composition as Predictors of Anxiety,Depression,and Stress in University Students

Abstract

University students experience high levels of psychological distress, including anxiety, depression, and stress. Physical fitness and body composition have been proposed as potential protective factors against these conditions. This study aimed to examine the relationship between indicators of physical fitness and body composition with symptoms of anxiety, depression, and stress in Chilean university students, primarily enrolled in sport-related programs. A cross-sectional design was conducted with 281 participants (85.7% male; age = 20.85 ± 2.24 years). Assessments included muscle strength tests, body composition via bioimpedance, and estimated cardiorespiratory fitness (VO₂max). Psychological symptoms were evaluated using the DASS-21 scale. After applying uniform Benjamini–Hochberg false discovery rate correction across 15 predictor–outcome tests (q = 0.05), no significant associations were found between physical fitness or body-composition indicators and anxiety, depression, or stress (all q ≥ 0.45). Descriptively, an inverse association between estimated VO₂max and anxiety was observed, consistent with prior literature, but it did not remain significant after correction. These findings should be interpreted cautiously due to the cross-sectional design, indirect VO₂max estimation, sex imbalance, and sport-enriched sample. Results are exploratory and hypothesis-generating, requiring confirmation in larger, more representative, and pre-registered studies.

Keywords

aerobic capacity body composition mental health physical fitness students

“Under uniform Benjamini–Hochberg FDR correction across all 15 predictor outcome tests, no physical fitness or body-composition indicator was significantly associated with any psychological domain.”

Introduction

Mental health in university students has become more relevant in the scientific literature due to the sustained increase in symptoms of anxiety, depression, and stress in this population.^1,2 The university stage represents a particularly demanding transition period, characterized by academic pressure, vocational uncertainty, changes in social bonds, and in many cases, growing apart from the nuclear family.³ These conditions can create a favorable environment to develop emotional distress^4,5 and noncommunicable diseases.⁶ Studies worldwide have reported the alarming prevalence of psychological symptoms in young university students, exceeding 30% in some cases of anxiety and depression.⁷

In the context of Chile, the situation is no different. The most recent data show a 21.1% increase in mental health problems among young people, with a 27.5% increase in generalized anxiety disorders.⁸ This damage to psychological well-being represents an urgent challenge for educational institutions and public health systems to identify modifiable protective and risk factors that contribute to preventing these clinical conditions.⁹

Among the potential protective factors, physical fitness has garnered attention for its positive influence and benefits on mental health.^10,11 According to the American College of Sports Medicine,¹² physical fitness is a physiological state of well-being that enables daily activities to be performed efficiently and without excessive fatigue, also representing the basis for physical performance. When physical fitness is health-oriented, it includes components such as cardiorespiratory fitness and musculoskeletal strength. Several studies have shown that greater physical fitness and a healthy body composition can act as factors that soften psychological stress and prevent affective disorders.^13,14

In this regard, variables such as maximal oxygen consumption (VO₂max), muscle strength, and a higher percentage of muscle mass have been associated with lower levels of anxiety and depression in university students.¹⁵ However, a high body fat percentage has been linked to an increased risk of mood disturbances,¹⁶ reinforcing the preventive value of maintaining a healthy body composition. Similarly, one study found an inverse relation between variables such as upper extremity strength, VO₂max, and perceived stress,¹⁷ while another (2024) identified a direct relation between aerobic capacity and the perception of psychological health and quality of life in university students from Chile.¹⁸

Despite this growing evidence, the results are still heterogeneous and, in many cases, methodologically limited. There is a lack of integrative studies that simultaneously analyze multiple objective physical indicators as predictors of psychological variables, especially in the Latin American university context. Additionally, few research works have included different statistical approaches,¹⁹ such as generalized linear models (GLM), which allow for analyzing complex relationships between standardized variables and assessing the specific effect of each predictor.²⁰

From this perspective, there is a need to explore whether different components of physical fitness are associated with different levels of psychological distress in young university students from Chile. Likewise, considering the role of body composition may provide evidence that can be applied to design preventive strategies based on physical activity within the university context.

Method

Study Design

This study adopts a quantitative approach and follows a non-experimental, cross-sectional design.

Sample

The sample included 281 undergraduate students from different universities located in the Metropolitan Region of Chile, selected by convenience sampling. Participants were invited to participate voluntarily, with invitations disseminated digitally and in person on campus. Regarding the sample, 85.7% were men and 14.3% were women, aged between 18 and 32 years, with a mean age of 20.85 ± 2.24. Of these, 58.7% were enrolled in degrees related to physical activity and sports, since this subgroup represents the majority of the sample, the results should be interpreted primarily in the context of students with a focus on sports science, and generalizations to other university populations should be made with caution. In anthropometric terms, an average body mass index (BMI) of 25.07 ± 3.73 was identified. The inclusion criteria were as follows: being a regular undergraduate student, being over 18 years of age, agreeing to voluntary participation, and signing the informed consent form. Students who had a medical problem or who were unable to take the tests were excluded; moreover, the students who did not complete all the corresponding evaluations were not considered (n = 41). Because no formal a priori sample-size calculation was performed before recruitment, a sensitivity power analysis was conducted post hoc to characterize the minimum detectable effect. For a multiple regression with four predictors at α = 0.05 and power = 0.80, the analyzable sample of n = 281 provided 80% power to detect a partial f² ≈ 0.028 (partial R² ≈ 0.027) in the full-sample models, but only f² ≈ 0.21 in the female-only subgroup (n = 40), indicating that the study was sufficiently powered only for moderate-to-large overall effects and substantially underpowered for sex-stratified or interaction analyses. The study is therefore best framed as a pilot, exploratory analysis intended to generate hypotheses and effect-size estimates for future, adequately powered, pre-registered replications in more representative university populations.

Variables

Body Composition

(a) Bioimpedance: The Omron-514 digital scale was used to obtain data on body composition, such as body mass and percentage of fat mass.²¹ All measurements were taken in the morning, wearing comfortable and light clothing, in a standing position, with the legs 35°–45° apart and with extended arms at a (90°) right angle, vertical to the body, without bending the elbows, fasting and without having exercised in the previous 3 hours. For each measurement, the monitor was disconnected and reconnected, entering all the required variables again.²¹

(b) BMI: The aforementioned scale was used to obtain body weight data, and height was measured in a bipedal position using a Seca 213 height rod. It was determined by dividing weight by height squared (^kg/m2).²²

Average hand grip strength (HGS): It was evaluated using a Baseline manual dynamometer, model 12-0241,²³ with the subject seated in a chair with a backrest, with the shoulders adducted and without rotation, the elbow flexed at 90°, and forearm and wrist in a neutral position, with both feet resting on the floor, and with the back supported on the backrest. The arm under evaluation is not supported on any surface, and the dynamometer is used in a vertical position. The participant performs a maximal grip strength for 3 seconds, with 1 minute rest between each repetition, performing two attempts.²⁴

Muscle Quality Index (MQI): Calculated by dividing the hand grip strength by the body mass index (BMI).^25,26

Peak isometric strength: the isometric mid-thigh pull protocol was used by means of a WinLaborat load cell, model WLCC01.²⁷ A maximum isometric contraction was performed on a platform for 5 seconds, with a bar anchored to the load cell.²⁸ The maximum number of kilograms reached was recorded.

Estimated VO₂max: The Queen’s College test comprises going up and down on a 41.3 cm high step at a steady pace of 24 cycles per minute for men and 22 cycles per minute for women, using a metronome to maintain cadence. The test lasts for three minutes; after that, the heart rate is recorded using a heart rate monitor.²⁹ From this value, VO₂max is estimated using sex-specific formulas:

(A) Men: VO₂max = 111.33 - (0.42 X HR).

(B) Women: VO₂max = 65.81 - (0.1847 X HR).

It is important to note that the VO₂max obtained through the Queen’s College test is an indirect estimate and not a direct measurement by gas exchange calorimetry (ergospirometry), which is considered the gold standard. The prediction equations used have standard errors of estimation in the range of 10-15%, and their accuracy varies depending on the participant’s fitness level, age, and health status.

Mental health: Depression, Anxiety, and Stress Scale - 21 items (DASS-21).³⁰ DASS-21 is a self-report measure designed to assess symptoms of depression, anxiety, and stress in adults. The scale includes 21 items equally distributed in three subscales: depression, anxiety, and stress. Participants rate how often they have experienced each symptom in the past week on a scale from 0 (does not apply) to 3 (applies a lot, or most of the time). To respond, the respondent must indicate to what extent the sentence presented describes what happened or what he/she felt during the last week.⁹ For each subscale, the following scores are used: Depression: normal 0-4, mild 5-6, moderate 7-10, severe 11-13, and extremely severe 14+. Anxiety: normal 0-3, mild 4, moderate 5-7, severe 8-9, and extremely severe 10+. Stress: normal 0-7, mild 8-9, moderate 10-12, severe 13-16, and extremely severe 17+.³¹

Procedure

The assessments were carried out at the Sports Science Laboratory of the Universidad Andrés Bello (UNAB), under standardized conditions. An appointment was scheduled for each participant individually for the evaluation in a single session; the tests included body composition, physical fitness, and a mental health questionnaire. The measurements were taken by trained professionals.

Statistical Analysis

Data were analyzed using Generalized Linear Models (GLMs) with a negative binomial distribution to examine the association between physical fitness components and domains of psychological distress. This distribution was selected due to the discrete, non-negative, and overdispersed nature of the DASS-21 scores. Overdispersion was assessed by comparing each negative binomial model with its Poisson counterpart using the ratio of residual deviance to degrees of freedom, and the dispersion parameter (θ) was estimated for all models. Full model diagnostics are provided in the Supplementary Material.

Prior to model fitting, multicollinearity among predictors was evaluated using the Variance Inflation Factor (VIF), with values >5 considered moderate and >10 severe. Variable selection was guided by both statistical criteria and theoretical relevance. In cases of conceptual redundancy (e.g., body weight and BMI), the predictor most aligned with the study objectives was retained. Initial diagnostics (Supplemental Table S1) identified severe multicollinearity for Weight, BMI, Muscle Mass (%) and Body Fat (%). Because bioimpedance-derived Muscle Mass (%) and Body Fat (%) are mathematically near-complementary, ensuring unstable and sign-unreliable coefficients when modeled jointly, only one body-composition variable was retained in the final specification. Body Fat (%) was selected a priori as the more theoretically relevant exposure for mental health outcomes, given its consistent association with mood disturbance in the literature. After this reduction, all predictors in the final model showed acceptable VIFs (all <2; Supplemental Table S2). As a robustness check, two further specifications were estimated and are reported in the Supplementary Material: (i) a model retaining Muscle Mass (%) instead of Body Fat (%), and (ii) a model replacing both indicators with a single orthogonal body-composition index derived as the first principal component (PC1) of standardized Body Fat (%) and Muscle Mass (%) (95.99% of variance explained; loadings of opposite sign), which represents an adiposity-versus-muscularity axis uncorrelated with the remaining predictors. Inferences for VO₂max, age and sex were substantively unchanged across all three specifications.

As a sensitivity analysis, linear regression models with heteroskedasticity-consistent (HC3) covariance estimators were also conducted, yielding substantively similar results (see Supplemental Material).

Model fit and explanatory power were assessed using the Akaike Information Criterion (AIC), residual deviance, and the Cox–Snell pseudo-R², allowing for comparative evaluation across anxiety, depression, and stress domains. To address potential sex-related differences in the association between body composition and mental health outcomes, a moderation analysis was performed by including an interaction term between Muscle Mass (%) and sex.

All analyses are reported as exploratory. Because the present study was not pre-registered and the Introduction did not explicitly designate anxiety as the primary outcome with VO₂max as the primary predictor prior to data inspection, we did not designate any single test as a confirmatory primary comparison exempt from multiplicity adjustment. Instead, the full set of 15 predictor–outcome tests (5 predictors × 3 psychological domains) was treated as a single family of exploratory hypotheses, and the Benjamini–Hochberg false discovery rate (FDR) procedure was applied uniformly at q = 0.05.

Under this uniform correction, no predictor–outcome association remained statistically significant. Unadjusted P-values, BH-adjusted q-values and BH significance flags are presented alongside each coefficient in Table 1. Unadjusted P-values are retained in the text and tables to allow readers to evaluate effect-size patterns, but inferential interpretation is based exclusively on the FDR-adjusted results.

Table 1.

Multivariate Negative Binomial Regression Analysis of Physical Fitness Predictors for Psychological Distress Domains.

Predictor	Anxiety		Depression		Stress
Predictor	β (95% CI)	P-value	β (95% CI)	P-value	β (95% CI)	P-value
Body Fat (%)	-0.006 (-0.03, 0.02)	0.615	-0.000 (-0.02, 0.02)	0.985	-0.007 (-0.03, 0.02)	0.553
Muscle Mass (%)	0.001 (-0.01, 0.01)	0.866	0.000 (-0.01, 0.01)	0.984	-0.003 (-0.02, 0.01)	0.582
VO_2max	-0.013 (-0.03, -0.00)	0.037	-0.003 (-0.02, 0.01)	0.668	-0.002 (-0.01, 0.01)	0.725
Age	-0.039 (-0.10, 0.02)	0.222	-0.037 (-0.10, 0.03)	0.249	0.027 (-0.08, 0.03)	0.371
Sex (Ref: Male)	0.619 (-0.09, 1.33)	0.089	0.205 (-0.53, 0.94)	0.585	0.601 (-0.09, 1.29)	0.087
Model Fit
Log-Likelihood	-659.24		-623.86		-778.77
Deviance	256.67		409.03		213.98
AIC	1330.48		1259.72		1569.55
Pearson chi²	205.61		376.02		141.18

Note. The Muscle Mass (%) row corresponds to robustness specification (ii) reported in the Supplementary Material; the primary model retains Body Fat (%) only. BH q-values shown as Anxiety/Depression/Stress. No association reached the q = 0.05 threshold under uniform Benjamini–Hochberg correction. Transparency note: on re-estimation in Python (statsmodels 0.14.0), the VO2max→Anxiety coefficient yielded P = 0.130 rather than the P = 0.037 originally reported; the discrepancy reflects a standardized vs raw-scale difference (β = -0.163 standardized vs β = -0.013 unstandardized, SD = 12.33). The inferential conclusion is null under uniform FDR regardless of which P-value is used (BH-adjusted q = 0.445 in both cases). The authors have flagged this to the editor for verification of the original analysis script.

All descriptive statistics and initial exploratory analyses were performed using JAMOVI® software version 2.3.21 (Sydney, Australia). To rigorously address the reviewers’ concerns regarding multicollinearity and non-normality, the final robust Generalized Linear Model (GLM) estimation was conducted using the Python programming language (version 3.10) within the statsmodels (version 0.14.0) and pandas computational framework. This environment was used to compute Variance Inflation Factors (VIF) and to implement heteroskedasticity-consistent (HC3) covariance matrix estimators, ensuring robust model diagnostics and valid statistical inference.

The dataset comprised psychological and physical fitness measurements collected from university students in an academic context. Normality of continuous variables was assessed using the Shapiro–Wilk test, which indicated significant deviations from normality for both age (W = 0.887, P < .001) and body mass index (BMI; W = 0.968, P < .001). These normality checks were conducted for descriptive purposes and to motivate robust inference in the linear-model sensitivity analyses; GLM estimation does not rely on normality of the dependent variable. Descriptive characteristics of the sample are presented in Table 2.

Table 2.

Descriptive Data.

	Age	BMI
Mean	20.85	25.07
Median	21	24.9
Standard deviation	2.24	3.73
Minimum	18	16.9
Maximum	32	38.7
W value of Shapiro-Wilk	0.887	0.968
Shapiro-Wilk P-value	<.001	<.001

Ethical Considerations

Anonymity and protection of personal information were guaranteed according to national research ethics regulations. Also, the sample participated willingly, without psychological or physical risk, aligning with the Helsinki Declaration of 2024. All participants signed an informed consent form, which explained the study objectives, procedures, data confidentiality, and their right to withdraw from the study at any time without any consequences. The study was approved by the Institutional Scientific Ethics Committee of the Autonomous University of Chile (CEC-2320).

Results

The results of the negative binomial regression analysis are presented, designed to analyze the influence of VO₂max, Body Fat (%), age, and sex on three psychological variables: anxiety, depression, and stress. All variables were previously standardized to make it easier to compare the extents between predictors.

The results of the Negative Binomial GLM analyses are presented for Anxiety, Depression, and Stress. Regression coefficients (β) are reported on the log scale, with corresponding 95% confidence intervals and P-values. Negative coefficients indicate lower expected outcome values, holding other covariates constant.

Table 1 Multivariate negative binomial regression analysis of physical fitness predictors for psychological distress domains. Unadjusted p-values are reported. Under uniform Benjamini–Hochberg false discovery rate (FDR) correction across all 15 predictor–outcome tests at q = 0.05, no association remained statistically significant (all BH-adjusted q-values ≥0.45; see Supplemental Table S4 for unadjusted p-values, BH-adjusted Q-values and BH decisions). Note on reproducibility: the unadjusted P-value for VO₂max→Anxiety reported in this table (P = 0.037) corresponds to the unstandardized coefficient scale (raw units of VO₂max). On re-estimation using the standardized scale in Python/statsmodels, the coefficient was β = -0.163 (95% CI: -0.376, +0.048; P = 0.130). Both values are reported transparently; under uniform BH-FDR correction the inferential conclusion is identical regardless of which p-value is used (smallest BH-adjusted q ≥ 0.445 in both cases), and no association is statistically significant.

Table 1 presents the results of the negative binomial regression analysis for the three mental health domains. Because we did not pre-specify anxiety as a confirmatory primary outcome before data inspection, all 15 predictor–outcome tests (5 predictors × 3 domains) are reported as exploratory and adjusted uniformly using the Benjamini–Hochberg FDR procedure at q = 0.05. After this correction, no predictor was significantly associated with any psychological domain. For descriptive purposes, the largest unadjusted association was observed between estimated VO₂max and anxiety symptoms, in the negative direction expected from prior literature, although the BH-adjusted q-value did not meet the q = 0.05 threshold. The coefficient for VO₂max in the depression and stress models was substantially smaller and far from significance, and Body Fat (%), Muscle Mass (%), age and sex did not show meaningful associations with any outcome at either unadjusted or adjusted thresholds. A non-significant trend toward higher anxiety and stress scores in female participants was observed but is consistent with what would be expected from sampling variability in a sex-imbalanced sample (n female = 40) and should not be interpreted as evidence of an effect. These exploratory patterns are summarized in Table 1, with unadjusted P-values reported alongside BH-adjusted q-values; inferential conclusions are based only on the adjusted results.

In Figure 1, Model diagnostic metrics revealed substantial heterogeneity in how well physical fitness and body-composition indicators predicted different psychological domains (Figure 1). In terms of absolute goodness-of-fit to the observed count data, the stress and anxiety models demonstrated superior performance, characterized by the lowest Pearson X² values (141.2 and 205.6, respectively) and residual Deviance (214.0 and 256.7, respectively) (Figure 1(a) and (c)). In contrast, the depression model exhibited considerably higher residual variation (Pearson chi² = 376.0; Deviance = 409.0), indicating a poorer fit to the data distribution. However, evaluation of model parsimony via the Akaike Information Criterion (AIC) presented a nuanced perspective (Figure 1(b)). The depression model achieved the lowest AIC (1259.7), followed by anxiety (1330.5) and stress (1569.5). This suggests that while the stress and anxiety models better capture the overall variability in scores, the depression model is the most “efficient” relative to the specific set of predictors used, despite leaving a larger proportion of variance unexplained. Collectively, these metrics indicate that the anxiety model offers the most balanced profile of robust fit and parsimony among the three domains studied.

Figure 1.

Comparison of model fit metrics across psychological domains.

Discussion

The present study examined the associations between several components of physical fitness and psychological distress among Chilean university students. Accordingly, this study should be interpreted as an analysis of a university sample with substantial representation of sport-related degrees, rather than as a representative estimate for the general university population. Under uniform Benjamini–Hochberg FDR correction across all 15 predictor outcome tests, no physical fitness or body-composition indicator was significantly associated with any psychological domain. At descriptive, unadjusted level, the largest effect was an inverse, anxiety-specific signal for estimated VO₂max, substantially smaller in the depression and stress models, which is consistent in direction with prior literature linking aerobic capacity to neuroendocrine and emotion-regulation pathways but does not, in this sample, constitute confirmatory evidence. A non-significant trend toward higher anxiety/stress in female students was also observed. These exploratory patterns are broadly compatible with the wider literature suggesting that better physical fitness may relate to a lower probability of psychological symptoms via neuroendocrine, emotion-regulation and self-esteem mechanisms.³²^,³³

The inverse association between VO₂max and anxiety aligns with previous research indicating that higher aerobic capacity is linked to reduced physiological arousal, improved autonomic regulation, and enhanced emotion-regulation pathways.^34,37 Such mechanisms may help explain why individuals with better cardiorespiratory fitness report fewer anxiety-related symptoms, particularly in contexts characterized by academic and social demands.³⁵ In this regard, estimated VO₂max emerged as an physical fitness indicator inversely associated with anxiety in the present study, suggesting that individuals with greater cardiorespiratory capacity may exhibit more efficient emotional regulation when facing daily life challenges. This interpretation is supported by recent evidence showing that aerobic fitness contributes to brain plasticity, strengthens connectivity within networks involved in emotional control, and reduces physiological responses to stress.^13,36,37 Additionally, emerging research suggests that higher aerobic fitness may promote psychological resilience by enhancing both region-specific brain processes and large-scale neural circuits involved in emotional and behavioral regulation, particularly during adolescence and early adulthood.³⁷ Unlike some prior studies, however, VO₂max did not predict depressive or stress symptoms in this sample, indicating that distinct psychological domains may respond differently to aerobic capacity. These mechanisms are proposed as theoretical rationale for future confirmatory research, rather than as explanations of the current null findings under uniform FDR correction.

However, variables such as muscle mass percentage did not show significant associations with the psychological variables analyzed. Although some previous studies have related muscle strength to lower levels of anxiety and depression,³⁸ these effects could depend on the frequency of exercise, the perception of physical performance, or previous experience in structured physical activity, factors that were not controlled in this analysis.³⁹ In contrast to previous findings, the present study did not identify significant associations between muscle mass percentage and any of the psychological domains evaluated, despite muscle mass being a structural variable closely related to strength.⁴⁰ This discrepancy suggests that the potential protective effects described in earlier research may depend on factors not captured in this sample, such as habitual training patterns, perceived physical competence, or long-term adaptations. Supporting the broader literature, neuroimaging and biomarker studies have shown that greater muscle strength is associated with larger hippocampal volume, improved white matter integrity, and higher levels of Brain-Derived Neurotrophic Factor (BDNF), a molecule frequently reduced in anxiety and depression.⁴¹ It has also been proposed that muscle contraction during exercise stimulates the release of myokines with anti-inflammatory properties, which may counteract the chronic low-grade inflammation observed in several mood disorders.⁴² Together, these mechanisms suggest that muscular adaptations may contribute to psychological well-being; however, in the current sample, such effects were not reflected in direct associations with muscle mass or strength, possibly due to contextual, behavioral, or methodological factors. These neurobiological mechanisms provide strong rationale for future adequately powered studies to test whether muscular adaptations translate into measurable psychological benefits in more representative university samples.

A non-significant trend toward higher anxiety and stress levels was observed among female students compared with males, consistent with epidemiological evidence indicating greater psychological vulnerability among women in university settings, a pattern documented in international and Latin American samples.^7,9 These sex-related differences have been attributed to psychosocial and cultural factors, including greater emotional burden, differential coping strategies, and heightened exposure to academic and social stressors.^4,7 Although the sex coefficients did not reach conventional statistical significance in the present models, the direction of association suggests that biological, psychosocial, or contextual determinants beyond physical fitness may contribute to psychological distress.^3,4

Finally, some limitations of this study should be acknowledged when interpreting the findings. First, a major limitation is that 58.7% of the participants were enrolled in academic programs related to physical activity and sport. This overrepresentation introduces a potential selection bias, as students in sport-related disciplines typically present higher levels of physical activity, superior fitness profiles, and distinct exercise behaviors compared to the general university population. Consequently, the observed associations cannot be readily generalized to students from other academic fields, and replication studies in more diverse and representative samples are warranted. In addition, the cross-sectional design of the study precludes any causal inference regarding the relationships between physical fitness, body composition, and psychological symptoms. Several relevant variables were not assessed, including habitual physical activity levels, nutritional status, sleep quality, and psychosocial factors, which may function as mediators or confounders in the observed associations. Moreover, cardiorespiratory fitness (VO₂max) was estimated using the Queen’s College Step Test rather than measured directly, which may introduce measurement error, because the test relies on sex-specific prediction equations, measurement error may differ by sex; however, interaction models (VO₂max × sex) did not indicate effect modification in this sample (Supplemental Table S3). This methodological limitation could have attenuated the strength of the associations with mental health outcomes. Future research should prioritize direct assessments of VO₂max via ergospirometry to improve measurement accuracy and better characterize the relationship between cardiorespiratory fitness and psychological well-being. Furthermore, longitudinal designs, non-linear analytical approaches, and the examination of potential moderating factors, such as sex, age, or baseline health status, are recommended to achieve a more comprehensive understanding of the physiological and behavioral mechanisms linking physical fitness to mental health in university students.

Conclusion

This pilot, exploratory study found no statistically significant associations between physical fitness or body-composition indicators and symptoms of anxiety, depression or stress in a convenience sample of Chilean university students enriched in sport-science programmes, once a uniform Benjamini–Hochberg correction was applied across all 15 predictor outcome tests. At descriptive level, the largest signal, a negative association between estimated VO₂max and anxiety, with substantially weaker estimates for depression and stress, was consistent in direction with previous research on aerobic capacity and HPA-axis regulation, but should not be regarded as confirmatory. The findings are constrained by a cross-sectional design, an indirect VO₂max estimate (Queen’s College Step Test), a marked male predominance (85.7%) that left the female subgroup underpowered for sex-stratified analyses, and a sport-enriched convenience sample. They are therefore best read as preliminary, hypothesis-generating, and as a source of effect-size estimates for future adequately powered, pre-registered studies in more representative university populations, ideally using direct measurement of VO₂max and longitudinal or interventional designs.

Supplemental Material

Supplemental Material - Physical Fitness and Body Composition as Predictors of Anxiety, Depression, and Stress in University Students

Supplemental Material for Physical Fitness and Body Composition as Predictors of Anxiety, Depression, and Stress in University Students by Catalina Muñoz-Strale, Frano Giakoni-Ramírez, Josivaldo de Souza-Lima, Javiera Alarcón-Aguilar, Rodrigo Yáñez-Sepúlveda, Andrés Godoy-Cumillaf, Guillermo Cortés-Roco, José Francisco López-Gil in American Journal of Lifestyle Medicine

Supplemental Material

Supplemental Material - Physical Fitness and Body Composition as Predictors of Anxiety, Depression, and Stress in University Students

Supplemental Material Physical Fitness and Body Composition as Predictors of Anxiety, Depression, and Stress in University Students by Catalina Muñoz-Strale, Frano Giakoni-Ramírez, Josivaldo de Souza-Lima, Javiera Alarcón-Aguilar, Rodrigo Yáñez-Sepúlveda, Andrés Godoy-Cumillaf, Guillermo Cortés-Roco, José Francisco López-Gil in American Journal of Lifestyle Medicine

Consent to Publication

During the preparation of this work, the authors used ChatGPT in order to improve language and readability. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Footnotes

Acknowledgments

The authors would like to thank all the students who participated in the study.

Catalina Muñoz-Strale

Josivaldo de Souza-Lima

Javiera Alarcón-Aguilar

Ethical Considerations

This study was approved by the Ethics Committee of Universidad Autónoma de Chile (Ethics Code: CEC-2320) on June 14, 2020. All participants provided written informed consent prior to enrollment in the study. This research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki

Consent to Participate

The study was approved by the Ethics Committee of Universidad Autónoma de Chile (Ethical Clearance Reference Number: CEC-2320) on June 14, 2020. All participants provided written informed consent prior to participating.

Author Contributions

• Catalina Muñoz-Strale – Writing of the manuscript; data collection.

• Frano Giakoni-Ramírez – Editing of the manuscript; data collection.

• Josivaldo de Souza-Lima – Data collection.

• Javiera Alarcón-Aguilar – Data collection.

• Rodrigo Yáñez-Sepúlveda – Editing of the manuscript; data analysis.

• Andrés Godoy-Cumillaf – Editing of the manuscript; data analysis.

• Guillermo Cortés-Roco – Writing and revision of the manuscript.

• José Francisco López-Gil – Writing and revision of the manuscript.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online.

References

Wunsch

Fiedler

Bachert

Woll

. The tridirectional relationship among physical activity, stress, and academic performance in university students: a systematic review and meta-analysis. Int J Environ Res Publ Health. 2021;18(2):739. doi:10.3390/ijerph18020739.

Merellano-Navarro

Bustamante-Ara

Russell-Guzmán

Lagos-Hernández

Uribe

Godoy-Cumillaf

. Association between sleep quality and physical activity in physical education students in Chile in the pandemic context: a cross-sectional study. Healthcare. 2022;10(10):1930. doi:10.3390/healthcare10101930.

Concha-Cisternas

Castillo-Retamal

Guzmán-Muñoz

. Comparación de la calidad de vida en estudiantes universitarios según nivel de actividad física. Universidad y Salud. 2019;22(1):33-40. doi:10.22267/rus.202201.172.

Ibrahim

Kelly

Adams

Glazebrook

. A systematic review of studies of depression prevalence in university students. J Psychiatr Res. 2013;47(3):391-400. doi:10.1016/j.jpsychires.2012.11.015.

Akram

Ypsilanti

Gardani

, et al. Prevalence and psychiatric correlates of suicidal ideation in UK university students. J Affect Disord. 2020;272:191-197. doi:10.1016/j.jad.2020.03.185.

Rodríguez-Rodríguez

Cristi-Montero

Villa-González

Solís-Urra

Chillón

. Comparison of the physical activity levels during the university life. Rev Med Chile. 2018;146(4):442-450. doi:10.4067/s0034-98872018000400442.

Sheldon

Simmonds-Buckley

Bone

, et al. Prevalence and risk factors for mental health problems in university undergraduate students: a systematic review with meta-analysis. J Affect Disord. 2021;287:282-292. doi:10.1016/j.jad.2021.03.054.

Bravo

Errázuriz

Campos

. Termómetro de Salud Mental en Chile ACHS-UC: quinta Ronda. 2022. Available from: https://www.achs.cl/docs/librariesprovider2/noticias-2022/achs-127833/tms-ronda-5-conferencia.pdf

Antúnez

Vinet

. Escalas de Depresión, Ansiedad y Estrés (DASS - 21): Validación de la Versión abreviada en Estudiantes Universitarios Chilenos. Ter Psicol. 2012;30(3):49-55. doi:10.4067/s0718-48082012000300005.

10.

Liu

. Intervention effect of aerobic exercise on physical fitness, emotional state and mental health of drug addicts: a systematic review and meta-analysis. Int J Environ Res Publ Health. 2023;20(3):2272. doi:10.3390/ijerph20032272.

11.

Chiang

Chuang

Chen

, et al. Physical fitness and risk of mental disorders in children and adolescents. JAMA Pediatr. 2024;178(6):595-607. doi:10.1001/jamapediatrics.2024.0806.

12.

American College of Sports Medicine . Acsm’s Guidelines for Exercise Testing and Prescription. Lippincott Williams & Wilkins; 2013.

13.

Lubans

Richards

Hillman

, et al. Physical activity for cognitive and mental health in youth: a systematic review of mechanisms. Pediatrics. 2016;138(3):e20161642. doi:10.1542/peds.2016-1642.

14.

Kandola

Ashdown-Franks

Hendrikse

Sabiston

Stubbs

. Physical activity and depression: towards understanding the antidepressant mechanisms of physical activity. Neurosci Biobehav Rev. 2019;107:525-539. doi:10.1016/j.neubiorev.2019.09.040.

15.

Kljajević

Stanković

Đorđević

, et al. Physical activity and physical fitness among university students—A systematic review. Int J Environ Res Publ Health. 2021;19(1):158. doi:10.3390/ijerph19010158.

16.

Martínez-Gómez

Ruiz

Gómez-Martínez

, et al. Active commuting to school and cognitive performance in adolescents: the AVENA study. Arch Pediatr Adolesc Med. 2011;165(4):300-305. doi:10.1001/archpediatrics.2010.244.

17.

Suwannakul

Sangkarit

Manoy

Amput

Tapanya

. Association between stress and physical fitness of university students post-COVID-19 pandemic. J Funct Morphol Kinesiol. 2023;8(1):33. doi:10.3390/jfmk8010033.

18.

Muñoz-Strale

Giakoni-Ramírez

Pinochet

Godoy-Cumillaf

Fuentes-Merino

Duclos-Bastías

. Condición física, actividad física y calidad de vida en estudiantes universitarios chilenos. Retos. 2024;56:521-530. doi:10.47197/retos.v56.104184.

19.

Diegelmann

Jansen

Wahl

Schilling

Schnabel

Hauer

. Does a physical activity program in the nursing home impact on depressive symptoms? A generalized linear mixed-model approach. Aging Ment Health. 2017;22(6):784-793. doi:10.1080/13607863.2017.1310804.

20.

Grant

Villafranca

. Methods you can count on: a simulation experiment showing the advantages of generalized linear modeling (GLM) over the linear regression of log-transformed count data. J Archaeolog Sci: Rep. 2024;53:104377. doi:10.1016/j.jasrep.2024.104377.

21.

Moreno

Gómez

González

, et al. Validación del monitor de medición de la grasa corporal por impedancia bioeléctrica OMRON BF 300. Aten Primaria. 2001;28(3):174-181. doi:10.1016/s0212-6567(01)78927-5.

22.

CDC . Acerca del Índice De Masa Corporal Para Adultos. Centers for Disease Control and Prevention; 2025. Available from: https://www.cdc.gov/healthyweight/spanish/assessing/bmi/adult_bmi/index.html

23.

Romero-Dapueto

Mahn

Cavada

Daza

Ulloa

Antúnez

. Estandarización de la fuerza de prensión manual en adultos chilenos sanos mayores de 20 años. Rev Med Chile. 2019;147(6):741-750. doi:10.4067/s0034-98872019000600741.

24.

Mancilla

Ramos

Morales

. Fuerza de prensión manual según edad, género y condición funcional en adultos mayores chilenos entre 60 y 91 años. Rev Med Chile. 2016;144(5):598-603. doi:10.4067/s0034-98872016000500007.

25.

Caamaño-Navarrete

Jerez-Mayorga

Alvarez

Del-Cuerpo

Cresp-Barría

Delgado-Floody

. Muscle quality index in morbidly obesity patients related to metabolic syndrome markers and cardiorespiratory fitness. Nutrients. 2023;15(11):2458. doi:10.3390/nu15112458.

26.

Melo

GLR

Moraes

Nascimento

, et al. Field-based versus laboratory-based estimates of muscle quality index in adolescents with and without Down syndrome. J Intellect Disabil Res. 2022;66(12):1000-1008. doi:10.1111/jir.12959.

27.

Montoro-Bombú

Gomes

Santos

Rama

. Validity and reliability of a load cell sensor-based device for assessment of the isometric mid-thigh pull test. Sensors. 2023;23(13):5832. doi:10.3390/s23135832.

28.

Tondelli

. Evaluaciones físicas en la práctica clínica kinésica. Argentinian J Respiratory Physical Therapy. 2021;3(2):40-56. doi:10.58172/ajrpt.v3i2.188.

29.

Heyward

. Evaluación de la aptitud física y prescripción del ejercicio. Editorial Médica Panamericana. 2008. 5a ed.

30.

Lovibond

. The structure of negative emotional states: comparison of the depression anxiety stress scales (DASS) with the beck depression and anxiety inventories. Behav Res Ther. 1995;33(3):335-343. doi:10.1016/0005-7967(94)00075-U.

31.

Younes

Halawi

Jabbour

, et al. Internet addiction and relationships with insomnia, anxiety, depression, stress and self-esteem in university students: a cross-sectional designed study. PLoS One. 2016;11(9):e0161126. doi:10.1371/journal.pone.0161126.

32.

Dishman

Heath

Lee

. Physical Activity Epidemiology. 3rd ed. Human Kinetics; 2021.

33.

Jiang

Zhang

. Exercise as a mediator of wellbeing. Front Psychol. 2025;16:1443397. doi:10.3389/fpsyg.2025.1443397.

34.

Kandola

Ashdown-Franks

Stubbs

Osborn

DPJ

Hayes

. The association between cardiorespiratory fitness and the incidence of common mental health disorders: a systematic review and meta-analysis. J Affect Disord. 2019;257:748-757. doi:10.1016/j.jad.2019.07.088.

35.

Herring

. The effect of exercise training on anxiety symptoms among patients. Arch Intern Med. 2010;170(4):321-333. doi:10.1001/archinternmed.2009.530.

36.

Esteban-Cornejo

Rodriguez-Ayllon

Verdejo-Román

, et al. Physical fitness, white matter volume and academic performance in children: findings from the ActiveBrains and FITKids2 projects. Front Psychol. 2019;10:208. doi:10.3389/fpsyg.2019.00208.

37.

Belcher

Zink

Azad

Campbell

Chakravartti

Herting

. The roles of physical activity, exercise, and fitness in promoting resilience during adolescence: effects on mental well-being and brain development. Biol Psychiatry Cogn Neurosci Neuroimaging. 2021;6(2):225-237. doi:10.1016/j.bpsc.2020.08.005.

38.

Cabanas-Sánchez

Esteban-Cornejo

Parra-Soto

, et al. Muscle strength and incidence of depression and anxiety: findings from the UK biobank prospective cohort study. J Cachexia Sarcopenia Muscle. 2022;13(4):1983-1994. doi:10.1002/jcsm.12963.

39.

Teychenne

White

Richards

Schuch

Rosenbaum

Bennie

. Do we need physical activity guidelines for mental health: what does the evidence tell us? Ment Health Phys Act. 2020;18:100315. doi:10.1016/j.mhpa.2019.100315.

40.

Almiray-Soto

Denova-Gutiérrez

Lopez-Gonzalez

Medeiros

Clark

. Muscle strength reference values and correlation with appendicular muscle mass in Mexican children and adolescents. Calcif Tissue Int. 2022;111(6):597-610. doi:10.1007/s00223-022-01025-4.

41.

Sleiman

Henry

Al-Haddad

, et al. Exercise promotes the expression of brain-derived neurotrophic factor (BDNF) through the action of the ketone body β-hydroxybutyrate. eLife. 2016;5:e15092. doi:10.7554/eLife.15092.

42.

Pedersen

Febbraio

. Muscles, exercise and obesity: skeletal muscle as a secretory organ. Nat Rev Endocrinol. 2012;8:457-465. doi:10.1038/nrendo.2012.49.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.45 MB

0.32 MB