The health diet as a mediator in relationship between stress and sleep quality of college students with physical exercise as a moderator

Abstract

Background and objectives

With the development of society, contemporary young people, especially college students, are facing increasing levels of stress. Stress is often associated with poor sleep quality in individuals, but the underlying mechanisms between these two factors need further exploration. Therefore, this study aims to investigate the relationship between stress and sleep quality among college students, with stress as the independent variable, sleep quality as the dependent variable, healthy diet as the mediating variable, and physical exercise as the moderating variable, to further elucidate the underlying connections between stress and sleep quality in college students.

Methods

A total of 4090 college students (1503 males and 2587 females; mean age 19.05 ± 1.16 years) were included in this study. Data on stress, healthy diet, physical exercise, and sleep quality were collected through a cross-sectional subjective survey. The basic characteristics of the variables were described using SPSS, and correlation analyses were performed. Finally, a moderated mediation model was constructed using the PROCESS macro.

Results

The study found that stress was significantly negatively correlated with healthy diet (r = -0.162, p < 0.001) and sleep quality (r = -0.492, p < 0.001) among college students. Healthy diet mediated the relationship between stress (β = -0.161, SE = 0.015, p < 0.001) and sleep quality (β = 0.160, SE = 0.014, p < 0.001) among college students. Physical exercise significantly moderated the relationship between stress and healthy diet (β = -0.040, SE = 0.015, p < 0.01).

Discussion

This study further elucidated the mechanisms underlying the relationship between stress and sleep quality among college students by incorporating behavioral variables (healthy diet and physical exercise). It is recommended that college students actively engage in physical exercise, which can help alleviate stress, promote healthy eating, and thereby improve sleep quality.

Introduction

From psychological and biological perspectives, stress refers to any environmental or physical pressure that elicits a response from an organism [1]. In most cases, stress promotes survival by forcing organisms to adapt to rapidly changing environmental conditions. However, when an organism's response to stress is insufficient or the stress is too intense, it may lead to illness or even death [1]. In this study, the term “stress” focuses on psychological distress, specifically emotional tension, anxiety, and worry that interfere with daily Life, including sleep. This is consistent with our measurement methods and previous Literature. A meta-analysis involving 2,088,032 global students during the COVID-19 pandemic found that the perceived stress prevalence rate of 31% [2]. Meanwhile, a study in China reported that the prevalence of stress among college students (n = 1586; Perceived Stress Scale [PSS-10] ≥ 14) reached 67.5% [3]. After leaving the strictly regulated high school environment, college students enter a more liberal university campus, where they face various challenges and associated stressors, such as interpersonal stress [4], academic pressure [5], financial stress [6], cultural adaptation stress [7], and employment pressure [8]. Therefore, this study aims to explore the impact of stress on Chinese college students and identify protective factors.

Sleep quality among college students is another important topic widely investigated by scholars. The National Sleep Foundation-led expert panel, after reviewing 277 studies, defined sleep quality as comprising four elements: sleep latency, awakenings, wake after sleep onset, and sleep efficiency [9]. In the relatively free university environment, Chinese college students may indulge in their lifestyles and neglect healthy sleep routines [10]. A recent cross-sectional survey involving 6284 college students showed that over 30% had poor sleep quality [11]. According to the hyperarousal model of insomnia [12], psychobiological disorders stemming from chronic psychological distress (excessive rumination) and physiological imbalance (disruption of the hypothalamic–pituitary–adrenal axis) may lead to poor sleep quality among college students. One important variable in this theory is stress, which has been found to be significantly positively correlated with poor sleep quality [13][14]. In a large-sample cross-sectional study of Chinese college students (n = 8178), perceived stress was found to significantly predict poor sleep quality among college students, even after introducing additional mediating variables (presence of meaning in life and depression), with a significant predictive effect (β = 0.14, p < 0.001) [15]. In a review study on factors affecting sleep quality among college students, the authors concluded that "stress is one of the important psychological factors affecting sleep quality" [16].

Dietary patterns may mediate the relationship between stress and sleep quality among college students. In an extensive survey conducted in 2023, involving 222,401 adolescents from 61 countries and regions, a systematic review was carried out on four types of unhealthy eating patterns. The study revealed that globally, the proportions of adolescents with insufficient fruit intake, inadequate vegetable consumption, excessive daily beverage consumption, and overconsumption of fast food on a weekly basis were 37.0%, 28.5%, 50.0%, and 57.4%, respectively [17]. Additionally, among college students, 28.2% exhibited disordered eating behaviors, while 40.2% engaged in binge eating [18]. To unify the terminology of variables, this study defines a healthy diet as consuming fewer sugary or caffeinated foods and more fruits and vegetables [19]. Research has found a significant positive correlation between stress and sugary food consumption, and stress also strengthens the longitudinal relationship between depression and sugary food consumption [20]. Moreover, studies have shown that higher stress levels are associated with more frequent consumption of sweets and fast food, while reducing the frequency of fruit and vegetable consumption [21]. A randomized controlled trial found that alleviating individual stress levels helps increase their consumption of fruits and vegetables [22]. Additionally, research has shown a significant relationship between diet and sleep quality [23, 24]. In a longitudinal study in China with three follow-ups (each six months apart), carbonated beverages were found to significantly predict future insomnia levels among female students [25]. A meta-analysis of 22 original studies (12 adult and 10 child studies) found that short sleep duration and delayed optimal sleep timing in adults were associated with increased energy drink consumption (OR: 1.58; 95% CI: 1.31–1.90), and compared to children with sufficient sleep, short-sleeping children had a higher likelihood of sugar consumption (OR: 1.16; 95% CI: 1.10–1.21). Overall, individuals with sufficient sleep had lower intake of sugary drinks [26]. A study involving 1444 Americans aged 21–30 years showed that better sleep was associated with higher fruit and vegetable consumption through secondary analysis [27]. After considering covariates, men with better sleep quality and shorter sleep latency had higher fruit and vegetable intake.

Physical exercise is defined as any bodily movement produced by skeletal muscles that requires energy expenditure [28]. In addition to its well-known physical health benefits, a growing body of research has demonstrated that physical activity is closely linked to psychological stress reduction and improved sleep quality. Meta-analytic findings suggest that regular exercise helps to lower cortisol levels and increase the release of endorphins and serotonin, which can buffer the physiological effects of stress [29]. Furthermore, physical exercise has been shown to enhance sleep quality by promoting sleep onset, increasing deep sleep, and reducing nocturnal awakenings [30]. Theoretically, physical activity may also serve a social-regulatory function. According to social bond theory [31], engaging in group or structured exercise can strengthen interpersonal connections, increase perceived social support, and reduce feelings of isolation—all of which are protective against stress and poor sleep. In addition, some studies have found that physical activity reduces psychological inflexibility, which is a key cognitive risk factor in stress-related sleep disorders [13][32]. Despite these findings, few studies have examined whether physical exercise can moderate the pathways between stress, diet, and sleep quality in a unified model, particularly in college student populations. Thus, the current study introduces physical exercise as a moderating variable to explore its potential buffering effect on both behavioral and psychological responses to stress.

While the individual relationships between psychological stress, sleep quality, healthy diet, and physical exercise have been established in previous studies, most have examined these factors in isolation or used different analytic frameworks. The current study contributes to the literature in three important ways. First, it investigates a comprehensive moderated mediation model in a large sample of Chinese college students, a population facing increasingly high levels of psychological stress but underrepresented in integrated behavioral health models. Second, we conceptualize healthy diet as a mediator based on the behavioral self-regulation perspective. Psychological stress can impair decision-making and increase impulsivity, leading individuals to choose unhealthy food options (e.g., high sugar, high fat), which may then impair sleep through mechanisms such as metabolic disruption or inflammation. While the reverse pathway is also plausible, we focus on this direction because it reflects a stress-induced behavioral cascade with actionable intervention targets. Third, we introduce physical exercise as a moderator on two separate pathways—the effect of stress on healthy diet, and the direct effect of stress on sleep quality. This dual-pathway moderation approach enables a more nuanced understanding of how exercise may attenuate the negative impacts of stress across both behavioral and psychological domains. Therefore, this study proposes the following hypothesis:

• H1: There is a significant negative correlation between stress and sleep quality among college students.

• H2: The healthy diet significantly mediates the relationship between stress and sleep quality among college students.

• H3: The physical exercise significantly moderates the relationships between stress and healthy diet.

• H4: The physical exercise significantly moderates the relationships between stress and sleep quality.

The hypothesized model is shown in Fig. 1.

Fig. 1.

Hypothetical model diagram

Methods

Participants

This study used convenience sampling to recruit college students from seven universities located in Southwest, Northwest, Central, and East China between October and November 2024. The study was approved by the Ethics Committee of Xizang Minzu University before the initiation of the project (Grant number: 2024–19). Online electronic questionnaires were distributed through class group chats, with an informed consent form placed on the first page of the questionnaire. Participants could proceed with the survey only after reading and clicking "agree." Data were collected only from participants who completed the entire questionnaire. Participants who disagreed to participate or exited during the survey were excluded, and their data were not collected. Their right to freely participate was fully respected, and no adverse effects were imposed on them. The first page of the questionnaire explained the survey content, data anonymity, confidentiality, and usage. Participants typically completed the electronic questionnaire within 10 min. Data were screened to exclude samples with extreme response times and obvious patterned answers. The final analytic sample included 4090 participants (mean age 19.05 ± 1.16 years), including 1,503 males and 2,587 females; 1,019 were only children and 3,071 had siblings. Regarding paternal education, 229 participants reported "unclear," 844 indicated "primary school or below," 2,568 "middle school," 418 "college," and 31 "graduate." For maternal education, 245 reported "unclear," 1,117 "primary school or below," 2,317 "middle school," 385 "college," and 26 "graduate."

Measurement tools

Considering the daily academic stress of participants and the instability of online questionnaires, as well as minimizing disruption to participants' or teachers' regular teaching schedules, we selected convenient measurement tools. Simple measurement tools offer numerous advantages [33, 34], such as lower cognitive and time costs for participants, reduced response bias, higher satisfaction, and lower data processing costs.

Stress

A single-item measure of stress level was used to assess stress [35]. The question was: "Have you felt nervous, uneasy, or anxious, or had your thoughts so disturbed that you couldn't sleep at night, in the last two weeks?" It was scored on a 4-point scale, ranging from 1 (never) to 4 (nearly every day). The scale scores ranging from 1 to 4. A higher score indicates a higher stress level. This tool has been widely used in previous studies [36, 37][38].

Healthy eating

Healthy eating was assessed using three questions: "In the past week, how often did you (1) drink carbonated beverages? (2) eat fruit (excluding fruit juice)? (3) eat vegetables (including salads and non-fried potatoes)?" Response options ranged from 0 days to 5 or more days [19]. The total healthy eating score was calculated by summing fruit and vegetable intake and subtracting carbonated beverage consumption, with higher scores indicating healthier dietary habits. This measurement tool has been used in prior studies [39][38].

Sleep quality

A single-item measure was used to evaluate sleep quality [40]. The question was: "Please reflect on your overall sleep quality over the past seven days. This includes aspects such as the number of hours you slept, how easily you fell asleep, the frequency of nocturnal awakenings (excluding bathroom visits), waking up earlier than necessary in the morning, and feeling refreshed upon waking. How would you rate your sleep quality over the past seven days?" This question was scored on a 10-point scale, with higher scores indicating better sleep quality. This measurement approach has been extensively used in prior research [41–44][14].

Physical exercise

Physical activity was assessed with a single item: "In the past 7 days, how many days did you engage in at least 20 min of physical exercise or physical activity that made you sweat or breathe hard?" Response options ranged from 0 to 7 days [19]. This measurement question has been widely used in prior studies [39, 45, 46][47–51].

Covariates

In the present study, demographic variables mentioned in "Participants" section were treated as covariates when estimating the mediation model and moderated mediation model. In the current study, only-child status and parental education were included as covariates. These variables were selected because previous research suggests that family structure (e.g., being an only child) and parental educational background may influence psychological stress perception, eating behaviors, and sleep-related health outcomes in college students [52].

Data analysis

Before analyzing the data, a common method bias (CMB) test was conducted. Following the recommendations of Podsakoff et al. [53], a threshold of 40% was set to determine whether significant bias existed in the data. Descriptive statistics (means and standard deviations) for the 4 focal variables—physical exercise, sleep quality, stress, and healthy diet—were then calculated. Independent-samples t-tests, one-way ANOVAs, and Pearson correlations (r) were performed in SPSS 26.0. The moderated mediation model was estimated with the PROCESS 4.0 macro. All variables were standardized beforehand. Stress was specified as the independent variable, sleep quality as the dependent variable, healthy diet as the mediator, and physical exercise as the moderator. Demographic variables (only-child status and parental education) were entered as covariates [54]. Model 8 in PROCESS was employed, which extends Model 4 by adding a moderator that influences (a) the path from the independent variable to the mediator and (b) the path from the independent variable to the dependent variable. The moderation effects include both the main effect of the moderator and the interaction between the moderator and the independent variable. A statistical diagram illustrating these relationships was produced [55]. Conditional effects were plotted by selecting “Mean – 1 SD, Mean, Mean + 1 SD" under “Conditioning values" in PROCESS, generating simple slope plots of the moderator’s effect [56, 57]. To assess model fit and estimate 95% confidence intervals (95% CI), 5000 bootstrap resampling iterations were performed to ensure the robustness of the analysis [58]. The significance level was set at 0.05.

Results

Common method bias test

The results of the common method bias test in this study revealed the presence of four factors with eigenvalues greater than 1. The first factor accounted for 18.93% of the total variance, which is below the threshold of 40%. This indicates that there is no significant risk of common method bias in the current study.

Descriptive analysis

The analyses of the study’s key variables—stress, healthy diet, sleep quality, and physical exercise—across gender, only-child status, and grade are presented in Table 1.

Table 1.

Analysis of the differences of stress, health diet, sleep quality and physical exercise among gender, only child status and grade (*: p < 0.05; ***: p < 0.001)

Variables		Stress		Health diet		Sleep quality		Physical exercise
		Mean	SD	Mean	SD	Mean	SD	Mean	SD
Gender	Male	1.55	0.68	4.86	2.75	7.10	2.07	3.63	2.18
	Female	1.55	0.63	5.25	2.80	7.11	1.86	2.75	2.08
	t	0.124		−4.264***		−0.262		12.654***
Only-child status	Only	1.54	0.67	5.06	2.88	7.19	1.99	2.96	2.19
	Not only	1.55	0.65	5.12	2.76	7.08	1.92	3.11	2.15
	t	−0.407		−0.579		1.453		−1.925
Grade	Freshman year	1.47	0.61	5.21	2.82	7.28	1.88	3.40	2.11
	Sophomore	1.63	0.68	4.98	2.78	6.92	1.99	2.69	2.10
	Junior in college	1.53	0.58	5.30	2.57	7.33	1.85	3.71	2.38
	Senior year	1.75	0.91	5.35	2.57	6.65	2.18	1.95	2.40
	F	21.194***		2.720*		12.620***		45.779***

Correlation aanalysis

As shown in Table 2, stress was significantly negatively correlated with healthy eating (r = −0.162, p < 0.001) and sleep quality (r = −0.492, p < 0.001) among college students. Healthy eating was significantly positively correlated with sleep quality (r = 0.246, p < 0.001) and physical exercise (r = 0.197, p < 0.001). Additionally, physical exercise was significantly positively correlated with sleep quality (r = 0.085, p < 0.001).

Table 2.

Correlation analysis

Variables	Mean	SD	1	2	3	4
1 Stress	1.55	0.62	-
2 Health diet	5.11	2.79	−0.162***	-
3 Sleep quality	7.11	1.94	−0.492***	0.246***	-
4 Physical exercise	3.08	2.16	−0.012	0.197***	0.085***	-

^***p < 0.001

Moderated mediation model test

After controlling for demographic variables, the results shown in Table 3, Fig. 2, and Fig. 3 indicate that stress significantly negatively predicted healthy eating (β = −0.161, SE = 0.015, p < 0.001) and sleep quality (β = −0.465, SE = 0.016, p < 0.001) among college students. Healthy eating significantly positively predicted sleep quality (β = 0.160, SE = 0.014, p < 0.001). Physical exercise significantly positively predicted healthy eating (β = 0.218, SE = 0.015, p < 0.01) and sleep quality (β = 0.048, SE = 0.014, p < 0.01). Finally, the interaction term between physical exercise and stress significantly negatively predicted healthy eating (β = −0.040, SE = 0.015, p < 0.01).

Table 3.

Mediation model testing

Outcome variables	Predictor variables	β	SE	t	R2	F
Health diet	Stress	−0.161	0.015	−10.638***	0.079	38.964**
	Physical exercise	0.218	0.015	14.142***
	Stress × physical exercise	−0.040	0.015	−2.712**
Sleep quality	Stress	−0.465	0.014	−34.176***	0.274	153.734***
	Health diet	0.160	0.014	−34.176***
	Physical exercise	0.048	0.014	3.443**
	Stress × physical exercise	−0.004	0.013	−0.275

^*p < 0.05

^**p < 0.01

^***p < 0.001

Fig. 2.

Moderated Mediation Model (Stress as the independent variable, healthy diet as the mediator, sleep quality as the dependent variable, and physical exercise as the moderator; **p < 0.01, ***p < 0.001)

Fig. 3.

Slope diagram (Left panel: Physical exercise significantly moderates the relationship between stress and healthy diet. Right panel: Physical exercise does not significantly moderate the relationship between stress and sleep quality)

Further examination of the moderating effect of different levels of physical exercise on the relationship between stress and healthy eating revealed that physical exercise significantly moderated this relationship at different levels (see Table 4).

Table 4.

The moderating effect of mean ± 1SD levels of physical exercise

The moderated relationship	Moderating variable (Physical exercise)	Effect size	SE	t	Lower Limit 95%CI	Upper Limit 95%CI
Stress and health diet	Mean −1SD	−0.121	0.021	−5.669***	−0.163	−0.079
	Mean	−0.161	0.015	−10.638***	−0.190	−0.131
	Mean + 1SD	−0.201	0.021	−9.604***	−0.242	−0.160
Stress and sleep quality	Mean −1SD	−0.461	0.019	−24.285***	−0.499	−0.424
	Mean	−0.465	0.014	−34.176***	−0.492	−0.438
	Mean + 1SD	−0.468	0.019	−24.963***	−0.505	−0.432

^***p < 0.001

Discussion

This study further examined the relationship between stress and sleep quality among college students, exploring how behavioral factors such as healthy diet and physical exercise may contribute to this association. The findings revealed significant negative correlations between stress and sleep quality, as well as between stress and healthy eating among college students. Moreover, the relationship between stress and sleep quality was mediated by healthy eating. Physical exercise significantly moderated the relationship between stress and healthy eating but did not significantly moderate the relationship between stress and sleep quality.

The negative correlation between stress and sleep quality among college students is consistent with our initial hypothesis (H1). Previous studies have identified stress as one of the key psychological predictors of poor sleep quality in this population [16]. This relationship is supported by physiological evidence: chronic stress can disrupt the homeostasis of the hypothalamic–pituitary–adrenal (HPA) axis, leading to fragmented sleep and reduced sleep quality [59]. Stress has also been shown to interfere with serotonin secretion, a neurotransmitter that plays a crucial role in regulating sleep cycles and quality [60–62]. From a psychological perspective, longitudinal studies have confirmed that stress not only directly predicts poorer sleep quality but also exerts indirect effects through rumination and emotional dysregulation [63]. For instance, Liu et al. found that stress significantly predicted insomnia symptoms among Chinese undergraduates via cognitive arousal and emotion-focused coping [64]. Similarly, recent cross-cultural surveys have indicated a robust link between academic stress and sleep disturbances in university students, especially in East Asian populations where academic pressure is high [65]. These findings are consistent with our results, which showed that stress was significantly negatively correlated with sleep quality in college students.

Previous studies have established a link between psychological stress and unhealthy eating behaviors, as well as between dietary patterns and sleep quality. For instance, experimental research has demonstrated that individuals under stress are more likely to choose unhealthy foods over healthy options, often due to the immediate pleasure or emotional relief provided by high-sugar or high-fat items [66]. Structured interview studies have similarly shown that high stress levels increase the likelihood of emotional eating [67]. Moreover, dietary choices also appear to affect sleep: high intake of sugary foods has been associated with shorter sleep duration and lower sleep quality [68], while a longitudinal intervention study found that increasing fruit and vegetable intake significantly improved insomnia symptoms, sleep quality, and sleep latency over a three-month period [69]. Biologically, this may be due to the role of nutrients found in fruits and vegetables in enhancing the secretion of melatonin and other sleep-regulating hormones [70]. In Line with these findings, our study confirmed Hypothesis 2 (H2): stress was significantly negatively associated with healthy eating, and healthy eating was significantly positively associated with sleep quality. These results suggest that under higher levels of stress, college students may be more prone to unhealthy eating behaviors, which in turn impair sleep quality. The mediating role of healthy diet between stress and sleep quality provides empirical support for understanding how behavioral factors connect psychological and physical health outcomes in this population. Our findings also extend prior research by validating this mediation model within a large sample of Chinese university students, which has been underrepresented in the literature. By highlighting the intermediary function of dietary behavior, this study underscores the importance of addressing stress-related eating patterns in sleep interventions among young adults.

As shown in Fig. 2, the interaction term between stress and physical exercise significantly predicted healthy diet (β = –0.040, p < 0.01), providing direct statistical support for Hypothesis 4 (H4). This result indicates that the strength of the association between stress and healthy diet varies across levels of physical exercise. Based on this interaction, we conducted a simple slope analysis (Fig. 3 and Table 4), which revealed that the negative effect of stress on healthy eating was strongest among participants with high levels of physical activity. This pattern may reflect higher health standards, stricter self-regulatory routines, or stronger appearance-related motivations among physically active college students. These findings are consistent with prior research suggesting that physical activity is positively associated with better dietary choices, including increased consumption of fruits and vegetables and reduced intake of sugary and processed foods [71]. Physical exercise has also been shown to improve self-regulation and inhibitory control [72, 73], which may help individuals resist unhealthy eating behaviors under stress [74]. Moreover, physical activity has been identified as an effective buffer against stress through its regulation of the hypothalamic–pituitary–adrenal (HPA) axis, reducing cortisol levels and promoting psychological flexibility [75]. Social engagement and improved mood associated with physical activity can also reduce stress-induced emotional eating tendencies (Socioeconomic differences in leisure-time physical activity: the role of social participation and social capital in shaping health related behaviour). Taken together, these findings support the moderating role of physical activity in reducing the negative impact of stress on dietary behavior.

In contrast, physical exercise did not significantly moderate the relationship between stress and sleep quality (β = –0.004, p > 0.05), and thus Hypothesis 5 (H5) was not supported. Although previous research has consistently shown that moderate physical activity is associated with improved sleep outcomes [76], its moderating effect on stress-related sleep disturbances appears to be more complex. One possible explanation lies in the measurement tools used in this study, which assessed fewer dimensions of sleep and may have lacked the sensitivity to detect subtle interaction effects [77]. Beyond methodological concerns, recent evidence suggests that the moderating role of physical activity on stress-related sleep outcomes is both complex and context-dependent. While moderate-intensity exercise is generally beneficial for sleep quality [78], its protective effect may be diminished under certain conditions. For example, vigorous or late-evening exercise can increase physiological arousal and interfere with sleep onset and maintenance, particularly among individuals already experiencing high levels of stress [79]. Moreover, cognitive-emotional factors—such as rumination and anxiety—may exert a greater influence on stress-induced sleep disturbances than behavioral factors like exercise alone [80]. Therefore, physical activity may not fully buffer the negative impact of stress on sleep unless combined with psychological interventions targeting maladaptive thought patterns. Additionally, university students often face irregular schedules, high academic demands, and prolonged screen time, all of which can impair sleep quality regardless of physical activity habits [16]. These external stressors may offset the potential moderating effects of exercise. Taken together, our findings are consistent with prior research suggesting that while physical activity contributes to better overall sleep, its role in moderating the stress–sleep relationship may not be consistently observed, especially in populations with complex behavioral and psychosocial risk profiles.

This study collected a relatively large sample through a cross-sectional survey to examine the subjective reports of healthy eating, stress, physical exercise, and sleep quality among college students. It analyzed their interrelationships and constructed a moderated mediation model with stress as the independent variable, sleep quality as the dependent variable, healthy eating as the mediator, and physical exercise as the moderator. This study revealed the underlying psychological and behavioral mechanisms between stress and sleep quality among college students and further expanded the theoretical basis for this relationship. Based on the results, this study encourages college students to actively engage in physical exercise to alleviate stress, promote healthy eating, and thereby improve sleep quality. Additionally, future research is recommended to explore the underlying mechanisms through higher-quality studies, such as randomized controlled trials or basic experimental research, to provide theoretical references and evidence for improving sleep quality among college students.

Although this study possesses several strengths, it is not without limitations. First, the cross-sectional design precludes causal inferences and objective associations between variables. Future investigations should incorporate longitudinal tracking combined with objective measurements to establish stronger causal relationships. Second, while simplified measurement tools were employed for cost and feasibility considerations, they may have constrained our capacity to examine the nuanced effects of individual variable dimensions and obtain deeper insights. Additionally, the operational definitions of stress and sleep quality presented to participants exhibited conceptual overlap, potentially confounding their observed relationship [81]. Subsequent studies should employ multidimensional, validated instruments to enhance measurement precision while avoiding circular definitions between constructs. Third, despite the large sample size, the reliance on convenience sampling may have introduced selection bias. Future research should implement more rigorous sampling strategies, including participants from diverse cultural backgrounds, to improve generalizability. Lastly, although our findings indicate that physical activity moderates the stress–diet relationship, further investigation is warranted to determine whether extremely intense or excessive exercise could, under certain conditions (e.g., in high-performance-pressure populations or rigid training regimens), exacerbate rather than alleviate stress.

Conclusions

This study found significant negative correlations between stress and sleep quality, stress and healthy eating, and stress and sleep quality among college students. Healthy eating mediated the relationship between stress and sleep quality, while physical exercise moderated the relationship between stress and healthy eating. It is recommended that universities encourage college students to engage in physical activities and promote healthy eating to alleviate stress and improve sleep quality.

Authors’ contributions

Yacheng Zhu12456, Qing Zhang12345. 1 Conceptualization; 2 Methodology; 3 Data curation; 4 Writing—Original Draft; 5 Writing—Review & Editing; 6 Funding acquisition.

Funding

2025 Shaanxi Province University United Front Theory Research and Practice Project (Project No.: 2025GXTZ45); General Project on Teacher Education Reform and Teacher Development in Shaanxi Province in 2025 (Project No. SJS2025YB070); Xizang Autonomous Region Education Science Research 2023 Key Project (Project Approval No.: XZEDKP230015).

Data availability

The datasets generated and/or analysed during the current study are not publicly available due [our experimental team's policy] but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the Ethics Committee of Xizang Minzu University before the initiation of the project (Grant number: 2024–19). And informed consent was obtained from the participants before starting the program. We confirm that all the experiment is in accordance with the relevant guidelines and regulations such as the declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.