China launched its 5G network services in June 2019, ushering in an era of enhanced data transmission, reduced latency, and expanded connectivity. These innovations have provided significant technological support and opened new opportunities in sectors such as the Internet of Things (IoT), industrial automation, autonomous vehicles, telemedicine, and virtual reality (1). The rapid evolution of mobile networking, along with decreasing smartphone prices, has made smartphones the primary means of internet access. Consequently, smartphone penetration in China has steadily increased, reflecting their growing adoption across the population.

According to the 53rd Statistical Report by the China Internet Network Information Center (CNNIC), China’s internet user base reached 1.092 billion, with an internet penetration rate of 77.5%. Among these, 99.9% were mobile internet users, totaling 1.091 billion. The 10–29 age group constitutes 28.4% of the online population (2), with college students making up 21.0% of users (3). On average, these students spend 26.1 h per week online (2). As internet usage becomes an indispensable part of modern life, the growing reliance on smartphones has led to smartphone addiction (SA), a nonsubstance-related addiction (4, 5) negatively affecting physical and psychological health, social interactions (6), and academic performance, with severe cases linked to suicidal ideation (7).

Physical activity (PA) plays a key role in promoting health and well-being. Research shows PA alleviates stress (8), reduces symptoms of depression and anxiety (9–11), and enhances overall mental health. In college students, PA is an essential countermeasure to sedentary behaviors driven by academic demands and extensive digital engagement (12). It fosters resilience and helps mitigate the adverse effects of prolonged screen time, providing a healthy outlet for stress and maintaining psychological balance.

A recent study showed that college students often fail to meet recommended PA levels, partly due to prioritizing academic success under parental expectations. This highlights the need to promote PA as a protective factor against excessive smartphone use (13).

The psychological effects of SA are well-documented, with research exploring its impact on anxiety (14, 15), depression, stress (15), self-esteem (16), and social isolation (5). However, the relationship between PA and SA remains under-explored. Previous research by our team found an inverse association between PA, exercise intensity (EI), and internet addiction (IA) among college students. Higher PA levels were linked to a lower risk of IA (17), suggesting a similar relationship may exist between PA and SA. This aligns with recent findings showing that PA and sedentary behavior are correlated with academic performance among college students, particularly freshmen (18).

The potential of PA to reduce SA among college students warrants further investigation, especially given the limited evidence in this area.

Western China, as defined by China’s economic and geographical classifications, consists of 12 administrative regions, including Chongqing, Sichuan, Yunnan, and others. These areas account for 27.2% of the country’s population (19), but their economies are less developed compared to Eastern and Southern China.

Western China presents a unique context for studying the relationship between PA and SA. The region is characterized by diverse cultural practices, socio-economic disparities, and distinct lifestyle behaviors. The geographical and economic challenges may influence both PA levels and smartphone usage patterns. Despite these features, Western China has received limited attention in SA and PA research. By focusing on college students in this region, this study aims to fill this gap and provide insights for culturally and regionally tailored interventions to promote healthier behaviors.

A 2015 report highlighted that international Chinese students are particularly vulnerable to SA, which significantly affects them emotionally and financially (20). This vulnerability may stem from the transition to university life and the unfamiliarity of the new environment. A more recent report from Sichuan Province found that SA is closely associated with self-control issues and rumination, especially in the context of psychological distress, such as loneliness (21).

This study aims to explore the correlation between PA, EI, and SA, specifically investigating how PA and EI impact SA to better understand the nature of this relationship.

This study explores the link between PA, EI, and SA among college students in Western China. Findings reveal an inverse relationship: students engaging in higher-intensity PA had a lower risk of SA. This remained significant after adjusting for confounders, with subgroup analyses confirming the robustness of the results across demographic and lifestyle factors. Promoting PA may effectively manage SA, aligning with research on PA’s protective role against behavioral addictions like internet addiction (20, 36). The large sample size enhances the reliability of these findings.

Although IA and SA share similarities, a key distinction is that smartphones are more portable and convenient to use, whereas computers often provide greater power for gaming with enhanced psychological rewards. Supporting this, Rozgonjuk et al. found that while risk factors generally had stronger associations with problematic internet use than with smartphone use, fear of missing out was more strongly linked to problematic smartphone use (37).

Our findings are in line with previous studies from other regions in China, such as Central (38) and Northern China (39), showing a similar pattern among college students. For instance, research in Central China demonstrated a significant reduction in SA after students engaged in basketball and Baduanjin for 12 weeks, with benefits sustained for an additional 2 months (40). This effect, observed in students from a different geographical context, highlights the potential of PA as a preventive approach to SA across various Chinese ethnics/races, consistent with the idea that the benefits of PA in mitigating SA also apply to college students in Western China.

Findings from Taiwan region contrast with ours, as a six-month longitudinal study reported no significant link between PA and SA (41). These differences may reflect cultural variations in exercise habits, smartphone use, attitudes toward PA, or age. Similar inconsistencies between mainland China and Taiwan region in health behaviors (36, 42) suggest that distinct forms of PA or fitness norms among mainland students may influence the PA-SA link. Further research should explore these cultural differences to validate our findings.

The analysis of EI highlights PA’s role in managing SA, as students engaging in moderate-to-vigorous PA had significantly lower odds of SA. Higher-intensity PA may offer protective benefits by fostering resilience and improving mood, reducing smartphone dependency (43, 44). Vigorous PA promotes endorphin release, which buffers against stress and anxiety linked to SA (20).

The inverse association between PA and SA remained consistent across subgroups, including gender, ethnicity, academic discipline, and social factors, suggesting PA’s protective effect is broadly applicable. This indicates that PA’s impact on SA likely stems from fundamental psychological or physiological mechanisms, rather than context-specific factors. The similar benefits across relationship statuses and academic disciplines support PA as a general preventive measure, due to its wide-ranging mental health benefits (45, 46). This uniformity suggests that PA interventions can be widely implemented without extensive customization, making them accessible to large student populations.

Our comparisons with studies from Central and Northern China, as well as Taiwan region, highlight the importance of cultural and environmental contexts in PA and SA research. Mainland Chinese students may prioritize fitness differently from students in Taiwan, influenced by unique societal expectations, which could shape how PA affects SA. Cultural factors significantly influence health behaviors, explaining the differing impact of PA across these populations (36, 42). Future research should explore these cultural influences to tailor intervention programs to regional behaviors and preferences.

Specific interventions could include tertiary institutions promoting greater student participation in physical activity through educational campaigns, providing suitable facilities, and organizing competitive sporting events (47).

It is important to note that China is divided into six major regions—North China, Northeast China, East China, South Central China, Southwest China, and Northwest China—by the National Bureau. This regional division helps better understand the nation’s characteristics, including its economy, climate, geography, governance, and administrative divisions (48).

Our current findings were collected solely from Western China, which has a disadvantaged economic status, partly due to its geographic features, such as high mountains and desert regions, as well as lower educational opportunities. Students from these areas may face less academic pressure due to fewer “tiger parents” in the region. While our data provides valuable insights into PA and SA, it may not be generalizable to students across all of China or the world. To address this limitation, we plan to conduct future longitudinal studies with larger cohorts from various regions of China—Eastern, Western, Southern, Northern, and Central—using a well-designed approach. Furthermore, we will minimize bias in cohort recruitment by accounting for factors such as ethnicity, sex, age, study years, academic discipline, parental education levels, and professions. This methodology is expected to provide more reliable and representative outcomes that can be generalized more effectively.

Since self-response bias is a key issue in psychometrics, Rosenman et al. developed a unique stochastic frontier estimation (SFE) method that substantially reduces, if not eliminates, this bias (49). We plan to adopt this system in our future study.

Although the current study’s cross-sectional design cannot establish causality between PA and SA, it reveals a significant association. The data cannot confirm whether PA directly reduces SA or if students with lower SA are more likely to engage in PA. Unmeasured factors, such as personality traits, lifestyle habits, or support systems, may influence both PA and SA. Previous research emphasizes the importance of longitudinal and experimental designs to clarify causality in behavioral addiction studies (27, 50). Future research using longitudinal and intervention-based approaches could better assess the causal relationship between PA and SA.

Our findings suggest mechanisms through which PA might reduce SA, including stress reduction, mood enhancement, and increased social interaction. PA has been shown to decrease stress, anxiety, and depressive symptoms, which could reduce students’ reliance on smartphones as coping tools (43, 51). Team-based PA also provides opportunities for socialization, offering an alternative to smartphone-based interactions and potentially reducing excessive smartphone use (44). These psychological benefits support PA’s inclusion in mental health and SA prevention programs, reinforcing its role in student well-being.

In conclusion, promoting PA, especially at higher intensities, could help manage SA among college students. By integrating these findings into university wellness programs, institutions can address SA and encourage healthier lifestyles. Future research should explore cultural variations, causative mechanisms, and tailored interventions to maximize PA’s impact on reducing SA.