The status and influencing factors of health behavior with first-ever ischemic stroke in young adults: a structural equation model

Abstract

Objectives

To investigate the status of health behavior and its influencing factors in stroke among young adults.

Background

Previous research on young stroke patients has focused mainly on clinical risk factors, with limited attention paid to health behavior. Factors associated with health behavior in younger adults is unclear.

Methods

A cross-sectional survey of 220 Chinese younger adults was conducted. A structural equation model(SEM) identified factors associated with health behavior and their effect magnitude.

Results

The young adult participants had a mean age of 39.62 ± 4.85 years(18 to 44 years), with male accounting for the majority (74.5%, n = 164). Mean the Health Behavior Scale for Stroke Patients (HBS-SP) score was 65.64 (± 12.30). The results showed that personal mastery, social support, health literacy, self-efficacy, health belief, occupation and comorbidities were independently associated with HBS-SP(p < 0.05). SEM showed HBS-SP had the strongest association with personal mastery, followed by health literacy, Health belief, Self-efficacy, social support.

Conclusions

Future research on tailored health behavior interventions should prioritize the enhancement of personal mastery, health literacy, and health beliefs, alongside the strengthening of self-efficacy and the expansion of social support systems. These interconnected psychosocial and contextual factors serve as critical underpinnings for sustainable health behavior change, and targeted efforts to optimize them will likely yield more effective and equitable intervention outcomes in clinical, community, or public health settings.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12883-026-04842-4.

Introduction

The World Stroke Organization (WSO) has projected a continued increase in the global burden of stroke, particularly among young and middle-aged individuals (i.e., those under 55 years). What was once considered a condition mainly affecting the elderly has now become a major health issue in younger populations. In China, stroke among young adults—defined here as individuals under 45 years—is associated with substantial morbidity, and its prevalence is on the rise [1]. Between 1990 and 2021, the age-standardized incidence rate of stroke among Chinese adults aged 20–49 years increased from 18.28 to 24.93 per 100,000 [2]. By 2035, the incidence of ischemic stroke among young adults in China is expected to rise further, reaching 25.40 per 100,000 [3]. Since young adults often serve both as wage earners and caregivers, stroke in this group can lead to reduced productivity, loss of life expectancy, and considerable socioeconomic consequences.

Behavioral factors significantly influence health outcomes, and numerous studies have highlighted the link between modifiable risk factors and stroke. Compared with the general population, younger stroke survivors exhibit a higher prevalence of unhealthy lifestyle behaviors, including daily smoking, obesity, low physical activity (< twice/week), excessive salt intake, and frequent alcohol consumption [4]. Meta-analyses have further quantified the associated risks: smoking increases the risk of recurrent stroke by 1.21-fold, while alcohol consumption elevates the risk by 1.28-fold [5]. Moreover, excessive leisure sedentary time (≥ 8 h/day) is associated with a 4.5-fold increased risk of long-term stroke [6]. Health behaviors refer to a set of actions individuals take to promote better health [7, 8]. Although evidence confirms that positive health behaviors improve stroke prognosis, current management of these behaviors in stroke patients remains inadequate [9]. One cross-sectional study revealed that 78.4% of stroke patients exhibited moderately low levels of health-promoting behavior [10]. Given that stroke is closely tied to behavior, adopting a healthy lifestyle can effectively reduce both its occurrence and recurrence.

However, young adults who have experienced a stroke often struggle to maintain healthy behaviors. Previous research on young stroke patients has focused mainly on clinical risk factors, with limited attention paid to health behavior. It is therefore essential to identify factors associated with health behaviors in this population before developing and implementing targeted interventions.Therefore, this study aims to investigate the status of health behavior and its influencing factors in stroke among young adults, based on the initial hypothetical theoretical framework (Fig. 1).

Fig. 1.

Path diagram of the hypothetical model

Materials and methods

Design and participants

This was a cross-sectional study. From February to July 2025, the convenience sampling method was adopted to recruit patients from the inpatient departments of the Xuanwu hospital capital medical university in Beijing, China. The inclusion criteria were as follows: (1) Aged between 18 and 44 years [11]; (2)Diagnosed with first-ever ischemic stroke, confirmed by CT or MRI [12]. (3)Conscious and able to communicate effectively; (4)Willing to participate in the study and having provided written informed consent. The exclusion Criteria: (1)History of severe mental disorders; (2)Comorbidities such as severe heart failure, respiratory failure, or malignant tumors (i.e., end-stage diseases). The Ethics Committee of Xuanwu hospital capital medical university approved the study(2025-127-002). Written informed consent was obtained from patients before entering the study.

Measured variables

Demographic characteristics

A self-designed questionnaire( Supplementary material 1 ) that consisted of age, sex, education, marital status, healthcare payment, monthly family income, and comorbidities was used to collect demographic characteristics of the patients.

Health behavior

The Health Behavior Scale for Stroke Patients (HBS-SP), created by Wan Lihong et al. in 2017 [13]. It assesses health behaviors across six core dimensions: nutrition, physical activity, medication adherence, adherence to medical instructions, personal responsibility, and tobacco and alcohol use. The scale consists of 25 items, each rated on a 4-point Likert scale ranging from 1 (“never”) to 4 (“always”). The total score is positively correlated with the overall level of health behaviors, where a higher score indicates better practices. A mean item score of 2.5 or above is classified as a moderate level of health behavior. The Cronbach’ s αa values of this scale was 0.88 [14].

Health literacy

Health literacy scale for stroke patients(HLSSP) was developed by Liu Jiru et al. in 2020 [15]. The version used in this study consists of 20 items across three dimensions: basic knowledge, health behavior practices, and fundamental competencies. A five-point Likert scale is used for scoring, ranging from 1 (“strongly disagree”) to 5 (“strongly agree”). The total score ranges from 20 to 100, with higher scores indicating better health literacy among patients. The scale with a Cronbach’s α of 0.929.

Health belief

The Champion Health Belief Model Scale (CHBMS) was originally developed by Champion in 1984 [16]. This study utilized a stroke-specific health beliefs scale adapted from the original CHBMS by Peng Huijiao in 2010 [17]. The adapted scale consists of 36 items across six dimensions: perceived susceptibility, perceived severity, perceived benefits, perceived barriers, health motivation, and self-efficacy. Responses are collected using a 5-point Likert scale, ranging from “strongly disagree” (1 point) to “strongly agree” (5 points). The original version of this adapted scale reported a Cronbach’s α coefficient of 0.894.

Social support

Social support was assessed using the Social Support Rating Scale (SSRS), a tool constructed by Xiao and colleagues in 1994 [18]. This scale comprises three core domains: subjective support, objective support, and social support utilization, with a total of 10 items. Total scores on the SSRS range between 12 and 66, where higher scores reflect a greater degree of social support obtained by the participants. The reliability of the scale was verified with a Cronbach’s α coefficient of 0.8 [19].

The fear disease of progression

The Fear of Progression Questionnaire Short Form (FoP-Q-SF), developed by Mehnert et al. in 2006, is used to assess patients’ fear of disease progression [20]. It consists of 12 items across two dimensions: physical health and social family. Each item is rated on a 5-point scale from “never” (1 point) to “always” (5 points), yielding a total score ranging from 12 to 60. Higher scores indicate a more severe fear of disease progression. A total score of 34 or higher suggests that the fear of progression has led to significant psychological distress. The Cronbach’s αof the scale was 0.886.

Personal mastery

The Personal Mastery Scale (PMS), developed by Pearlin et al., is used to assess an individual’s perceived level of control over their life and personal events [21]. The scale comprises 7 items, each rated on a 5-point Likert scale ranging from 1 to 5. The total score ranges from 7 to 35, with higher scores indicating a stronger sense of personal mastery. The scale with a Cronbach’s α of 0.882.

Self-efficacy

The General Self-Efficacy Scale (GSES) was formulated by Schwarzer et al., a team of German researchers [22]. Comprising 10 items, the scale adopts a 5-point Likert-type scoring system, with response options spanning from “completely incorrect” (assigned 1 point) to “completely correct” (assigned 5 points). Total scores on the GSES range between 10 and 40, with higher scores reflecting stronger self-efficacy among participants. The scale exhibited favorable internal consistency, as evidenced by a Cronbach’s α coefficient of 0.87.

Prior to the official investigation, a pre-survey was conducted. Specifically, 10 young patients who met the study’s eligibility criteria were recruited via convenience sampling. Based on the issues identified during the pre-survey, the questionnaire content was appropriately revised, and the corresponding data collection methods were adjusted to optimize survey quality.

Data collection and quality control

Comprehensive quality control was implemented throughout data collection to ensure research reliability. Pre-survey, all researchers received standardized training on survey procedures, questionnaire interpretation, and participant communication. The study’s purpose, procedures, and ethical considerations were fully explained to participants and their guardians; written informed consent was obtained from guardians, and verbal assent from participants, in line with human research ethical guidelines.

During on-site surveys, researchers provided only basic guidelines for questionnaire completion. All questionnaires were completed and collected immediately, with real-time checks for completeness and logical consistency—missing or invalid information was promptly supplemented.

Post-collection, data exported from Questionnaire Star were double-entered by two independent staff and cross-validated against original questionnaires to minimize entry errors. Strict adherence to predefined inclusion and exclusion criteria was maintained throughout participant recruitment to ensure sample representativeness.

Statistical analysis

SPSS 26.0 and Mplus 8.3 software were utilized for data analysis. Descriptive statistics were applied to characterize all study variables, with scale scores expressed as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) and nonparametric tests were employed to explore factors correlated with health behavior. Multiple linear regression analysis was performed to pinpoint key determinants influencing health behavior, while structural equation modeling (SEM) was adopted to examine the comprehensive influencing mechanism of health behavior. All statistical tests were two-tailed, and a p-value < 0.05 was deemed statistically significant.

Results

Demographic information

For the present study, a total of 236 questionnaires were administered, with 220 valid responses retrieved, yielding an effective response rate of 93.2%. The young adult participants had a mean age of 39.62 ± 4.85 years, ranging from 18 to 44 years. The gender distribution of the sample was skewed, with male accounting for the majority (74.5%, n = 164). A majority (90.0%) of the participants were married, while 88.6% held an educational attainment of high school or lower. Additionally, 51.8% (n = 114) of the respondents had been diagnosed with comorbidities. Detailed sociodemographic characteristics of the young adult participants are summarized in Table 1.

Table 1.

The general information of young adults

Characteristics	N = 220(%)	Characteristics	N = 220(%)
Gender		Monthly income(yuan)
Male	164(74.55)	< 2000	2(0.91)
Female	56(25.45)	2001–4000	10(4.55)
Age(year)		4001–6000	152(69.09)
18–20	4(1.82)	> 6000	56(25.45)
21–30	7(3.18)	Occupation
31–40	91(41.36)	mental labor	125(56.82)
41–44	118(53.64)	physical labor	79(35.91)
Average Age(SD)	39.62(4.85)	unemployed	16(7.27)
Marital status		Comorbidities
Married	198(90.00)	Yes	114(51.82)
Unmarried	19(8.64)	No	106(48.18)
Divorced	3(1.36)	Smoking
Education level		Yes	44(20.00)
Elementary school	71(32.27)	No	176(80.00)
Middle school	68(30.91)	Drinking
High school	56(25.45)	Yes	47(21.36)
Undergraduate	21(9.55)	No	173(78.64)
Bachelor’s degree or above	4(1.82)

Current health-related factors status

Table 2 summarizes the scores of health-related factors among young adults. The HBS-SP had a mean score of 65.64 (SD = 12.30). For the HLSSP, the mean was 47.80 (SD = 7.21). The CHBMS showed a relatively high average of 82.34 (SD = 5.99). On the SSRS, participants scored an average of 41.33 (SD = 7.95). The FoP-Q-SF yielded a mean of 58.29 (SD = 3.95). Regarding the PMS, the average score was 23.49 (SD = 5.07). The GSES had a mean of 27.66 (SD = 3.54). Overall, these results provide a quantitative overview of the current status of various psychological and behavioral health factors in this population.

Table 2.

The Health-related factors scores in young adults

Variables	Mean	SD	Range
HBS-SP	65.64	12.30	52
HLSSP	47.80	7.21	28
CHBMS	82.34	5.99	35
SSRS	41.33	7.95	43
FoP-Q-SF	58.29	3.95	25
PMS	23.49	5.07	28
GSES	27.66	3.54	24

HBS-SP Health Behavior Scale for Stroke Patients, HLSSP Health literacy scale for stroke patients, CHBMS The Champion Health Belief Model Scale, SSRS Social Support Rating Scale, FoP-Q-SF The Fear of Progression Questionnaire Short Form, PMS The Personal Mastery Scale, GSES The General Self-Efficacy Scale

Multiple linear regression analysis of factors associated with HBS-SP in young adults

To explore the factors correlated with HBS-SP among young adults with AIS, we incorporated statistically significant sociodemographic variables, along with scores from HLSSP, CHBMS, SSRS, FoP-Q-SF, PMS, and GSES, as independent variables into the multiple linear regression analysis, adopting the “stepwise” selection method. The results indicated that HLSSP, CHBMS, SSRS, PMS, GSES, occupation, and comorbid chronic diseases were independently associated with health behaviors in this young population (p < 0.05), collectively explaining 97.1% of the variance (adjusted R² = 94.3%) (Table 3).

Table 3.

Multiple linear regression analysis of different factors communicable health behaviors among young adults

Variable	β	S.D.	Beta	T	P
HLSSP	0.316	0.056	0.185	5.629	< 0.001
CHBMS	0.833	0.093	0.406	8.948	< 0.001
SSRS	0.305	0.035	0.197	8.783	< 0.001
PMS	0.348	0.084	0.143	4.168	< 0.001
GSES	0.663	0.132	0.190	5.015	< 0.001
occupation	−0.707	0.330	−0.034	−2.141	0.033
comorbidities	1.707	0.442	0.069	3.861	< 0.001

HBS-SP Health Behavior Scale for Stroke Patients, HLSSP Health literacy scale for stroke patients, CHBMS The Champion Health Belief Model Scale, SSRS Social Support Rating Scale, FoP-Q-SF The Fear of Progression Questionnaire Short Form, PMS The Personal Mastery Scale, GSES The General Self-Efficacy Scale

SEM

Mediation model test

Subsequently, SEM was employed to examine the interrelationships between sociodemographic variables, HLSSP, CHBMS, SSRS, FoP-Q-SF, PMS, GSES, and HBS-SP. An initial conceptual model was constructed to assess model fit, with the evaluation encompassing key fit indices: chi-square/degrees of freedom ratio (χ²/df), root mean square error of approximation (RMSEA), comparative fit index (CFI), incremental fit index (IFI), goodness-of-fit index (GFI), normed fit index (NFI), and adjusted goodness-of-fit index (AGFI). In addition to these indices, modification indices, factor loadings, covariance estimates, and correlation coefficients were also analyzed to refine the model (Fig. 2). The values of these fit indices indicated that the model only partially fit the data: x² /df = 19.578, RMSEA = 0.291, CFI = 0.966, GFI = 0.9376, NFI = 0.965, IFI = 0.966, and AGFI = 0.410. The x²/df value (which is considered among the most important goodness of fit model indices) of 7.548 was lower than the accepted value(≤ 2 or 3). The initial SEM of factors associated with HBS-SP in younger adults is presented in Fig. 2.

Fig. 2.

The initial structural equation model of the factors influencing of the health behaviour in younger adults

Final model

Path analysis of the model of factors associated with HBS-SP in younger adults

To correct the model indices, a pathways with a non-significant pathway coefficient value was deleted. The main significant pathways of HBS-SP were as follows: SSRSGSESHBS-SP, indirect effect(0.11 × 0.64 = 0.07); SSRSCHBMHBS-SP, indirect effect(0.15 × 0.79 = 0.12); SSRS HBS-SP, direct effect(0.33); PMSGSESHBS-SP, indirect effect(0.46 × 0.64 = 0.29); PMSCHBMHBS-SP, indirect effect(0.51 × 0.79 = 0.40); PMS HBS-SP, direct effect(0.36); HLSSPGSESHBS-SP, indirect effect(0.13 × 0.64 = 0.08); HLSSPCHBMHBS-SP, indirect effect(0.47 × 0.79 = 0.37); HLSSP HBS-SP, direct effect(0.36); CHBMSHBS-SP, direct effect(0.79);GSESHBS-SP, direct effect(0.64). Then, the final model was obtained (Fig. 3); path analysis results are presented in Table 4. The final model fit reached the standard: x2/df = 1.862, RMSEA = 0.063, CFI = 0.999, GFI = 0.997, NFI = 0.999, IFI = 0.999, AGFI = 0.941.

Fig. 3.

The final structural equation model of the factors influencing of health behaviour in younger adults

Table 4.

The final model pathway coefficient for factors influencing of health behaviour in younger adults

pathway	Estimate	S.E.	C.R.	P
Self-efficacy <— social support	0.107	0.015	7.332	***
Health belief <— social support	0.150	0.020	7.353	***
Health belief <— personal mastery	0.512	0.030	17.215	***
Self-efficacy <— personal mastery	0.457	0.021	21.533	***
Self-efficacy <— health literacy	0.125	0.016	7.804	***
Health belief <— health literacy	0.473	0.023	21.040	***
Health behavior <— Self-efficacy	0.636	0.135	4.702	***
Health behavior <— Health belief	0.793	0.097	8.213	***
Health behavior <— health literacy	0.361	0.058	6.172	***
Health behavior <— social support	0.325	0.036	9.118	***
Health behavior <— personal mastery	0.364	0.090	4.053	***

***P < 0.01

Effect size analysis of a model of factors associated with HBS-SP in younger adults

Factors correlated with HBS-SP among young adults were ranked based on the magnitude of their predictive effects, in the following order: personal mastery (β = 1.061), health literacy (β = 0.816), health belief (β = 0.793), self-efficacy (β = 0.636), and social support (β = 0.512). Among these variables, PMS exhibited the largest total effect on HBS-SP scores in young adults. Further analysis revealed that self-efficacy and health belief exerted mediating effects in the relationships between personal mastery, social support, health literacy, and HBS-SP.

Discussion

This study demonstrates a moderate-to-high level of health behavior among young stroke patients, likely reflecting well-established habits such as regular physical activity and individualized dietary measures—often integrated into post-stroke rehabilitation and secondary prevention strategies. In contrast, health literacy was found to be low to moderate. This discrepancy may stem from a lack of patient-centered health education, as well as exposure to fragmented health information online, which impedes the accurate comprehension and application of stroke-related knowledge. It also can be attributed to the fact that 88.6% of the participants have a high school education or lower, as lower educational attainment is strongly associated with reduced health literacy. This educational background often limits an individual’s ability to access, understand, and apply health information effectively. Health belief scores were relatively high, suggesting that young stroke patients generally perceive health threats as serious and value preventive behaviors—a perspective likely strengthened by their personal experience with an acute cardiovascular event and ongoing encounters with health risk messaging. Social support was moderate. While young adults typically maintain extensive social networks, the predominantly virtual nature of their interactions may fail to deliver the practical and emotional support essential during recovery, creating a gap between available and perceived support. Notably, fear of disease progression scored extremely high, highlighting profound anxiety regarding stroke recurrence or functional decline [23]. This may be intensified by the psychological impact of experiencing a life-threatening event at a young age, increased awareness of health vulnerability, and frequent online exposure to distressing health-related content.

The path analysis revealed that social support (SSRS) is direct or indirect positive associated with health behavior. Patients at this stage are usually in a period of career development and heavy family responsibilities. They are psychologically independent yet vulnerable to unhealthy lifestyles, and these characteristics interact dynamically with social support. Adequate family and social support can relieve financial pressure, reduce anxiety and depression, improve self-efficacy and disease acceptance, promote healthy behaviors and active rehabilitation, and lower relapse risk. Conversely, career disruption, financial strain, and psychological barriers in young patients limit access to social support. Their desire for independence may lead to avoidance of support, and high risks of social isolation further weaken its effectiveness [24, 25]. The direct association can be explained by the practical resources and emotional encouragement provided by social support—for example, family members or friends may accompany individuals to exercise, remind them of healthy dietary habits, or offer financial support for health-related activities [26]. Such tangible and intangible support reduces the barriers to engaging in healthy behaviors and enhances adherence. Recent findings further underscore the relevance of social context during adolescence: a higher proportion of prosocial friends and a supportive peer environment were linked to fewer behavioral problems, and regular physical exercise was associated with the lowest mental health burden [27, 28]. These observations align with the pathways identified in the present study. The indirect pathways operate through the mediating roles of self-efficacy and health beliefs. From a social cognitive theory perspective, positive social interactions (e.g., praise for healthy behaviors, sharing of successful health experiences) can be linked to greater confidence in individuals’ ability to perform healthy behaviors [29]. Additionally, social support from significant others can lead to stronger individuals’ beliefs about the value of health behaviors, thereby relating to greater their willingness to adopt and sustain such behaviors. These findings align with previous studies highlighting social support as a critical protective factor for health behavior, as it addresses both the practical and psychological dimensions of behavior change [30].

Personal mastery (PMS) exhibits strong direct and indirect positive associations on health behavior, with the indirect pathway via health beliefs being the most prominent. Young adult stroke patients, who often belong to the “sandwich generation” facing employment pressure, parenting and elderly care burdens, suffer from great life stress and role dysfunction. Such stress significantly impairs their perceived personal control and self-efficacy, which further leads to poor medication adherence, inadequate rehabilitation participation and unhealthy lifestyles. In contrast, restoring personal control through phased return-to-work, family role adaptation and structured rehabilitation can promote positive health behaviors and better functional recovery. Thus, personal control acts as a key mediator between age-specific life stressors and health behaviors in young stroke patients, supporting the necessity of targeted psychosocial and behavioral interventions [31, 32]. The direct association of personal mastery stems from its core nature: individuals with high personal mastery perceive themselves as capable of controlling their health outcomes and believe that their actions can significantly influence their health status. This internal locus of control corresponds to proactive engagement in healthy behaviors, and persist in behaviors despite obstacles [33]. The indirect pathways through self-efficacy and health beliefs are consistent with self-determination theory, which emphasizes autonomy, competence, and relatedness as key drivers of intrinsic motivation. High personal mastery fulfills the need for competence, as individuals feel capable of mastering health-related tasks, which in turn related to self-efficacy. Furthermore, individuals with strong personal mastery are more likely to recognize the causal link between their behaviors and health outcomes, corresponding to health beliefs. This heightened belief in the effectiveness of health behaviors further related to greater motivation to act. The robust indirect pathway via CHBMS suggests that personal mastery is most strongly associated with health behavior through shaping individuals’ core health beliefs, underscoring the importance of internal psychological resources in driving behavior change.

Health literacy (HLSSP) demonstrates significant direct and indirect positive associations on health behavior. The direct effect is rooted in the functional role of health literacy: individuals with higher health literacy can effectively access, understand, evaluate, and apply health information. For example, they can interpret nutrition labels, understand the benefits of regular physical activity, and distinguish evidence-based health advice from misinformation—enabling them to make informed decisions about health behaviors [34]. In contrast, individuals with low health literacy may struggle to navigate complex health information, corresponding to inappropriate behaviors. The indirect pathway of health literacy are mediated by self-efficacy and health beliefs. From an information-processing perspective, high health literacy enhances individuals’ understanding of how to perform healthy behaviors, which relates to their confidence in their ability to execute these behaviors [35]. Additionally, health literacy helps individuals recognize the long-term benefits of healthy behaviors and the risks of unhealthy ones, thereby relating to stronger their health beliefs. The stronger indirect pathway via CHBMS compared to GSES suggests that health literacy is more strongly associated with health behavior through shaping individuals’ beliefs about health, rather than directly relating to self-efficacy. This highlights the need for health education interventions to not only improve knowledge acquisition but also emphasize the practical relevance and benefits of health behaviors to strengthen health beliefs.

Self-efficacy (GSES) and health beliefs (CHBMS) exert strong direct positive associations with health behavior and serve as key mediators in the relationships between social support, personal mastery, health literacy, and health behavior. The direct associations of these two variables are consistent with the Health Belief Model and social cognitive theory, which posit that individuals’ intentions to engage in health behaviors are related to their beliefs about the behavior’s effectiveness and their confidence in their ability to perform the behavior. Specifically, high self-efficacy corresponds to greater ability to overcome barriers and persist in healthy behaviors, while strong health beliefs related to individuals to initiate and maintain healthy behaviors [36]. As mediating variables, self-efficacy and health beliefs act as “psychological bridges” connecting external and internal factors to health behavior [37]. Social support, personal mastery, and health literacy—though distinct in nature—all correspond to higher self-efficacy and health beliefs, which in turn related to improved health behavior. For example, social support provides external validation and resources that related to stronger self-efficacy; personal mastery fosters an internal sense of control that strengthens health beliefs; and health literacy provides the knowledge foundation that supports both self-efficacy and health beliefs [38]. The strong mediating roles of self-efficacy and health beliefs highlight their centrality in the model—suggesting that interventions targeting health behavior should prioritize enhancing self-efficacy and health beliefs to maximize the predictive and practical relevance of other modifiable factors.

Limitations

Limitations of this study include its cross-sectional design, which precludes the establishment of causal relationships, and the reliance on self-reported measures, which may be subject to response bias. Future research should adopt longitudinal designs to explore the dynamic changes in these factors and their potential long-term relationships with health behavior, and incorporate objective measures to validate self-reported data. Second, the reliance on a convenience sample from a single tertiary hospital in Beijing introduces potential selection bias. While this approach facilitated data collection, it inherently limits the external validity of the study. Therefore, our findings may not be directly transferable to young stroke survivors in rural communities, where rehabilitation resources and cultural attitudes toward stroke may differ significantly. Future multi-center studies with more diverse, population-based sampling strategies are needed to validate and extend our conclusions. Third, This study had an imbalanced gender distribution, which may limit the generalizability of the findings to female populations.

Conclusion

This study comprehensively explores the current status of health-related factors and their associative pathways with health behavior among young adults. The findings indicate that young adults exhibit moderate-to-high levels of health behavior. Path analysis reveals that social support, personal mastery, and health literacy are associated with health behavior through both direct and indirect pathways, with self-efficacy and health beliefs serving as critical mediators. These findings have important practical implications for developing targeted health promotion interventions for young adults.

Authors’ contributions

K.F.: experimental design; data curation and analysis; writing-original. T.Y.:data collection. H.Y.: data collection. Y.Q.:experimental design; writing-review and editing. H.C.:experimental design; writing-review and editing. All authors reviewed the manuscript.

Funding

This research was supported by Scientific Research Promotion Program for Clinical Medicine (2024CMFA09).

Data availability

The datasets utilized and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Xuanwu Hospital, Capital Medical University (2025-127-002) and was performed in accordance with the Declaration of Helsinki. Prior to the initiation of the study, patients provided written informed consent for themselves.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.