Introduction

Substance Use Disorders (SUDs) present a global public health dilemma, primarily due to complex brain disorders associated with recreational substance use1. Repeated or prolonged substance consumption leads to significant changes in the brain's neuronal morphology, structure, and functionality. These changes not only deteriorate individual health but also obstruct societal progress by leading to extreme behaviors such as robbery, theft, aggression, and self-harm2. Despite deploying a variety of interventions to address the health consequences of substance abuse and dependence, the consistently high rates of relapse remain a significant challenge. This issue not only compromises the success of rehabilitation efforts but also inflates the costs associated with rehabilitation services3. Relapse, which is the resumption of substance seeking after successful detoxification4, stands as a formidable barrier to addiction recovery. Surveys report alarmingly high relapse rates: 78.2% among heroin users, 61.9% for cocaine users, and 52.2% for methamphetamine users5. Hence, pinpointing the factors that trigger relapse, alongside comprehending the mechanisms and boundary conditions at play, is crucial. Such understanding is indispensable for devising precise intervention strategies, ultimately enhancing the effectiveness of substance addiction treatment endeavors.

Stress and relapse

Stress triggers both physical and mental adjustments in humans to manage real or perceived threats and restore balance. Individuals with SUDs, upon reintegration into society after detoxification, experience heightened stress levels compared to the general population, due to stigmatization by family and friends, employment and lifestyle stressors, and financial hardships6,7,8,9. This stress negatively affects the brain's hippocampus, amygdala, and prefrontal cortex, leading to damage that exacerbates cravings and increases the risk of relapse10. According to the negative reinforcement model of addiction, the avoidance of negative emotions, exacerbated by adverse life events, drives addictive behaviors and increases vulnerability to relapse11. Empirical evidence supporting this model comes from research involving 182 individuals with SUDs, which showed that stress levels and craving intensity remained significantly elevated for up to 16 weeks after withdrawal treatment. Furthermore, stress, particularly when combined with environmental substance cues, significantly activates the dopamine system, greatly increasing the likelihood of relapse12. Studies have shown that stress during withdrawal not only boosts substance cravings but can also directly precipitate relapse13. It is therefore evident that stress plays a pivotal role in the relapse of substance users; the greater the stress experienced by addicts, the higher their susceptibility to relapse. However, the specific mechanisms underlying this relationship are yet to be fully understood.

Self – control as a mediator

Self-control, a trait unique to humans, enables individuals to regulate irrational thoughts, emotions, and actions to meet both personal and societal norms14. Evidence suggests a significant negative correlation between self-control and psychological disorders, such as depression, anxiety, and stress15,16,17. The strength model of self-control proposes that tasks requiring self-regulation—such as inhibiting reactions, managing impulsivity, and making social decisions—drain limited mental resources18. Stress management, which also necessitates cognitive and emotional control, is linked to "ego depletion," characterized by reduced effectiveness in self-regulation19. This depletion particularly affects the prefrontal cortex, critical for self-control, by activating the stress response system and triggering cortisol release20. Supporting this, an international study, corroborated by extensive two-week diary analysis, confirmed a consistent negative relationship between self-control traits and stress levels21. Additionally, research on internet addiction shows an association between stress and individuals’ misinterpretation of external challenges or their own coping abilities, which is correlated with failures of self-control22.

Furthermore, enhanced self-control is paramount for resisting temptations, mitigating impulsivity, and prioritizing long-term goals over immediate gratification23. The self-regulation deficit model suggests that a lack of self-control can lead to pathological behaviors24, with individuals exhibiting lower levels of self-control more prone to impulsive, risk-taking, and shortsighted actions, which increases their risk of criminal involvement. Conversely, those with heightened self-control tend to avoid criminal activities by recognizing the long-term consequences and outweighing the risks of such engagement25. Research demonstrates that individuals with SUDs who bolster their self-control exhibit a reduced tendency towards substance use, thus diminishing their relapse rates26. Duckworth27 contends that self-control inherently counters impulsivity, enabling individuals with enhanced self-control to more effectively manage impulsive behaviors. A significant correlation between impulsivity and the likelihood of relapse among individuals with SUDs has been identified28. Nonetheless, the research exploring the mediating role of self-control between stress and relapse in individuals with SUDs is scant, underscoring the necessity for additional direct evidence to validate this relationship.

Social support as a moderator

Social support involves the subjective care and assistance received from close friends or family, covering emotional support (e.g., empathy), instrumental support (e.g., assistance with household tasks), and informational support (e.g., academic guidance)29. The main-effect model of social support asserts that these supportive relationships are vital for reducing negative reactions and bolstering proactive coping strategies in the face of adversity30. This foundational support becomes especially crucial when individuals encounter traumatic events, paving the way for their effective recovery and building resilience.

Social support is crucial for stress recovery in individuals who have experienced trauma or disaster. It boosts self-efficacy, eases social pressures, diminishes negative emotions, and upholds mental health, all of which lower the risk of relapse in substance addiction31,32. Additionally, the positive involvement of family and friends can counter the euphoric effects of substance use and lessen the negative impact of social stress33. Research shows that social support directly reduces the likelihood of relapse by alleviating withdrawal symptoms and enhancing outcomes of cessation efforts34. It also indirectly minimizes relapse risks through the enhancement of exercise self-efficacy and the improvement of health-related quality of life35. Despite the accumulating evidence on the advantages of continuous social support in lowering perceived stress and the probability of relapse, inconsistencies remain. For instance, studies on alcohol dependence reveal that social support does not significantly impact the stress-relapse relationship36, highlighting the complex role of social support in the recovery from substance dependence.

Building on the established importance of social support, recent studies have focused more on the dynamics among stress, social support, and self-control. Pilcher and colleagues argue that social support enhances the necessary resources for effective self-control, especially in stressful contexts37. Research indicates that social support can mitigate acute stress symptoms by satisfying psychological needs and boosting individuals' sense of autonomy, thereby lessening stress's detrimental effects on self-control38. Furthermore, social support, alongside self-acceptance, plays a vital role in enabling individuals to navigate the challenges encountered in substance treatment and rehabilitation, highlighting its moderating effect on the interplay between stress and self-control39. However, while the positive impacts of social support on reducing stress and improving self-control are documented, its moderating role, particularly in substance dependence scenarios, requires more in-depth investigation.

The Present Study

Existing literature consistently demonstrates a significant correlation between perceived stress in individuals with SUDs and their propensity for relapse, yet the mechanisms underlying this relationship and its boundary conditions remain unexplored. This study aims to investigate the mediating role of self-control and the moderating effect of social support on this relationship. Accordingly, we propose the following hypotheses:

H1

Perceived stress in individuals with SUDs directly correlates with their tendency to relapse.

H2

Self-control mediates the relationship between perceived stress and the propensity for relapse.

H3

Social support moderates both the direct impact of perceived stress on the tendency to relapse and the first half of the mediating effect of self-control.

Figure 1 depicts the proposed model for the study.

Fig. 1
figure 1

Proposed model.

Full size image

Methods

Participants

This study recruited 420 male individuals with SUDs (Mage = 38.73, SD = 8.53; minimum = 19, maximum = 59) from rehabilitation facilities in Guangxi, China, through convenience sampling. Due to missing values on the dependent variable (propensity for relapse), the mediating variable (self-control), and the moderating variable (social support), data from 19 participants were excluded, leaving 401 valid questionnaires and achieving a 95.5% recovery rate. Participant demographic characteristics are detailed in Table 1.

Table 1 Descriptives of the sample.
Full size table

Measures

Chinese version of perceived stress scale (Chinese 14-item PSS)40

This questionnaire measures the stress levels respondents have experienced in the past four weeks. The Chinese version includes two dimensions: tension and loss of control, with a total of 14 items, scored from 1 (never) to 5 (always). Higher total scores signify greater perceived stress. The scale demonstrates good reliability with an internal consistency coefficient of 0.85.

Self-control scale (SCS)41

This questionnaire measures five dimensions of self-control: impulse control, resistance to temptation, focus on work or study, establishment of healthy habits, and moderation. It comprises 19 items rated on a five-point Likert-type scale, from 1 (totally disagree) to 5 (totally agree), with 15 items scored inversely. Therefore, higher scores reflect a higher level of self-control. The scale demonstrates excellent reliability, indicated by an internal consistency coefficient of 0.91.

2.2.3 Perceived Social Support Scale (PSSS)42 The scale evaluates three dimensions of social support—family, friends, and others—through 12 items on a 7-point scale, ranging from 1 (strongly disagree) to 7 (strongly agree). Higher scores indicate stronger perceptions of social support. In this study, the scale showed excellent reliability, evidenced by a Cronbach's alpha of 0.95.

Relapse inclination questionnaire (RIQ)43

This scale measures five dimensions: confidence in substance withdrawal, the real influence of substance, the objective environment, the level of physical and psychological harm, and the support system. It comprises 18 items evaluated on a six-point Likert scale from 0 (least severe) to 5 (most severe). Elevated scores indicate a heightened risk of relapse. The scale's reliability in this investigation is confirmed by a Cronbach's alpha value of 0.85.

Process

Two brigades were randomly selected from a male compulsory isolation treatment center in Guangxi Province, with all members completing the questionnaire survey. Participants were adults over 18 years of age without intellectual disabilities. To reduce fatigue effects, the survey was conducted over two distinct days. Graduate students in psychology, with assistance from police officers at the compulsory drug rehabilitation center, carried out the study. They offered standardized instructions to participants, emphasizing the importance of authenticity, independence, and completeness in their responses. The study received approval from the university's Ethical Committee for Scientific Research.

Data analysis

Initially, all study variables were evaluated using questionnaires, and Harman's single factor test was applied to investigate the presence of common method bias. Subsequently, descriptive statistics for these variables were analyzed. Finally, the SPSS Process plugin was employed to perform a moderated mediation model test and bootstrap analysis, designating perceived stress as the independent variable, relapse as the dependent variable, self-control as the mediator, and social support as the moderator.

Ethics statement

The studies involving human participants were reviewed and approved by Ethics Committee of the School of Psychology at the Beijing Sport University. The patients/participants provided their written informed consent to participate in this study.

Informed consent

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000. Informed consent was obtained from all patients to be included in the study.

Results

Common method deviation test

The Harman single factor test facilitated an unrotated exploratory factor analysis of all items. Results revealed that the first common component accounted for merely 20.43% of the total variance, below the 40% criterion established by Zhou & Long44, suggesting a negligible common method bias in this research.

Preliminary analyses

The mean value and standard deviation (M ± SD) for each variable were as follows: perceived stress (2.80 ± 0.57), self-control (3.53 ± 0.54), social support (4.10 ± 1.08), relapse(1.30 ± 0.61). Figure 2 presented the correlation matrix among these variables, which revealed that perceived stress had a negative link with self-control (r = − 0.75, p < 0.01), a positive correlation with relapse (r = 0.41, p < 0.01), and no correlation with social support(r = 0.08, p > 0.05). There was no correlation between self-control and social support (r = 0.03, p > 0.05), however there was a negative connection between self-control and relapse (r = − 0.44, p < 0.01). Relapse was negatively related to social support (r = -0.34, p < 0.01) .With this in mind, we can further explore the interplay between the aforementioned factors.

Fig. 2
figure 2

The correlation matrix among variables is being examined. The lower triangular section of the matrix presents specific correlation coefficient values, whereas the upper triangular section illustrates the correlations' direction and strength via the ellipses' color and shape. Blue ellipses signify positive correlations, and red ellipses indicate negative ones. A darker or more flattened ellipse denotes a stronger correlation. PS: Perceived Stress, PS_1: tension, PS_2: loss of control; SC: Self-Control, SC_1: impulse control, SC_2: resistance to temptation, SC_3: focus on work or study, SC_4: establishment of healthy habits, SC_5: moderation; SS: Social Support, SS_1: family support, SS_2: friends support, SS_3: others support; RI: Relapse Inclination, RI_1: confidence in substance withdrawal, RI_2: the real influence of substance, RI_3: the objective environment, RI_4: the level of physical and psychological harm, RI_5: the support system.

Full size image

Moderated mediation model

First, Model 4 in the PROCESS macro for SPSS was utilized to assess the mediating role of self-control. As indicated in Table 2, perceived stress had a direct predictive effect on the relapse of individuals with SUDs (β = 0.41, p < 0.01). Even after introducing the mediating variable of self-control, perceived stress continued to positively predict relapse (β = 0.22, p < 0.01), negatively predicted self-control (β = − 0.75, p < 0.01), and self-control negatively predicted relapse (β = − 0.26, p < 0.01). A percentile Bootstrap test with bias correction confirmed that the partial mediating effect of self-control was significant, with an indirect effect value of 0.20 and a 95% confidence interval of [0.09, 0.32]. The mediating effect accounted for 48.78% of the total effect.

Table 2 Mediating model of self-control between perceived stress and relapse (n = 401).
Full size table

The moderating effect of social support on the relationship between perceived stress and both self-control and relapse was analyzed using PROCESS Model 8 of the SPSS macroprogram. Results, as shown in Table 3, demonstrated that the interaction between perceived stress and social support significantly predicted self-control (β = 0.17, p < 0.01), indicating that social support moderates the impact of perceived stress on self-control. Similarly, this interaction negatively influenced relapse (β = -0.19, p < 0.01), demonstrating that social support also moderates the effect of perceived stress on relapse.

Table 3 A moderated mediation model.
Full size table

To elucidate the moderating role of social support, it was categorized into high and low groups using the mean plus or minus one standard deviation. Subsequent simple slope analysis revealed distinct outcomes. As depicted in Fig. 3, among individuals with high social support, the negative effect of perceived stress on self-control was statistically significant (β = − 0.50, p < 0.01). This negative effect was more pronounced in the low social support group (β = − 0.87, p < 0.01). Furthermore, Fig. 4 showed that for the high social support group, the effect of perceived stress on relapse was not statistically significant (β = 0.08, p = 0.318). Conversely, in the group with lower social support, perceived stress significantly predicted an increase in relapse (β = 0.48, p < 0.01).

Fig. 3
figure 3

The moderating effect of social support on perceived stress and self-control.

Full size image
Fig. 4
figure 4

The moderating effect of social support on perceived stress and relapse.

Full size image

Discussion

This study establishes a positive correlation between perceived stress and relapse among individuals with SUDs, providing critical insights for drug rehabilitation strategies. However, considering the uncontrollable nature of stress events, this study additionally underscores the critical roles of self-control and social support in mediation and moderation. These elements, unlike stress events, are adaptable and can be strengthened, effectively reducing relapse risk. Enhancing self-control through cognitive-behavioral therapy, emotional regulation training, and willpower-strengthening exercises helps develop effective self-management strategies. Moreover, creating a supportive environment through family therapy, community support groups, and healthy social networks is crucial for recovery. This holistic approach is designed to bolster internal psychological resilience while fostering a supportive external environment, collaboratively minimizing the risk of relapse. Thus, this study provides a theoretical and practical basis for developing more effective relapse prevention measures, specifically emphasizing the improvement of self-control and social support systems, guiding future rehabilitation efforts.

Stress and relapse

This study identified a significant positive correlation between perceived stress and the tendency to relapse among individuals with SUDs, highlighting that increased perceived stress augments the risk of relapse. Supporting evidence reveals that anhedonia, restlessness, depressive symptoms, and heightened stress intensify psychological cravings and the towards relapse45. In stressful situations, individuals with substance addiction might experience a strong motivational state, enhancing the probability of substance reuse36. Stress profoundly affects the dopamine system, crucial for reward and pleasure. Sinha and colleagues' research demonstrates that stress exacerbates substance cravings and relapse risk by impacting the dopamine system13. This implies individuals might use substance to counteract the decreased activity in the reward system induced by stress, underscoring the vital role of stress management in relapse prevention. Moreover, an interaction exists between the stress response system, such as the HPA axis, and the brain's reward circuitry, including the ventromedial prefrontal cortex, amygdala, and nucleus accumbens. Stress activates the HPA axis, raising cortisol levels, which, in turn, influences the reward system and leads to heightened substance cravings. This underlines the importance of biological factors in assessing relapse risk46. Furthermore, stress alters the functioning of neural circuits linked to substance cravings and relapse. Research indicates that stress and substance-related cues activate brain areas associated with substance cravings, like the prefrontal cortex and amygdala, elucidating how stress amplifies substance cravings and the risk of relapse by affecting specific brain regions47.

Future drug rehabilitation programs could include detailed assessments of perceived stress levels among individuals with substance addiction. Identifying and tackling the root causes of stress allows treatment providers to assist these individuals in developing effective coping strategies, consequently minimizing stress's influence on their relapse vulnerability. Moreover, integrating stress management techniques into these programs offers significant benefits. Techniques such as mindfulness meditation48, relaxation exercises49, and stress-reduction activities like yoga50 play a critical role in cultivating healthier stress management coping mechanisms. Equipping individuals with these advanced stress management strategies can significantly reduce the impact of stress on their likelihood of relapse.

The Mediating Role of Self—control

Self-control functions as a partial mediator between perceived stress and the likelihood of relapse in individuals with substance dependency. There is a notable connection between perceived stress and diminished self-regulatory capacity in those with SUDs. Studies suggest that stress correlates with changes in the structure and function of the prefrontal cortex (PFC). Specifically, chronic stress is linked to neuronal atrophy and the activation of microglial cells in the PFC, both of which are associated with local inflammation. This inflammation, in turn, contributes to impairments in PFC functions, adversely affecting cognitive control and decision-making capabilities51. Additionally, chronic stress is intimately tied to reductions in synaptic plasticity and long-term potentiation in the PFC, essential for working memory and behavioral flexibility—key components of self-control52. Furthermore, stress increases the activity of protein kinase C in the PFC. This upsurge is linked to elevated levels of dopamine and norepinephrine, neurotransmitters crucial for managing stress responses and maintaining self-control53. The subcortical balance model emphasizes the crucial role of subcortical emotional processing areas in cognitive control. Under stress or other stimuli, these bottom-up processes are believed to augment the influence of subcortical brain functions in reaction to specific, potent cues, potentially leading to lapses in self-control54. Finally, while glucose supplementation can temporarily enhance self-control, the ongoing energy demands of chronic stress might exhaust glucose reserves, thus undermining self-control55.

On the other hand, self-control was negatively related to relapse. Research identifies addiction as arising from an imbalance between two interacting neural systems: the amygdala system, which triggers immediate pain or pleasure responses, and the reflective prefrontal cortex system, which processes future pain or pleasure signals. Individuals with substance addiction who exhibit higher levels of self-control can utilize the prefrontal cortex to assess the long-term consequences of substance use, effectively reducing the likelihood of relapse. Conversely, a weakened control system may enhance the impulse system's activity, initiating bottom-up, unconscious signals from the amygdala and thereby escalating the difficulty of resisting substance temptations56. Furthermore, the theoretical model of relapse behavior suggests that relapse is influenced by both external factors, like social pressure and environmental triggers, and internal factors, notably the individual's self-control capacity. This insight emphasizes the importance of relapse prevention strategies that bolster the individual's internal psychological resilience in addition to providing external environmental support57.

Based on the findings, treatment programs should prioritize enhancing self-control skills. Integrating evidence-based methods, including cognitive-behavioral therapy, can significantly aid individuals in recognizing and managing triggers, formulating impulse control strategies, and improving problem-solving skills58. This comprehensive approach effectively strengthens self-control and diminishes the risk of relapse.

The moderating role of social support

The moderated mediation model test revealed that social support moderated the direct effect and the first half path of the mediation effect. According to the stress-buffering theory, social support serves as a protective barrier, not just supplementary external assistance but a crucial mechanism influencing the stress response system. This support shields individuals from the adverse effects on physical and mental health due to stressful life events59.In comparison to divorced or widowed individuals, married persons typically take on more family responsibilities and receive support from their spouses and children, enhancing their ability to manage psychological stress60. This stable family environment contributes to the restoration of depleted self-control resources and the success of drug rehabilitation efforts. Additionally, peer support during the withdrawal phase and active participation in support groups significantly strengthen self-control and reduce the likelihood of relapse61,62. Concurrently, government agencies are escalating their efforts to promote a substance-free society by educating the public on the dangers of substance use and implementing strategies to combat illegal substance activities. The reinforcement provided by these social institutions establishes a solid foundation for combating stress and preventing relapse, thus serving as crucial factors in enhancing individual self-control63.

The moderating effect of social support emphasizes the necessity of reinforcing social support systems in substance treatment programs. Treatment providers must proactively involve family members, friends, and support groups in the rehabilitation process. Implementing strategies such as promoting participation in support groups, providing family therapy sessions, and establishing connections with extensive social networks are critical. These actions furnish individuals with the requisite social support to adeptly navigate stress and minimize the probability of relapse.

Limitations

This study's comprehensive exploration of the determinants and mechanisms of relapse in SUDs patients reveals significant insights, yet it is not without its limitations. First, the sample's scope is narrow, drawing primarily on data from male substance addicts in compulsory isolation detoxification centers in China's Guangxi region. This limitation could affect the generalizability of the findings. To address this, future research should broaden the sample to include diverse regions, female addicts, and individuals in non-compulsory isolation settings, thereby enhancing the representativeness and applicability of the results. Second, the study's cross-sectional design hampers the ability to establish causality. A longitudinal approach, which would monitor participants over time, is recommended for future studies to more precisely determine the causal links between perceived stress, self-control, social support, and relapse tendencies. Third, the reliance on self-reported data through validated scales introduces the potential for bias, notably social desirability bias. Future studies should integrate objective measurement techniques, such as physiological indicators or behavioral observations, to secure a more comprehensive dataset.

Conclusion

This study offers new insights into the complex interplay between perceived stress, social support, self-control, and relapse in Chinese substance abusers. It identifies self-control as the key internal mechanism mediating the stress-relapse relationship, while social support sets the contextual conditions. These findings are instrumental in designing psychological interventions aimed at reducing relapse rates. Strategies that minimize individuals' with SUDs exposure to stress or diminish their stress sensitivity emerge as promising measures to decrease relapse likelihood. Furthermore, enhancing emotional and social support from family, friends, and the community can improve coping mechanisms, foster self-control, and subsequently lower the tendency towards relapse.