Failure-Related State Orientation and Self-Esteem Change in Inpatient Addiction Treatment for Illicit Drug Addiction

Illegal drug addiction is one of the most serious and costly public health problems in Germany (Rauschert et al., 2022). The annual costs of the consumption of illegal drugs were estimated at 5.2–6.1 billion euros in 2010. Cannabis is the most frequently used illegal drug in Germany with a 12-month prevalence of 8.8%. A rise in the prevalence of use in recent years has been reported throughout Europe. With a prevalence of 1.6%, cocaine is the second most commonly used illegal drug followed by amphetamine (1.4%), new psychoactive substances (1.3%), and methamphetamine (0.2%) (Rauschert et al., 2022).

In terms of relapse prevention, improving psychosocial functioning is an important component of drug addiction treatment (Committee on Developing Evidence-Based Standards for Psychosocial Interventions for Mental Disorders, 2015; Dutra et al., 2008; Silverman et al., 2008). Drug addiction can be caused by low psychosocial functioning, which can be attributed to unfortunate dispositions, for instance, neuroticism (Clemente-Suárez et al., 2022; Juchem et al., 2024), impulsivity (Verdejo-García et al., 2007; Zhuo et al., 2021), or sensation seeking (Shen et al., 2023). Especially low self-esteem has been associated with the development of drug addiction (Fieldman et al., 1995; Zeng et al., 2024). Lowered self-esteem has been consistently found to occur in several psychiatric disorders (Hilbert et al., 2019; Orth & Robins, 2022; Silverstone & Salsali, 2003). According to one explanation, drugs may be used to counteract feelings of negative mood and low self-esteem (Alavi, 2011; Samuels & Samuels, 1974; Yan et al., 2020). On the other hand, low esteem is also regarded as a consequence of drug addiction (Ayres & Taylor, 2025).

Self-esteem is understood as an individual’s general self-assessment that is associated with a positive or negative attitude towards oneself (Baumeister et al., 2003; Buhrmester et al., 2011; Twenge & Campbell, 2002). In a more complex approach, self-esteem is conceptualized as a hierarchical, multi-layered structure that distinguishes emotional, social, physical, and academic components of self-esteem (Fleming & Courtney, 1984; Katrin Rentzsch et al., 2021).

Effective interventions that raise self-esteem in individuals with substance use disorders can increase their chances of success in important life domains (Zell & Johansson, 2025). It was shown that high self-esteem helps individuals to make friends, to have more satisfying romantic relationships, to be more successful at school and work, and to experience higher psychological well-being (Orth 2022). On the other hand, the experience of failure in acquiring sufficient self-esteem may be associated with emotional instability, self-doubt, and feelings of insufficiency (Henriksen et al., 2017; Sowislo & Orth, 2013). The negative effects of low self-esteem are addressed in the so-called vulnerability model, in which low self-esteem is conceptualized as a stable personality trait that predisposes people to mental health problems such as anxiety or depression (Orth et al., 2016; Steiger et al., 2015). Low self-esteem was found to be associated with social problems, resulting in destructive attitudes and high-risk behaviors such as drug use (Mann et al., 2004).

As already stated, drug addiction can also lead to low self-esteem. Accordingly, individuals with substance use disorders are shown to be dissatisfied with themselves, suffering under feelings of loneliness, depression, and interpersonal conflicts (Gossop, 1976; Silverstone & Salsali, 2003; Zeng et al., 2024). For instance, stigmatization by friends or colleagues due to negative prejudices about individuals with drug addiction may reduce self-esteem via difficulties in job employment, interpersonal rejection, devaluating thoughts about oneself, seclusion, or avoidance of intimate contacts (Ayres & Taylor, 2025; Khalid et al., 2020). It is supposed that self-esteem of stigmatized individuals may parallel the degree to which they are devaluated by the culturally dominant group because of internalizing the negative view held by the society at large (Cama et al., 2016; Major & O'Brien, 2005).

Low self-esteem following drug addiction may be maintained by state orientation. The distinction between action and state orientation refers to individual differences in self-regulation, particularly under stress (Baumann et al., 2005, 2007; Kuhl, 1981). According to Kuhl’s Personality Systems Interactions theory, state orientation can manifest in two forms: decision-related state orientation, characterized by difficulties generating positive affect in demanding situations, and failure-related state orientation (AOF), defined by impaired regulation of negative affect after failure or setbacks. State-oriented individuals are less able to initiate effective action because they tend to lose sight of self-congruent goals. As a result, they may commit to unrealistic or need-incongruent goals more frequently than action-oriented individuals. The capacity to maintain self-congruent goal pursuit under stress is considered protective for mental health and abstinence (Koole et al., 2019, 2023). Accordingly, state orientation may represent a vulnerability factor for low self-esteem, as it can hinder successful coping and goal attainment. Empirical research links state orientation to procrastination, alienation, depression, anxiety, and reduced well-being (Koole & Fockenberg, 2011; Koole et al., 2019). Among the two forms of state orientation, AOF is particularly relevant in the present context. The pursuit of abstinence is often accompanied by setbacks and perceived failures. Individuals who struggle to regulate negative affect following failure may be especially vulnerable to self-doubt and diminished self-evaluation during this process. Difficulties in recovering from setbacks may therefore undermine both goal persistence and self-esteem. We therefore hypothesized that higher AOF would be associated with lower self-esteem.

In general, inpatient addiction treatment aims to achieve and maintain an abstinent lifestyle (Deutsche Rentenversicherung Bund, 2025). Through interventions such as group therapy, psychoeducation, social skills training, and structured stress-testing phases, rehabilitants are encouraged to reflect on personally meaningful goals and to develop more adaptive coping strategies. These processes may foster a more stable and positive self-evaluation, particularly as individuals experience mastery, social integration, and improved self-regulatory functioning during treatment. Although AOF is a well-established vulnerability factor for low self-esteem, little is known about its role in self-esteem development during addiction treatment. Prior research has largely relied on cross-sectional designs or non-clinical samples, leaving open the question of whether dispositional differences in self-regulatory functioning merely reflect stable between-person differences in self-esteem or whether they also influence within-person changes during structured inpatient treatment. Given that treatment explicitly targets self-regulatory capacities and adaptive goal pursuit, self-esteem may be expected to improve over the course of treatment. Clinically, it would be particularly desirable to achieve comparable or even stronger increases in self-esteem among individuals with higher AOF, given their heightened vulnerability to setbacks during abstinence. We therefore hypothesized that self-esteem would increase during inpatient addiction treatment.

The study was planned as a longitudinal study (within-subjects design) and conducted from December 2019 (12/17/2019 first patient in) to September 2022 (09/27/2022 last patient out). It was conducted exclusively with rehabilitants from the Fachklinik Alte Flugschule Großrückerswalde (IGB – Institut für Gesundheit und Bildung gGmbH Alte Flugschule, 2025), Germany, which specializes exclusively in the inpatient treatment and rehabilitation of adolescent and adult patients with addiction to illegal drugs. In Germany, withdrawal treatment for addiction to illegal drugs lasts 24 weeks and can only be considered to have been completed after the full duration has elapsed. A prerequisite for admission is an approval by a funding body based on an assessment of the prospects of treatment success (Deutsche Rentenversicherung Bund, 2025). Before beginning the inpatient addiction treatment, all rehabilitants receiving an approval have to undergo detoxification treatment in a specialized psychiatric clinic.

All rehabilitants who met the a priori defined inclusion and exclusion criteria were approached by their primary therapist at the start of their treatment regarding participation in the study. If informed consent was given, the patients were included in the study. In addition to mandatory initial diagnostics, the research questionnaire was presented to all participants for completion (t0). After collecting the demographic data (e.g., age, gender, employment status, previous withdrawal treatments, and prison experience), ten further measures were used (e.g., International Classification of Functioning, Disability and Health (ICF) (Linden, 2017). For our investigation, only demographic data, the results of the Action Control Scale (ACS-90, German Version: HAKEMP-90) (Diefendorff et al., 2000; Kuhl, 1994), and the Multidimensional Self-esteem Scale (Multidimensionale Selbstwertskala, MSWS) (Schütz et al., 2025) were analyzed.

The level of action versus state orientation was measured using the ACS-90 (Diefendorff et al., 2000; Kuhl, 1994). State orientation, as assessed by the ACS-90, reflects a dispositional self-regulation style that is assumed to be relatively stable over time. The ACS-90 allows the determination of three dimensions of action versus state orientation, whose subscales are called: (1) Action orientation subsequent to failure vs. preoccupation (AOF), (2) Prospective and decision-related action orientation vs. hesitation (AOD), and (3) Action orientation during successful performance of activities as intrinsic orientation vs. volatility (AOP) (Kuhl, 1994). For our research question, the AOF dimension is particularly relevant because its underlying characteristics may have a decisive influence on self-esteem, as it relates to how well individuals can deal with failures and negative emotions and how quickly they can recover from their motivational “breakdowns.” The corresponding subscale comprises 12 items, each describing a fictitious failure situation. Two answer options are available for each item, with one answer expressing a state-oriented attitude and one answer expressing an action-oriented attitude. For example, the failure situation: “If I have lost something valuable and every search has been in vain, then …” can be answered either by selecting the option: … “I have difficulty concentrating on something else” (state orientation) or by selecting the option: … “I don’t think about it much longer” (action orientation). After adding up all answers, a total score of up to 12 can be achieved. Higher scores indicate a stronger tendency toward action orientation, whereas lower scores reflect higher state orientation.

The MSWS (Schütz et al., 2025) was developed to comprehensively assess a person’s self-esteem across multiple domains. It assumes that self-esteem is not a single, unified feeling but manifests itself in multiple dimensions. The measure originally comprised six subscales, which can be subsumed into four subscales according to the present conception. First, the emotional self-esteem subscale is designed to assess the extent to which a person sees themselves as emotionally valuable, accepted, and capable of maintaining positive emotional states. High emotional self-esteem can be associated with a stable sense of self and better coping with emotional challenges. Second, the performance-related self-esteem subscale measures the extent to which a person derives their self-worth from their achievements and successes. People with high performance-related self-esteem define their worth primarily through goal achievement, recognition, or success. Third, the social-related self-esteem subscale measures how valued and accepted a person feels in social situations. It captures self-image in terms of social skills, the feeling of being liked by others, and confidence in one’s own social abilities. Fourth and finally, the body-related self-esteem subscale captures the extent to which someone positively assesses their own body image and perceives themselves as valuable and satisfactory in terms of physical characteristics and external appearance. The MSWS is characterized by good validity and feasibility, making it suitable for use across different age groups and contexts.

Longitudinal data were analyzed using linear mixed-effects models with random intercepts for participants. Measurement time, AOF, and their interaction were specified as fixed effects, with age and gender included as covariates. Separate models were estimated for each self-esteem dimension. Statistical significance was evaluated based on fixed-effect tests using a two-sided alpha level of .05. All analyses were conducted using SPSS (version 28.0.1.0). An ethics board approval was obtained from the Sächsischen Landesärztekammer.

A total of N = 261 rehabilitants were initially included in the study. Of these, N = 134 (51.3%) participants completed the inpatient addiction treatment regularly and were included in the longitudinal analyses, whereas N = 127 participants discontinued treatment prematurely and were classified as treatment dropouts. Of these, 36 (13.6%) participants discontinued treatment prematurely on medical grounds, 6 (2.3%) terminated treatment with their physician’s consent, 44 (16.7%) did so against medical advice, and 32 (12.1%) did so for disciplinary reasons. Five participants (1.9%) were transferred to another clinic. In addition, four participants were excluded from analysis due to the non-processing of scales that were relevant for the research question. Consequently, data from 134 participants were available for longitudinal analysis. Baseline comparisons indicated only minor sociodemographic differences between treatment completers and dropouts (see Table 1). No meaningful differences were found with respect to AOF or baseline self-esteem across any MSWS subscale (see Table 2), suggesting substantial comparability between groups regarding the core psychological variables.

AOF was assessed at the beginning of treatment using the Action Control Scale (ACS-90). Baseline analyses examined whether AOF was associated with sociodemographic characteristics. Linear regression analyses indicated that AOF was not significantly associated with age (B = 0.18, SE = 0.17, p =.279). In addition, binary logistic regression analyses revealed no significant associations between AOF and partnership status (OR = 0.96, 95% CI [0.88, 1.05], p =.396), having at least one biological child (OR = 1.00, 95% CI [0.91, 1.10], p =.975), or prior imprisonment (OR = 0.99, 95% CI [0.90, 1.09], p =.833).

Overall, AOF appeared largely independent of the examined sociodemographic variables at baseline. However, men showed higher AOF scores than women (M = 5.85, SD = 2.76 vs. M = 4.72, SD = 2.99), t(132) = 2.11, p <.05, d = 0.40. Baseline AOF and self-esteem across all MSWS subscales (emotional, performance-related, social, and body-related self-esteem) are reported separately for treatment completers and treatment dropouts in Table 2.

In line with the research question, subsequent analyses examined whether AOF was associated with self-esteem. At the beginning of treatment (t0), higher levels of AOF were associated with lower emotional, social, and body-related self-esteem, but not with performance-related self-esteem. Linear mixed-effects models with random intercepts for participants were estimated separately for each self-esteem dimension. Measurement time (t0 vs. t1), AOF, and their interaction (measurement time × AOF) were specified as fixed effects. Age and gender were included as covariates in all models.

Across the sample analyzed, a significant increase during treatment was observed for emotional, social, and body-related self-esteem, whereas performance-related self-esteem did not show a significant change over time (see Table 3). Moreover, higher state orientation was associated with lower emotional self-esteem across measurement occasions, F(12, 230.225) = 4.697, p <.001. The time × AOF interaction was not significant, F(11, 226.572) = 1.457, p =.149, indicating that changes in emotional self-esteem over time did not differ as a function of AOF. Similarly, participants with higher state orientation reported significantly lower social self-esteem than participants with lower state orientation, F(12, 257.599) = 5.780, p <.001, with no significant time × AOF interaction, F(11, 256.767) = 1.542, p =.117. The same pattern was observed for body-related self-esteem, where higher state orientation was associated with lower self-esteem, F(12, 228.157) = 3.642, p <.001, and the interaction between time and AOF was not significant, F(11, 224.353) = 1.000, p =.447.

In contrast, performance-related self-esteem was not significantly associated with AOF, F(12, 346.997) = 1.292, p =.221. Likewise, the time × AOF interaction was not significant, F(11, 359.538) = 0.806, p =.634, indicating that performance-related self-esteem did not develop differently over time as a function of AOF.

Taken together, the present findings indicate that emotional, social, and body-related self-esteem increased over the course of inpatient addiction treatment, whereas performance-related self-esteem did not show significant change. Higher levels of AOF were consistently associated with lower emotional, social, and body-related self-esteem, whereas no such association was observed for performance-related self-esteem. Across all four self-esteem dimensions, AOF did not moderate changes in self-esteem over time.

The present study was conducted in a specialized German inpatient addiction setting providing a standardized 24-week inpatient addiction treatment following mandatory detoxification. Although all participants entered treatment under comparable structural conditions, only about half completed the program and could be included in the longitudinal analyses. Within this clinical context, the observed pattern of results suggests that self-esteem development during treatment differs across domains and is not uniformly shaped by dispositional differences in AOF. In particular, the finding that comparable levels were reached only for body-related self-esteem highlights potential domain-specific mechanisms of change, which are discussed in more detail below.

Hypothesis 1 was partially supported. AOF was associated with lower emotional, social, and body-related self-esteem, whereas no association was observed for performance-related self-esteem. Regarding Hypothesis 2, the present findings indicate that self-esteem increased over the course of treatment in the three domains emotional, social, and body-related self-esteem, suggesting that core aspects of self-evaluation may change during treatment. In contrast, performance-related self-esteem remained largely stable over time. Importantly, AOF did not moderate these changes, indicating that increases in self-esteem were comparable across levels of dispositional state orientation. These findings may also be interpreted within a self-regulatory framework. Inpatient addiction treatment explicitly targets adaptive coping, goal pursuit, and emotion regulation. Improvements in these processes may contribute to changes in self-evaluation, even if dispositional state orientation itself remains relatively stable. From this perspective, self-esteem development during treatment may reflect not only stable individual differences but also changes in regulatory functioning, an assumption that warrants further empirical investigation.

These findings are consistent with prior research indicating that self-esteem is a modifiable construct that can be strengthened through psychological intervention (Bhattacharya et al., 2023; Sowislo & Orth, 2013). Accordingly, the observed increases in emotional, social, and body-related self-esteem suggest that inpatient addiction treatment can facilitate meaningful changes in core self-evaluative processes. However, despite comparable treatment conditions, state-oriented individuals did not reach the same levels of self-esteem as action-oriented individuals, indicating persistent vulnerability-related differences. One contextual factor that may have contributed to improvements in certain self-esteem domains concerns the social comparison environment within inpatient treatment. Self-evaluations are strongly shaped by reference groups and comparison standards (Festinger, 1954; Mussweiler & Strack, 2000). The relatively homogeneous and protected setting of inpatient rehabilitation may reduce exposure to unfavorable upward comparisons and provide more comparable peer reference standards. Such conditions may facilitate more positive self-evaluations, particularly in domains that are sensitive to social comparison processes. However, comparable convergence between state- and action-oriented individuals was not observed across all self-esteem domains, suggesting that comparison-related improvements alone cannot fully account for the broader pattern of results.

There are several limitations to the present study. A primary limitation concerns selective attrition. Approximately half of the initially enrolled rehabilitants discontinued treatment prematurely and could therefore not be included in the longitudinal analyses. Selective dropout may, in principle, bias estimates of treatment-related change if individuals who discontinue treatment differ systematically from those who complete it. However, comparative analyses indicated that treatment completers and dropouts did not differ meaningfully at baseline with respect to AOF or any dimension of self-esteem, which were the central variables of interest in the present study. Differences between groups were limited to sociodemographic characteristics, with a slightly higher proportion of men and a higher prevalence of prior imprisonment among dropouts. These factors were statistically controlled in all longitudinal models. Accordingly, although the findings should be interpreted as applying to individuals who completed inpatient addiction treatment, the absence of baseline differences in the key psychological variables reduces concerns that selective attrition substantially biased the observed patterns of self-esteem change. A second limitation is the absence of follow-up assessments after treatment completion. Consequently, the present analyses are restricted to changes observed during treatment, and no conclusions can be drawn regarding the long-term stability of self-esteem following treatment termination. Given that relapses and psychological deterioration may occur after discharge, future research should include follow-up measurements to examine the sustainability of self-esteem changes over time. The third limitation concerns covariate selection. While age and gender were included as covariates in all linear mixed-effects models, other clinical background variables (e.g., prison history, previous treatment experience, substance type) were not controlled. Although such variables may be related to psychological functioning, they were not considered primary confounders of the association under study. Future research with larger samples should examine their potential moderating role. The fourth limitation concerns the measurement approach. All constructs were assessed via self-report instruments, which may be susceptible to response biases such as social desirability, particularly in inpatient treatment settings where therapeutic expectations may influence responding. Although no specific control measures were applied, future studies should consider combining self-report data with behavioral or clinician-rated indicators to further reduce potential bias. Although the MSWS and ACS are well-validated and widely used measures, their application in clinical populations with addiction disorders has been examined only to a limited extent. Accordingly, it cannot be fully ensured that these instruments capture the intended constructs with equal validity in this specific population. The fifth limitation concerns the assessment of gender. Gender was recorded using a binary classification (male/female), which does not capture the full diversity of gender identities. This may limit the generalizability of the findings to individuals who do not identify within a binary gender framework. Future research should employ more inclusive gender measures to better reflect gender diversity in clinical populations. Finally, the study was conducted in a single German clinic specializing in inpatient addiction treatment, which may limit the generalizability of the findings. Although inpatient addiction treatment in Germany is highly standardized regarding treatment duration, therapeutic components, and admission criteria, direct transferability to treatment systems in other countries cannot be assumed. At the same time, the psychological mechanisms examined in this study are grounded in established theoretical frameworks and are not specific to the German healthcare context. From this perspective, the findings may inform research in non-German treatment settings, including outpatient or shorter-term rehabilitation programs, although direct generalization should be made with caution.

Future research should address two closely related lines of inquiry emerging from the present findings. First, the persistent stability of performance-related self-esteem highlights the need to identify treatment strategies that more directly target this domain. Given the potential role of performance-related self-worth in mood regulation and relapse vulnerability, it may be beneficial to develop intervention components that explicitly address the self-esteem challenges associated with limited access to achievement-related roles. Psychoeducational approaches focusing on the structure and contingencies of self-esteem, as well as interventions aimed at fostering alternative sources of competence, mastery, and socially valued contribution, may represent promising avenues. Future studies should examine whether such approaches can facilitate improvements in performance-related self-esteem and contribute to more stable psychological functioning following treatment.

Second, the observed differences between state-oriented and action-oriented individuals across multiple self-esteem domains suggest that failure-related state orientation may represent a broader vulnerability factor extending beyond self-esteem alone. State orientation has been linked to impaired goal disengagement, reduced self-regulatory flexibility, and increased sensitivity to setbacks, which may affect multiple areas of functioning. Future research should therefore investigate whether targeted interventions can help state-oriented individuals develop more adaptive self-regulatory strategies and thereby achieve self-esteem levels comparable to those of action-oriented individuals. This includes examining whether therapeutic approaches that strengthen goal regulation, coping with failure, and adaptive self-reflection can promote more stable and generalized improvements in psychological functioning across domains.

Overall, the present findings highlight that self-esteem development during inpatient addiction treatment is domain-specific, shaped by both individual self-regulatory dispositions and structural opportunities for competence and social comparison.