Internalized shame drives suicidal thoughts in addiction, but social support can blunt the effect

People being treated for substance use disorder carry a burden that goes far beyond the pharmacology of addiction. Many have absorbed society’s harshest judgments about their condition, internalizing stereotypes of being dangerous, weak, or morally failed until those beliefs point inward as shame, guilt, and worthlessness. A new cross-sectional study published in BMC Psychiatry suggests that this internalized stigma, known as self-stigma, is significantly associated with suicide-related cognitions among adults in addiction treatment, and that the pathway runs primarily through depressive symptoms rather than anxiety or stress. Just as strikingly, the study finds that perceived social support from family, friends, and significant others can weaken the grip of self-stigma on suicidal thinking, pointing to a modifiable protective factor in a population already at dramatically elevated risk.

The research team, led by Zehra Uçar Hasanlı of the Alcohol and Substance Abuse Research, Treatment, and Education Center at Ankara Training and Research Hospital, recruited 270 adults receiving treatment for substance use disorder at a specialized addiction center and conducted analyses on 265 participants with complete data. All participants had a clinical diagnosis of substance use disorder established by a psychiatrist according to DSM-5 criteria, and the sample deliberately excluded alcohol use disorder because of its distinct stigma profile and epidemiological characteristics, restricting enrollment to individuals with illicit substance use disorders to reduce heterogeneity. Individuals with psychotic disorders, other severe psychiatric conditions, or significant cognitive impairment were excluded. The mean age of the sample was 33 years, with a range of 18 to 65, and 7.8 percent of participants were female, a distribution the authors note reflects the clinical population given lower treatment-seeking rates among women with substance use disorder.

The clinical stakes of this question are considerable. Prior research has reported lifetime suicide attempt rates of up to 40 percent among individuals receiving treatment for substance use disorder, and a recent meta-analysis indicated that the presence of substance use disorder increases suicide risk more than fivefold. Suicide mortality rates are significantly higher in individuals with substance use disorder than in the general population, and psychological autopsy studies identify mood disorders and substance use disorder as the most common psychiatric diagnoses among individuals who die by suicide. According to the World Health Organization, suicide is the third leading cause of death among individuals aged 15 to 29 years, with more than 720,000 deaths occurring globally each year. What has remained poorly understood, however, is the psychosocial machinery connecting the social experience of addiction to suicidal vulnerability, and this study is among the first to directly examine the relationship between self-stigma and suicide cognitions in this population.

The concept of suicide cognitions is central to the study’s theoretical framing. Drawing on Aaron Beck’s cognitive model of hopelessness and Mark Rudd’s Fluid Vulnerability Theory, the authors distinguish suicide cognitions from suicidal ideation. Whereas suicidal ideation reflects intent and planning, suicide cognitions represent a broader and potentially enduring cognitive vulnerability to suicide, characterized by core beliefs such as perceived unlovability, unbearability, low distress tolerance, and perceived burdensomeness. The authors reasoned that the shame, worthlessness, and self-blame inherent in self-stigma align closely with this suicide belief system, suggesting that internalized stigma may foster the development and maintenance of suicide-related dysfunctional beliefs. This makes suicide cognitions an earlier and theoretically grounded indicator of vulnerability than suicidal ideation, and a valuable target for early identification.

Participants completed four validated self-report instruments: the 40-item Substance Abuse Self-Stigma Scale, which measures the extent to which individuals internalize negative societal attitudes related to substance use; the 16-item Suicide Cognitions Scale-Revised, which assesses beliefs, attitudes, and expectations associated with vulnerability to suicidal behavior; the Depression Anxiety Stress Scale-21, a three-subscale measure of psychological distress with seven items each for depression, anxiety, and stress; and the Multidimensional Scale of Perceived Social Support, a 12-item instrument assessing subjective support from family, friends, and a significant other. Internal consistency of the measures in the sample was confirmed, with Cronbach’s alpha coefficients ranging from 0.78 to 0.93. Data screening included outlier diagnostics using standardized z-scores and Mahalanobis distance, normality checks against conventional skewness and kurtosis thresholds, and multicollinearity assessment using variance inflation factors and tolerance values, all of which indicated acceptable data quality.

The correlational findings painted a coherent picture. Self-stigma scores showed a strong positive association with depression (r = .67, p < .001) and moderate to strong positive associations with anxiety (r = .58), stress (r = .62), and suicide cognitions (r = .48). Suicide cognitions were also significantly positively correlated with depression (r = .55), anxiety (r = .48), and stress (r = .49), and negatively correlated with all three types of social support: family (r = -.39), friend (r = -.37), and significant other (r = -.39). Self-stigma itself was negatively associated with perceived family support (r = -.20), friend support (r = -.26), and support from a significant other (r = -.23). Depression, anxiety, and stress were highly intercorrelated, with r values ranging from 0.58 to 0.86, a pattern that would later prove consequential for interpreting the mediation results.

To test whether psychological distress transmits the effect of self-stigma onto suicide cognitions, the researchers conducted a parallel multiple mediation analysis using the PROCESS macro, entering depression, anxiety, and stress simultaneously as mediators and estimating indirect effects with 5000 bootstrap resamples and 95 percent bias-corrected confidence intervals. Self-stigma significantly predicted all three potential mediators, with unstandardized coefficients of 0.17 for depression, 0.13 for anxiety, and 0.14 for stress. Yet when all mediators competed for variance in predicting suicide cognitions, only depression emerged as a significant predictor (b = 0.78, p = .012), while neither anxiety (p = .58) nor stress (p = .54) was significantly associated with suicide cognitions in the presence of the other variables. The indirect effect of self-stigma on suicide cognitions through depression was significant (b = 0.13, SE = 0.06, 95 percent CI [0.01, 0.25]), whereas the indirect effects through anxiety and stress had confidence intervals that included zero. The direct effect of self-stigma on suicide cognitions remained significant after controlling for the mediators, indicating partial mediation.

The authors interpret this selective mediation through the specific cognitive content of depression. The depressive syndrome encompasses not only depressed mood, anhedonia, and vegetative symptoms but also feelings of worthlessness, guilt, and negative self-perception, cognitions that are directly related to the patterns underlying suicide cognitions. Feelings of worthlessness and guilt induced by self-stigma may reinforce these depressive schemas, thereby strengthening suicide-related beliefs. Anxiety and stress, by contrast, may reflect a shared negative affectivity with depression, while the cognitive content most characteristic of suicide cognitions belongs to depression. The authors also caution that the high intercorrelations among the DASS-21 subscales suggest potential multicollinearity, which may explain why anxiety and stress did not emerge as significant mediators when entered simultaneously, even though formal diagnostics indicated that multicollinearity did not reach problematic levels. The differential mediating roles of these variables should therefore be interpreted with caution.

The moderation analyses delivered the study’s most hopeful technical result. Separate models tested whether perceived social support from family, friends, or a significant other moderated the association between self-stigma and suicide cognitions, with all continuous variables mean-centered and interaction effects probed using simple slope analyses at low, mean, and high levels of each moderator. Every model was significant. The family support model explained 35 percent of the variance in suicide cognitions, the friend support model 32 percent, and the significant other model 35 percent. In each case, the interaction between self-stigma and support was significant, and simple slopes showed the association between self-stigma and suicide cognitions attenuating at higher levels of support. For family support, the simple slopes fell from b = 0.36 at low support to b = 0.16 at high support; for friend support, from b = 0.35 to b = 0.15; and for significant other support, from b = 0.37 to b = 0.13, all statistically significant. The pattern indicates a buffering effect: even when individuals have internalized prejudicial stereotypes from broader society, perceiving oneself as valued and understood by one’s social environment may counterbalance self-stigma-related cognitions such as worthlessness and unlovability, weakening their progression toward suicidal orientation.

The study’s limitations frame how far these findings can travel. The cross-sectional design does not allow causal inferences, the reliance on self-report measures for all variables introduces the risk of common method bias, and sociodemographic variables such as age, gender, and education were not controlled for in the analyses. The predominantly male sample, while consistent with epidemiological evidence, may limit generalizability, and the absence of analyses according to the type of substance used represents another constraint. Still, the clinical implications are concrete. The authors recommend routine assessment of self-stigma and depressive symptoms in substance use disorder populations due to their predisposition to suicidal cognitions, and argue that interventions aimed at reducing self-stigma, enhancing self-worth, addressing depressive symptoms, and increasing perceived social support should be considered integral components of suicide prevention strategies in this high-risk group. From a policy perspective, the findings highlight the importance of integrating stigma-reduction strategies into clinical training and public health initiatives, and of strengthening social support systems and access to psychosocial services. In a population where suicide vulnerability appears to be shaped by modifiable psychosocial processes, the message is that how society’s judgments are internalized, how depression takes hold, and how supported a person feels may together determine whether stigma hardens into suicidal belief or is softened by connection.

Subject of Research: Self-stigma and suicide cognitions in substance use disorder: the mediating role of psychological distress and the moderating role of social support

Article Title: Self-stigma and suicide cognitions in substance use disorder: the mediating role of psychological distress and the moderating role of social support

Article References: Uçar Hasanlı, Z., Varlı, M. R., Kurtoğlu, M. B., Sezer Katar, K., Zengin İspir, G., & Danışman, M. (2026). Self-stigma and suicide cognitions in substance use disorder: the mediating role of psychological distress and the moderating role of social support. BMC Psychiatry, 26(1), Article 716. https://doi.org/10.1186/s12888-026-08349-7

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08349-7

Keywords: substance use disorder, self-stigma, suicide cognitions, depression, social support, psychiatry, suicide prevention, psychological distress, suicide, cognitions, substance, disorder