The effects of posttraumatic stress symptoms on anxiety and depression in patients with breast cancer: mediating role of negative interpretation bias.

Introduction
Breast cancer is the most prevalent malignancy among women worldwide [1], and its diagnosis and treatment are frequently accompanied by significant psychological stress and severe emotional disturbances. Studies consistently indicate that anxiety and depression are the most common mental health challenges in breast cancer patients, with prevalence rates reaching 60.2% and 52.3%, respectively [2]. These emotional disorders are often intertwined with other psychological issues—such as fear of cancer recurrence and sleep disturbances—that not only impair quality of life but also hinder treatment adherence, ultimately compromising clinical prognosis [3, 4]. A meta-analysis involving over 280,000 breast cancer patients further revealed that depression significantly increases the risk of recurrence and mortality, whereas anxiety is associated with both recurrence and all-cause mortality [4]. These findings highlight the importance of identifying the psychological mechanisms that underlie emotional distress in this population to improve both psychological outcomes and overall well-being.
PTSS are frequently comorbid with anxiety and depression, and together they form a complex and reinforcing network of psychological disturbances [5–7]. Genetic studies suggest that PTSS, anxiety, and depression may share biological vulnerabilities, particularly involving stress-regulatory pathways such as the hypothalamic–pituitary–adrenal (HPA) axis [8]. These vulnerabilities may be amplified by environmental stressors, including traumatic experiences, through gene–environment interactions. Previous research has demonstrated that core PTSS features—such as avoidance, intrusive memories, and negative cognitive alterations—are strongly associated with excessive worry in anxiety and feelings of helplessness in depression [5]. Moreover, the chronicity and severity of PTSS have been closely linked to emotional dysregulation, further exacerbating psychological burden [6].
In the context of breast cancer, although a cancer diagnosis does not typically meet DSM-5 Criterion A for posttraumatic stress disorder (PTSD), it can operate as a persistent, life-threatening stressor that elicits trauma-like responses [9]. Accordingly, the diagnostic and treatment process itself constitutes a profound life stressor that exposes patients not only to physical trauma but also to severe psychological threats. PTSS has been increasingly recognized as a common outcome in this population, with prevalence rates ranging from 6.9% to 58%, depending on assessment criteria and timing [10, 11]. Consistent with this view, studies in breast cancer samples have documented substantial subthreshold PTSS, with estimates around 25% [12] and 25.8% in a Chinese cohort [13], and up to 75% in some reports [14]. PTSS typically presents with persistent psychological and somatic symptoms, including flashbacks, emotional numbing, trauma-related negative cognitions, and hyperarousal. These symptoms elevate overall emotional distress and may serve as precursors to full-blown anxiety and depressive disorders. For example, intrusive thoughts may fuel recurrent fears of cancer relapse, while emotional numbing can impair the ability to derive pleasure from social or interpersonal interactions, potentially leading to withdrawal and depressive states [15]. Despite accumulating evidence on the impact of PTSS in breast cancer, the specific cognitive pathways through which it contributes to anxiety and depression remain insufficiently understood.
Schnurr et al. [16] posited that PTSS disrupts mental health by impairing core cognitive processing functions. Consistent with this, several studies on breast cancer patients have shown that mild cognitive impairments—often observed before treatment initiation—are more strongly associated with psychological trauma than with the cancer itself or its medical treatment [17, 18]. For instance, breast cancer patients tend to exhibit attentional avoidance of emotionally salient stimuli and difficulty disengaging from positive cues, a pattern linked to heightened anxiety and depressive symptoms [19]. Additionally, survivors with high PTSS levels are more likely to rely on maladaptive emotion regulation strategies (e.g., emotional suppression), whereas those with lower PTSS tend to engage in more adaptive strategies such as cognitive reappraisal. These regulatory patterns are significantly associated with emotional health outcomes [20].
Brewin et al.'s dual representation theory of PTSD further elucidates how trauma-related memories may disrupt emotional functioning. According to this model, traumatic events are encoded in two distinct memory systems: contextualized representations (C-Reps), which are verbal, narrative-based, and integrated into autobiographical memory, and sensory-bound representations (S-Reps), which are vivid, emotionally intense, but lack temporal or spatial context [21]. During successful trauma processing, these systems are integrated, allowing the individual to cognitively frame traumatic experiences within a coherent autobiographical narrative. However, when integration is incomplete, S-Reps may intrude into awareness without contextual modulation, leading to flashbacks, threat overgeneralization, and persistent emotional dysregulation [22, 23].
Within this framework, anxiety and depression are conceptualized as secondary emotional responses—not direct reactions to trauma, but downstream consequences of unresolved trauma processing. Specifically, repeated reactivation of sensory-bound memories in the absence of contextual interpretation contributes to sustained negative affect and maladaptive emotional regulation [24]. This model provides insight into why patients with persistent PTSS often experience heightened anxiety and depression: the cognitive system fails to reconcile traumatic content within autobiographical memory, leading to chronic dysregulation.
Similarly, Ehlers et al.'s cognitive model of PTSD emphasizes that individuals with PTSS have difficulty evaluating trauma-related information adaptively. As a result, they become hypervigilant to perceived threats and tend to interpret ambiguous information in a negative or threatening manner—a cognitive distortion known as interpretation bias [25]. Among breast cancer patients, this bias may manifest as excessive concern about recurrence, magnified perceptions of treatment-related danger, or misinterpretations of supportive social cues [26, 27]. Indeed, studies have shown that those with higher fear of recurrence are more likely to interpret ambiguous information as threatening and to report more severe somatic symptoms such as pain [26, 28]. Notably, interpretation bias remains a robust predictor of recurrence-related fear, even when accounting for maladaptive thought patterns and metacognitive beliefs [26]. In line with this, Lam et al. [29] demonstrated that cancer patients with chronically elevated anxiety trajectories also exhibit heightened levels of interpretation bias. This cognitive bias not only reinforces trauma-related negative beliefs but also amplifies threat perception, thereby exacerbating anxiety and depressive symptoms.
Building on this account, numerous studies have examined associations among PTSS, anxiety, and depression in breast cancer; however, the underlying cognitive mechanisms have been less frequently tested. Previous work has linked interpretation bias to cancer-related distress—often via correlational or group-comparison designs [29] —and even tested mediation for other outcomes such as fear of cancer recurrence [28]; nevertheless, formal mediation models that evaluate interpretation bias as a mechanism linking PTSS to anxiety and depression in oncology populations remain scarce [30]. To address this gap, the present study investigates whether negative interpretation bias mediates the relationship between PTSS and emotional distress in breast cancer patients. We hypothesize that PTSS contributes to anxiety and depression both directly and indirectly, through increased interpretation bias. By constructing a mediation model including PTSS, interpretation bias, anxiety, and depression, this study aims to clarify one cognitive mechanism linking trauma symptoms to emotional comorbidity. These findings may inform early screening approaches and support the development of targeted cognitive interventions to improve psychological functioning and overall quality of life in breast cancer patients.

Methods

Participants

Sample size calculation
The required sample size was determined a priori using G*Power 3.1, following Cohen's [31] guidelines for repeated-measures ANOVA with a between-subjects factor. Parameters included: medium effect size (f = 0.25), α = 0.05, power (1 − β) = 0.80, two groups, nine measurements, and an assumed correlation of 0.5 among repeated measures. The minimum required sample size was 72. Given the cross-sectional nature of this study, a 10% increase was applied to account for potential data loss due to non-compliance, technical failure, or incomplete trials, resulting in a target of 80 participants. This approach is consistent with cognitive-affective paradigms reporting 5–15% exclusion rates [32, 33].

Participants and recruitment
Participants were recruited from the Breast Surgery Department of Weifang Municipal Hospital of Traditional Chinese Medicine between July 2020 and June 2023 through a two-stage screening process. In stage one, 234 breast cancer patients completed the IES during routine psychological assessment and were categorized into preliminary PTSS and non-PTSS groups. In stage two, 80 eligible participants (40 per group) were selected using consecutive sampling until the sample size was reached. Exclusions included incomplete data, ineligibility, or refusal to participate.Inclusion Criteria:Female, aged 30–79 years: This range reflects the peak breast cancer incidence in Chinese women [34, 35], while minimizing variability related to pediatric or geriatric cognitive profiles.

Primary diagnosis of breast cancer: To isolate the psychological impact of initial diagnosis.

Adequate Chinese literacy: Required for valid completion of self-report instruments and task instructions.

Awareness of diagnosis and voluntary participation: Ensured informed consent and psychological readiness.

Diagnosis within 12 months: Focused on the early adjustment phase, where trauma-related cognition and emotion are most salient.

Exclusion Criteria:Presence of recurrence or metastasis: May introduce distinct emotional and coping patterns.

Diagnosis of another malignancy: To avoid psychological confounds unrelated to breast cancer.

History of psychiatric disorders: To minimize pre-existing influences on interpretation bias and emotional distress.

Measures

Posttraumatic stress symptoms
The 15-item IES [36, 37] was used to assess PTSS via the intrusion and avoidance subscales. Items were rated on a 4-point scale, and scores above 19 suggest clinically relevant distress [36]. This threshold has been validated in cross-cultural breast cancer populations [38, 39]. Internal consistency in this sample was acceptable (total Cronbach's α = 0.81; intrusion = 0.80; avoidance = 0.66).

Anxiety and depression
The HADS [40] evaluated symptoms over the past month using 14 items (seven per subscale), scored from 0 to 3. Higher scores indicate greater severity. Internal consistency was satisfactory (total Cronbach's α = 0.81; anxiety = 0.70; depression = 0.70).

Stimuli and task design
Facial stimuli were selected from the Chinese Facial Affective Picture System(CFAPS) [41], comprising 40 images (10 males, 10 females, each showing happy and sad expressions). Images were matched for arousal and differed in valence (see Table 1). Each image was 260 × 300 pixels.

Morphing software (Fanta Morph 5.9.0.14, Abrosoft) generated 20 face sequences by blending each individual's happy and sad expressions across 11 steps with 10% increments. Two endpoint faces (0%, 100%) were excluded; the remaining nine faces (Steps1–9) were used (see Fig. 1). Steps 4–6 (high ambiguity) were presented three times, others once, totaling 300 randomized trials.

Procedure
The Ambiguous Emotional Face Recognition Task was administered in a quiet, temperature-controlled, sound-attenuated room using E-Prime 3.0. Each trial began with a 500 ms fixation cross, followed by a face image displayed for up to 2500 ms. Participants classified the expression as "happy"or "sad" via keypress. A 20-trial practice session included response feedback; no feedback was given in the main task (see Fig. 2). The formal task comprised 300 trials, divided into four blocks with short breaks. Participants received ¥50 for completing the task.

Interpretation bias index
Interpretation bias was assessed across emotional ambiguity levels 1–9. Level 5, representing the midpoint and highest ambiguity, was used as the primary indicator due to its sensitivity to group differences. The bias index was defined as the proportion of "sad "responses at this level. Since the task assessed interpretive tendency rather than accuracy, responses were not scored as correct or incorrect. Higher "sad" rates indicated stronger negative interpretation bias.

Statistical analysis
Data analysis was conducted using SPSS 21.0. Independent-samples t-tests compared group differences in PTSS, anxiety, and depression. ANOVA tested group differences in interpretation bias. Mediation analyses (PROCESS Model 4, Hayes, 2018) examined indirect effects of PTSS on emotional distress via interpretation bias, using 5,000 bootstrap samples and 95% confidence intervals.

Results

Sample characteristics
There were no significant differences in demographic or clinical variables between the PTSS and non-PTSS groups, ensuring comparability across groups (Table 2).

Posttraumatic stress symptoms, anxiety, and depression
Patients with breast cancer in the PTSS group reported significantly higher levels of anxiety and depression than those in the non-PTSS group (Table 3).

Negative interpretation bias
A 2 (Group: PTSS vs. non-PTSS) × 9 (Morph Level: 1–9) mixed-design ANOVA was conducted to examine the differences in "sad" response proportions across morph levels. The analysis revealed significant main effects of morph level [F (8, 71) = 877.22, p < 0.001, partial η² = 0.99] and group [F (1, 78) = 80.95, p < 0.001, partial η² = 0.51], as well as a significant interaction between group and morph level [F (8, 71) = 11.89, p < 0.001, partial η² = 0.57]. As shown in Fig. 3, participants with PTSS exhibited a consistently higher proportion of "sad"responses across all morph levels compared to the non-PTSS group.

Interpretation bias at morph level 5
Given that morph level 5 represented the highest ambiguity and showed the greatest group difference, it was selected as the primary indicator of negative interpretation bias. A one-way ANOVA revealed a significant group effect: PTSS group (M = 61.60, SD = 13.05) vs. non-PTSS group (M = 34.71, SD = 10.16), F (1, 78) = 105.73, p < 0.001, η² = 0.58 (Fig. 4).

Mediation analyses
Given that morph level 5 represented the highest level of emotional ambiguity and exhibited the most significant group differences in the proportion of "sad" responses, this morph level was selected as the primary indicator of negative interpretation bias in the mediation analysis.
To test whether interpretation bias mediated the relationship between PTSS and emotional distress, two separate mediation models were constructed using bootstrapping procedures (5,000 samples, 95% confidence intervals).

Model 1: anxiety as dependent variable
PTSS group was coded as the independent variable (1 = PTSS, 2 = non-PTSS), interpretation bias (i.e., proportion of "sad" responses at morph level 5) as the mediator, and anxiety as the outcome. Indirect effect: 0.85 (SE = 0.24), 95% CI [0.403, 1.338].
Direct effect: 0.64 (SE = 0.19), 95% CI [0.256, 1.024]. Interpretation bias partially mediated the association between PTSS and anxiety, accounting for 57.05% of the total effect (Fig. 5).

Model 2: depression as dependent variable
Using the same mediation structure, the model revealed: Indirect effect: 0.51 (SE = 0.26), 95% CI [0.017, 1.028]. Direct effect: 0.78 (SE = 0.25), 95% CI [0.280, 1.284].
The mediating effect accounted for 39.53% of the total association between PTSS and depression (Fig. 6).

Discussion
This study examined the interrelationships among PTSS, negative interpretation bias, and emotional distress—specifically anxiety and depression—in patients with breast cancer. Consistent with prior findings [13, 42], participants with PTSS reported significantly higher levels of emotional distress than those without PTSS. Importantly, they also exhibited stronger tendencies to interpret emotionally ambiguous facial expressions in a negative manner, particularly under conditions of maximal ambiguity (morph level 5). Taken together, these findings indicate that the ambiguous-face paradigm indexes negative cognitive appraisal under uncertainty rather than perceptual accuracy. Consistent with prior work, this paradigm reflects negative trauma-related cognitions in both trauma-exposed populations [43] and medical populations with high emotional distress [19], aligning with theoretical accounts that link interpretation bias to maladaptive threat appraisal [30]. This pattern underscores the presence of trauma-related cognitive vulnerability, which may potentiate adverse emotional outcomes in oncology populations.
Our findings are aligned with cognitive theories of trauma, especially the dual representation theory [21], which posits that traumatic experiences disrupt the integration of sensory-perceptual and contextual representations. This disruption impairs emotional regulation and leads to intrusive symptoms. Core features of PTSS—such as intrusive re-experiencing and avoidance—may intensify negative appraisals in situations involving emotional ambiguity. In breast cancer contexts, these appraisals may revolve around uncertainty regarding diagnosis, prognosis, or treatment, thereby exacerbating perceived threat and contributing to anxiety and depressive symptoms [42, 44]. Additionally, avoidant coping may further diminish engagement with social and daily activities, a hallmark of depression [45].
The morph-level analysis provided nuanced insights into interpretive processes under ambiguity. Patients with PTSS were more likely to respond with sadness across all morph levels, with the most pronounced differences at level 5, suggesting that heightened ambiguity amplifies cognitive distortions. These results are consistent with literature indicating that trauma survivors—including those in non-combat contexts—exhibit hypervigilance to threat under uncertainty [25, 43, 46]. In breast cancer patients, chronic emotional arousal may generalize beyond illness-related cues, enhancing global threat sensitivity and increasing the likelihood of negatively skewed interpretations. Moreover, previous studies have shown that cancer-related trauma can impair executive functioning [47], thereby further reinforcing negatively biased responses.
Interpretation bias was found to partially mediate the association between PTSS and emotional distress, indicating that distorted cognitive processing may represent a mechanism linking trauma symptoms to negative emotional outcomes. This is congruent with the vulnerability-stress model [48], which posits that pre-existing cognitive vulnerabilities, when activated by environmental stressors such as cancer diagnosis, precipitate psychopathological symptoms. Patients with elevated PTSS may interpret ambiguous information—whether facial cues or health-related messages—as inherently threatening, thereby entering a maladaptive cognitive-affective loop that reinforces distress [49]. These results point to interpretation bias as a modifiable risk factor in the trajectory of trauma-related psychological impairment.
It is also necessary to acknowledge the potential bidirectionality between cognitive distortions and emotional distress. Although our cross-sectional design positioned PTSS as the antecedent of interpretive bias, it is plausible that heightened anxiety or depression may contribute to the formation or persistence of cognitive distortions. For example, Everaert et al. [50] found that depressive symptoms were associated with both increased negative and decreased positive interpretations of ambiguous stimuli. Similarly, Hirsch et al. [30] demonstrated that individuals with social anxiety tend to display a consistent bias toward threat-related appraisals. Furthermore, deficits in cognitive flexibility—specifically, difficulties in revising initial negative appraisals—have been shown to perpetuate both anxiety and depression, suggesting that interpretation style and rigidity may jointly influence emotional outcomes. Future longitudinal and experimental studies are warranted to delineate causal directions and establish temporal precedence among these variables.

Clinical implications
The findings have meaningful clinical implications for psychooncology practice. Interpretation bias may serve as a non-invasive cognitive marker for identifying breast cancer patients at heightened risk for emotional disorders. While not diagnostic per se, tasks such as facial morph interpretation or ambiguous scenario judgment could be incorporated into survivorship care to complement traditional psychological assessments, potentially facilitating early detection of at-risk individuals—even prior to the emergence of full-blown psychiatric symptoms [30].
In addition to assessment, our findings suggest important implications for risk communication. Patients with negative interpretation bias may respond poorly to conventional, fear-based medical messaging. Negatively framed health information may be misinterpreted as threatening, further exacerbating distress. Instead, positively framed messages—emphasizing recovery, safety, and coping efficacy—may act as a protective buffer. Meta-analytic evidence indicates that gain-framed communication is particularly effective among individuals with high threat sensitivity [51, 52]. Integrating gain-framed communication into clinical interactions may help buffer emotional distress and enhance treatment engagement. This highlights the importance of trauma-informed communication strategies in oncology.
Moreover, psychological interventions targeting cognitive distortions—such as Cognitive Bias Modification (CBM) and Cognitive Behavioral Therapy (CBT)—may hold promise for improving emotional well-being in this population. Although CBM has shown effectiveness in reducing interpretation bias in anxiety and depression, its application in oncology remains underexplored. Future research should evaluate whether bias modification yields sustained improvements in psychological adjustment and cancer-related quality of life.

Limitations and future directions
Several limitations should be considered when interpreting these findings.
First, the cross-sectional design precludes causal inference. Although the mediation model revealed an indirect effect of PTSS on emotional distress via interpretation bias, temporal relationships remain uncertain. Longitudinal and experimental designs are needed to establish directionality and clarify potential bidirectional effects between cognitive distortions and emotional symptoms.
Second, the study 's generalizability is limited by its sample and cultural context. Participants were exclusively female breast cancer patients from a single hospital in China. Moreover, the use of the CFAPS, while culturally appropriate, may not generalize to populations with different emotional recognition norms. Interpretation bias likely reflects sociocultural influences, particularly in non-facial or interpersonal contexts [53]. Cross-cultural validation and multisite sampling will be essential to verify the robustness of these mechanisms.
Third, while the morphing task ensured experimental precision, its ecological validity was limited. Interpretation bias in real-world settings often arises in dynamic, multimodal interactions. Future research should employ more naturalistic assessment tools—such as interactive videos, simulated conversations, or virtual reality environments—to better approximate real-life cognitive processing.
Finally, the study did not control for potential confounds such as rumination, executive dysfunction, social support, or psychotropic medication use. These factors may influence both PTSS severity and interpretation patterns and should be integrated into future analytic models to improve explanatory precision.
Addressing these limitations will require longitudinal methods, culturally diverse samples, and ecologically valid paradigms to refine mechanistic understanding and inform the development of targeted, trauma-informed interventions in psycho-oncology.

Conclusions
This study provides novel empirical support for the role of negative interpretation bias as a partial mediator in the relationship between PTSS and emotional distress among breast cancer patients. These findings reinforce cognitive models of trauma-related psychopathology and identify interpretation bias as a clinically meaningful and potentially modifiable mechanism. Incorporating cognitive assessments into clinical screening, tailoring risk communication to patients’ cognitive profiles, and exploring the efficacy of bias-targeted interventions may enhance psychological outcomes in oncology settings. Longitudinal, cross-cultural, and mechanistically focused research is essential to refine these pathways and inform the development of trauma-informed cancer care.