Health anxiety (HA), alike known as illness anxiety, is a common issue, impacting up to 6% of general individuals at some point in their lives and up to twenty percent of medical outpatients (1). Excessive worry about one’s health and fear of getting sick are key features of HA, leading to constant monitoring of symptoms and unproductive facts seeking and safety measures (1). HA, while mild forms can encourage beneficial behaviors such as attending preventive medical check-ups, becomes a disabling condition when excessive. Severe health anxiety leads to significant distress, challenging patient-doctor relationships (2), increased healthcare utilization (3), and a considerable economic burden (4). It affects approximately 3.4% of the adult population (5) and is strongly associated with psychological distress and functional impairment (3, 5). As a serious and costly public health issue, untreated HA often becomes chronic (6), resulting in enduring consequences for individuals, healthcare systems, and society (7). The effect of HA on one’s health and the use of medical services can substantially elevate healthcare costs. In a natural longitudinal examination, it was revealed that individuals suffering from severe HA incurred 41–78% more healthcare expenses annually (2).

Numerous examinations have concentrated on the cognitive and emotional factors connected with a greater probability of experiencing excessive HA (8–10). In several studies (11–14), it has been discussed, investigated, and demonstrated that anxiety sensitivity (AS), emotion regulation (ER) and distress tolerance (DT) which are known as three transdiagnostic factors are linked to HA. These factors provide valuable insight into the underlying mechanisms of HA, offering potential targets for therapeutic interventions.

Anxiety sensitivity (AS) as transdiagnostic risk factor for psychiatric disorders refers to the negative emotional reactions to anxiety-related senses or symptoms. It involves the fear of bodily senses related to physiological arousal and the belief that these senses may result in negative physical, social, or psychological impacts (15, 16). AS is a result of dysfunctional beliefs and can lead to intense anxiety-related fear of physical sensations. Several studies highlight the connection between AS and illness anxiety disorder. As part of the cycle of health anxiety, AS can intensify an individual’s tendency to be overly vigilant about bodily senses and to experience distress when they occur (17, 18). AS has been observed to be linked with HA in both non-clinical and clinical adult samples. Research has found that greater levels of AS relates to greater HA (11, 19). In contrast, the absence of a significant prospective relationship between AS and HA is incompatible with some earlier work (12).

Distress tolerance (DT) represents the capacity to endure unfavorable psychological and physical states (20). It influences how distressing stimuli are appraised and the perceived capability to manage them (21). DT measures a person’s capability to endure uncomfortable emotional states and assesses their ability to handle distress (22). The behavioral index of DT measures the mastery to handle a challenging situation or task while striving toward a goal in front of negative feelings (20).

Studies indicate that clinically meaningful anxiety levels are related to more down levels of DT (23). According to a study (24), these results imply that DT may be an essential component in the development of anxiety disorders in overall. Persistent anxiety symptoms may result from actions linked to low DT (25). Studies suggest that DT and HA are negatively associated in a wide variety of nonclinical individuals (25, 26). Individuals with low DT are driven to flee from interior experiences, which exacerbates anxiety, since they believe they are unable to tolerate bad internal states, such as by believing that illness-related beliefs are dangerous and uncontrollable (25, 26). The concept of DT has obtained little consideration in the literature on HA despite several studies highlighting its significant role in general anxiety (26–29).

Emotional regulation (ER) involves trying to alter emotions by either evoking or sustaining emotional experiences, or by adjusting and effectively manage their frequency, strength, or duration (30). Our theoretical framework is anchored in Gross’s (2015) process model of ER, which stands as the most commonly utilized model of ER within the field of affective science. This model introduces a framework for emotion regulation that categorizes strategies into five distinct groups: the selection of specific scenarios, the adjustment of those scenarios, the way attention is focused, the reappraisal of cognitive perceptions, and the modification of responses (31).

Within the Gross’s model, cognitive reappraisal (CR) and expressive suppression (ES) serve as primary illustrations of strategies for regulating emotions. CR, which focuses on changing cognitive antecedents, requires a reinterpretation or intentional alteration in the perception of an emotional event or stimulus, aiming for a more balanced or positive perspective. Conversely, ES targets the modulation of emotional responses by restraining the outward expression of emotions, essentially inhibiting any reaction during the last phase of the ER process (31). In the field of ER, CR is often considered an adaptive strategy and ES considered a maladaptive strategy. Utilizing CR regularly is correlated with numerous beneficial long-term effects, such as enhanced overall well-being and social functional (32). On the other hand, excessive reliance on ES generally correlates with suboptimal outcomes, including diminished social functional and well-being (33). Certainly, mental health conditions like depression and anxiety disorders are often typified by a decreased reliance on CR and an increased reliance on ES (34, 35). The number of research exploring the contribution of and/or association between HA and ER In adults. ER difficulties have been positively correlated with HA (36, 37). Studies show that unhealthy emotional regulation techniques like suppression and catastrophizing are connected to HA (38, 39). Also (40), discovered that ER served as a mitigating factor for HA, but its impact was not direct. Exploring the relationship between ER and HA is valuable because the findings could enhance our understanding of the origins of HA, potentially leading to more effective, evidence-based interventions and assessments.

An important transdiagnostic factor is anxiety sensitivity (AS), which involves the fear of physiological arousal senses and the view that they could lead to negative outcomes (16). Another related transdiagnostic factor is difficulties in ER, which involves an individual’s capacity to detect, process, and reply to emotions (33, 41). Previous study in adults supports a connection between ER and AS over time (42). Prior research has found significant interacting effects despite the fact that AS and ER are distinctive transdiagnostic variables (43, 44). In fact, it has been demonstrated that AS and ER are connected to the emergence, maintenance, and management of anxiety-associated pathologies and symptoms (43–45). For instance, prior research indicates that reduced ER capacity may act to intensify AS, which in turn may impact a variety of other anxiety consequences such as worry and panic (43). Current empirical research and established theoretical models indicate that an individual’s ER capacity may have a role in the development of AS (44). On the other hand, elevated AS levels may theoretically make it more difficult for a person to use ER techniques that are necessary to reduce symptoms. Previous research in clinical populations demonstrates that modifications in ER and AS mediate abatement in anxiety symptoms in individuals with anxiety-related illnesses (e.g., social phobia, PTSD, and generalized anxiety disorder (45);. All of these findings point to the importance of ER in transdiagnostic anxiety processes, including sensitivity to anxiety and signs correlated with anxiety (41). However, it is not possible to interpret these effects as causative due to the cross-sectional character of these investigations.

The capacity of a person to withstand unfavorable emotional condition (DT) and their inclination to evaluate their own feelings of anxiety as detrimental (AS) are conceptually similar and yet can be considered as separate constructs. Schmidt et al. (2011) investigated the correlation between AS and DT, highlighting that DT involves a general capacity for enduring emotional distress, while AS pertains specifically to the endurance of anxiety and its associated feelings (41). They noted that DT encompasses a wide array of negative emotional experiences, but AS distinctly targets managing anxiety, suggesting the possibility that AS might act as an element within the broader trait of DT (22, 46). Accordingly, if individuals have lower grades of DT, they may be more likely to experience anxiety issues due to their reduced capacity to handle and endure negative emotional states, leading them to avoid or escape situations that provoke anxiety (28).

When DT and AS have been studied together, conflicting findings arise regarding their distinct connections with symptoms of anxiety. Some research shows that AS is strongly connected to anxiety symptoms, whereas DT is not (47), but alternate research suggests that DT is unusually associated with anxiety symptoms, whereas AS is not (24, 48). Differences in previous studies, findings are likely attributable to a variety of agents including disparities in the demographics of participants, such as their age and symptom severity, as well as methodological variations like the duration of the study and the methodologies applied (e.g., measures used).

Anxiety and AS are inversely correlated with DT (28, 49, 50). In both nonclinical and treatment-seeking populations, a correlation has been established between DT and AS (49, 51). Despite documented negative relationships between DT and AS, to our knowledge, no studies have clearly examined the associations among AS, DT, and symptoms of HA in one combined model

The research suggests that DT likely influences the choice of specific ER strategies an individual uses. In emotion research, high DT aids adaptive emotive reactions to life tension by serving as an ER skill. It enables individuals to select and implement effective strategies during challenging situations (52, 53). To express it another way, DT serves as a broad factor relating to individual differences that facilitates particular behaviors for regulating emotions. This viewpoint suggests that people with high DT are probable to engage in positive ER practices such as problem-solving and labeling emotions when faced with stress, while those with low DT may resort to less effective ER techniques like suppression and avoidance (21, 54). Linking adaptive ER behaviors might provide a reason for the relationship between DT and health outcomes (55). Quantitative study conclusions suggest that there are meta-analytic relationship between DT and various ER strategies. The investigation revealed that high DT showed moderate connections with maladaptive ER methods like rumination, worry, experiential avoidance and expressive suppression, with correlations ranging between -.19 and -.57. Conversely, DT had relatively weaker positive associations with effective ER including reappraisal, mindfulness and problem-solving, with correlations ranging between.08 and.38 (52). Previous investigation on DT in healthy individuals has also demonstrated that higher DT is linked to adaptive ER, encompassing impulse control and emotional awareness (56, 57).

The problem of HA is significant in clinical settings and requires further clarification. AS is a crucial factor that predisposes individuals to HA. AS has been identified in early studies as a significant risk factor for HA (58), playing a predictive role in this study.

While previous research has explored the relationships between AS and HA, a critical gap remains in understanding the mechanisms that explain this link. Specifically, few studies have examined the mediating roles of DT and ER, two constructs that are central transdiagnostic factors of emotional disorders. Our study is among the first to integrate these variables into a comprehensive model, highlighting their mediating roles in the AS-HA relationship. Additionally, understanding these mediators is crucial for developing targeted interventions. Unlike earlier studies that primarily focused on direct relationships, this research provides insights into how and why AS translates into HA, offering practical applications for clinical settings.

As mentioned earlier, both HA and AS are associated with abnormal processing of ER and DT, AS may decrease people’s DT and then cause HA symptoms. While ER and DT have each been separately associated with various anxiety outcomes, the current work explicitly models both factors as potential mediators. This dual-mediation design offers a more comprehensive view of the underlying mechanisms at play, thereby illuminating practical intervention targets. Therefore, the primary purpose of the study was to specify the role of ER and DT in the association between AS and HA (see Figure 1 ). We conceptualized AS as comprising of three dimensions: (1) physical concerns, (2) social concerns and (3) cognitive concerns, while HA was analyzed in terms of two dimensions: illness likelihood and negative consequences of illness. ER was defined with two dimensions: cognitive reappraisal (CR) and expressive suppression (ES), while DT was investigated in four dimensions: regulation, tolerance, appraisal, and absorption. Based on existing publications and previous results, we developed several assumptions. Prior research has indicated a link between AS and HA. Consequently, we hypothesized a positive relationship between AS and HA. Additionally, we proposed that DT, ES, and CR would act as mediators between AS and HA.

This study used a convenience sample of 971 participants. The Participants were from Tehran city located in Iran country. The age group of the participants varied between 18 and 65 (M_Total = 39.04, SD_Total = 10.64). 52.8% were female (513 females; Mean age = 38.63, SD = 10.80) and 47.2% were male (458 males; Mean age = 39.49, SD = 10.44). The educational background of the participants varied: 3.7% (n=36) had some undergraduate education, 21.1% (n=205) held a diploma, 39.1% (n=380) had earned a bachelor’s degree, 30.4% (n=295) had obtained a master’s degree, and 5.7% (n=55) had achieved a doctoral degree. Concerning employment status, 57.8% (n=561) were employed, 26.9% (n=261) were student, 10.9% (n=106) were unemployed, 3% (n=29) were retired and 1.4% had lost their jobs. Fifty-nine percent of the sample reported their marital status as married (n=573), 41% as Single/divorced/windowed (n=398).

The current study acquired approval from the Association for Ethics at the Iran University of Medical Sciences. Various social media platforms were utilized to recruit participants, from October to December 2023. Advertisements were placed on digital applications (such as WhatsApp, Instagram, Telegram) to invite participation in the investigation. A web link, which was provided in these advertisements, contained information about the study’s objective on its first page. By approving to participate and confirming the consent form, participants were then given access to do the surveys, submit demographic details, and receive a set of questionnaires. Participants must spend a minimum of 20 minutes completing questionnaires to ensure data quality. Participation required individuals to be a minimum of 18 years old. Participation was voluntary, with no compensation offered; moreover, participants were ensured confidentiality due to the unidentified essence of the questionnaire completion process.

Inclusion criteria comprised: (1) aged 18 to 65, (2) people who volunteered to take part in the study and (3) people who were able to done questionnaires on their own. Exclusion criteria included (1) inadequate proficiency in Persian and (2) no access to the Internet.

The study employed descriptive statistics to collection an overview of the participants’ socio-demographic characteristics and the scores of various study variables. To investigate the connections between AS and other variables, including mediators and outcomes, Pearson correlation analysis was conducted.

We applied a structural equation model (SEM) to test our hypothetical framework. Initially, the measurement models were assessed using the maximum likelihood (ML) approach, where all principal constructs were treated as latent variables. The suggested signs were described as follows: AS was represented by the combined scores of the three ASI-3 subscales; distress tolerance was reflected in the combined scores of the four DTS subscales; emotion regulation was denoted by the combined scores of the two ERQ subscales; and the total scores of health anxiety and negative consequences of illness denoted health anxiety. Ultimately, a SEM was designed to investigate the psychological pathways from AS to HA, considering sex and age as covariates for HA. The analysis was conducted using the ML method, and indirect effects were computed using a bootstrap technique with 5000 iterations. Statistical significance was determined at the 0.05 level if the bias-corrected bootstrap 95% confidence interval (CI) did not include zero (63). Based on the criteria proposed by Klin (2023), Chi-square index (χ2), comparative fit index (CFI; > 0.90), the goodness of fit index (GFI; < 0.95), the adjusted goodness of fit index (AGFI; < 0.90); in addition, values equal to or < 0.08 (values < 0.06 are more appropriate) for the RMSEA and SRMR suggest a good fit for the model (64). Statistical analyses were performed by utilizing the IBM SPSS (version 27.0) software package and the AMOS (version 24.0) software package.

Table 1 provides an overview of the means, standard deviations, ranges, skewness, and kurtosis for the total and subscale scores of the SHAI, DTS, ASI-3, and ERQ-10. To assess the normality of these scores, normal Q-Q plots, histograms, and boxplots were visually inspected, and the kurtosis and skewness ratios were examined for each measure. The analysis indicated that all four measures followed an approximately normal distribution. There were no notable discrepancies between the mean and the five percent trimmed mean for each scale, suggesting that the data did not contain any significant outliers that could skew the results. Furthermore, the Cronbach’s alpha values for each scale exceeded the commonly accepted threshold of 0.70, implying strong internal consistency.

To explore the linkages among the study variables, Pearson correlation coefficients were calculated, and the results are presented in Table 2 . The analysis revealed that AS had a statistically significant relationship with both DT and ER, and it was particularly positively correlated with HA.

The relationships between the study’s exogenous, mediator, and endogenous variables, both positive and negative correlations, are shown in Table 2 . The endogenous variable SHAI was moderately correlated with several variables: ASI-3 (r = .55, p <.001), DT (r = −.46, p <.001), ERQ-Expressive Suppression (ERQ-ES; r = .44, p <.001), and ERQ-Cognitive Reappraisal (ERQ-CR; r = −.42, p <.001). In contrast, the exogenous variable ASI-3 was negatively correlated with DT (r = −.40, p <.001), positively correlated with ERQ-ES (r = .44, p <.001), and negatively correlated with ERQ-CR (r = −.51, p <.001). Among the mediators, a significant negative relationship was discovered between DT and ERQ-ES (r = −0.40, p <.001), whereas DT and ERQ-CR exhibited a significant positive correlation (r = .38, p <.001).

The study assessed the mediating role of DT and ER on the relationship between AS and HA. To evaluate the proposed model, one SEM analysis was performed. Since there was adequate evidence supporting the assumption of normality, the MLE method was employed (64).

The model’s fit was evaluated using several criteria: chi-square (χ²), normed chi-square index (χ²/df), comparative fit index (CFI), goodness of fit index (GFI), adjusted goodness of fit index (AGFI), root-mean-square error of approximation (RMSEA), and standardized root mean squared residual (SRMR). The results showed that the model fit well with the data (χ² = 434.421, df = 144, χ²/df = 3.01, p < 0.001, GFI = .953, AGFI = .938, RMSEA = .046, SRMR = .052). The normed chi-square index was 3.01, which is below the acceptable limit of 5 (58). The RMSEA value for the model was.046, which is below the standard cut-off of.05 (64). The CFI, GFI, and AGFI values were.90,.95, and.93, respectively, all of which exceed the recommended value of.90 (58), indicating that the model fit the data well. Subsequently, standardized path coefficients for all proposed paths were analyzed, and the findings are depicted in Figure 2 .

Table 3 provides an overview of the standardized direct, indirect, and total effects within the proposed model. Bootstrap analysis confirmed that the total indirect effect of AS on HA is statistically significant (β = .45, p <.01). Specifically, the indirect pathways from AS to HA through DT (β = .10, p <.01), ER-ES (β = .11, p <.01), and ER-CR (β = .06, p <.01) were significant. Moreover, significant direct effects of DT (β = -.20, p <.01), ER-ES (β = .21, p <.01), and ER-CR (β = -.11, p <.01) on HA were observed. The direct effect of AS on HA remained significant even after considering the indirect effects, indicating that DT, ER-ES, and ER-CR act as partial mediators. Please see Table 3 for more information.

The findings of our study should be interpreted with caution due to several important limitations. These constraints are essential to consider when evaluating the results of this initial research. Firstly, the cross-sectional format of this study prohibits for conclusions about causality or the determination of temporal relationships. Although our findings suggest that DT and ER mediate the relationship between AS and HA, these conclusions should be considered exploratory and correlational rather than definitive. Longitudinal studies are needed to examine the directionality and potential causal pathways between AS, ER, DT, and HA over time. Without such evidence, we cannot definitively claim that improving DT or ER will directly reduce HA, or that these mediators are the underlying mechanisms of the relationship between AS and HA. Secondly, the sample composition must be acknowledged as a limitation. Our participants were not clinical subjects, and this limits the generalizability of the findings to populations with diagnosed health anxiety-related disorders, including illness anxiety disorder. The results may not apply to individuals with more severe manifestations of HA. Future studies should replicate this research with clinical populations to better understand how ER and DT operate in individuals with more pronounced HA symptoms. Thirdly, the use of convenience and snowball sampling methods in this study makes it difficult to ascertain whether the sample accurately represents the entire population of Iran. Future studies should consider using larger sample sizes and exploring the relationships between variables in diverse populations. Fourthly, the use of web-based surveys limits the ability to create random samples, as participants need internet access, leading to potential self-selection biases. It is recommended that this research be replicated on a larger national scale and that qualitative methodologies be used to uncover cultural nuances that quantitative methods cannot capture. Fifthly, although there are known disparities among ethnic groups in ER (62), this study did not control for potential cultural variances. In addition, there is a shortage of specific demographic data such as ethnicity, race, and socioeconomic status necessitates cautious interpretation of our findings. A further limitation is the reliance on self-report measures, which might potentially create common method variance. Future research could profit from adopting multimethod approaches, such as behavioral indicates of ER and DT.

On the clinical level, these findings show that those who fail to tolerate unpleasant emotions are at greater risk to develop different kinds of psychopathology related with emotion dysregulation, such as illness anxiety disorder. Our results indicate that treatments aimed at reducing HA symptoms by focusing on DT and ER could be particularly effective. Broadly targeting DT and ER may help lower HA levels. Understanding the role of DT as a mediator can assist physicians recognize those patients who will probably to benefit the most from specific therapies and establish suitable treatment suggestions. For example, people who have low DT levels might benefit more from Dialectical Behavior Therapy (DBT), which is specifically designed to enhance DT skills. DBT, a form of cognitive-behavioral therapy, incorporates Instruction in skills and functioning behavioral assessment alongside ideas of dialectical philosophy, mindfulness and styles of communication (83). DBT has been shown to improve ER in individuals with various anxiety and mood disorders (84). Since DBT targets ER, it may be an effective treatment for HA. Although no studies have yet employed DBT specifically for HA, research combining ER treatments with DBT has shown positive outcomes for individuals with HA. Additionally, DBT can educate persons who employ inefficient coping mechanisms in response to an unable to handle unpleasant negative emotions to find more effective strategies for addressing negative impact.

Our findings emphasize the importance of targeting both DT and ER in therapeutic interventions for HA. Cognitive-behavioral therapy (CBT) has been shown to be effective in enhancing DT by equipping individuals with skills to endure and manage distressing emotions (85). Additionally, metacognitive therapy (MCT) offers a promising approach to addressing difficulties in ER by focusing on modifying dysfunctional metacognitive beliefs (86). MCT aims to reduce the reliance on maladaptive forms of mental control and promote adaptive strategies for managing emotional distress. Incorporating these therapeutic approaches may improve outcomes for individuals experiencing health anxiety by addressing both the underlying mechanisms of DT and ER. These insights contribute to a more comprehensive understanding of the relationship between AS and HA and highlight the potential for integrative therapeutic approaches that address the interplay of DT, ER, and metacognitive processes.