research articles social anxiety disorder

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• PMC10108597

Recent advances in the understanding and psychological treatment of social anxiety disorder

Kate wolitzky-taylor.

1 Department of Psychiatry, University of California, Los Angeles, CA, USA

Richard LeBeau

2 Department of Psychology, University of California, Los Angeles, CA, USA

Social anxiety disorder (SAD) is characterized by persistent anxiety or avoidance of social situations because of a fear of negative evaluation. Cognitive behavioral therapy (CBT) (typically with an exposure component) is a first-line treatment for social anxiety, but there remains room for improvement with regard to treatment efficacy. Therefore, the field continues to better understand the mechanisms underlying SAD and its common and complex comorbidities in order to develop targeted interventions to improve symptom outcomes. Additionally, efforts are under way to improve the efficacy and accessibility of CBT. This review outlines major advances in understanding and treating SAD in adults over the past roughly 3 years (2019 to early May 2022). Themes are identified and discussed, as are recommendations for future research.

Current status of social anxiety disorder and its treatment

The diagnostic classification and criteria for social anxiety disorder (SAD), in contrast to those of many other prevalent disorders, have gone largely unchanged in recent decades. However, that is not to imply that research on SAD has been stagnant. In reality, an astonishing amount of cutting-edge research on SAD has been conducted in recent years. Such research is necessitated by the fact that SAD remains a prevalent condition, whose estimated 12-month prevalence in the United States is 8% 1 . It is also a markedly disabling condition with an early onset 2 and persistent course 3 . SAD is associated with significant impairment in social, occupational, and academic domains as well as decreased quality of life 4 .

Fortunately, several effective treatments for SAD exist. The gold-standard psychosocial treatment for SAD is cognitive behavioral therapy (CBT) delivered in individual or group format 5 – 7 . Increasing evidence also supports the effectiveness of acceptance and commitment therapy (ACT) for SAD 8 – 10 . Various medications, particularly selective serotonin reuptake inhibitors (SSRIs), have been shown to be effective for SAD as well 5 . However, the ability to access evidence-based treatments remains a challenge 11 , 12 . Even among those who are able to access evidence-based treatments, many fail to respond to treatment or relapse after completing treatment 13 . Thus, ongoing research is needed to figure out how to maximize the accessibility and effectiveness of current treatments as well as to develop novel treatments.

The purposes of this review were to provide a thematic glimpse into the most recent developments in the understanding and management of SAD and to guide researchers and clinicians to particularly important articles for a more in-depth subsequent review. We focused on this broad topic from a psychological lens. Although we acknowledge the many contributions from the fields of neuroscience, psychiatry, and pharmacology/pharmacogenetics, as well as child and adolescent psychology and psychiatry, we discuss advances in our understanding of the psychological processes underlying SAD and of behavioral (and cognitive) treatment advances for SAD in adults. Importantly, we also discuss the recent emphasis on understanding SAD and its treatment in the context of common and complex psychological comorbidities.

Methods for this review

Our review consisted of PubMed and Google Scholar searches for articles from 2019 to 2022 with the following search terms: “social anxiety disorder,” “social anxiety disorder treatment,” “social anxiety etiology,” and “social anxiety assessment.” We conducted an extensive review of abstracts that appeared and then we selected articles that reported on the relevant themes of this review. Review articles and meta-analyses that were published between 2019 and 2022 but that synthesized findings from previously published research were excluded. Note that the review was conducted in spring 2022, so any studies published after the first week of May 2022 were not reviewed.

Recent advances in improving the understanding of SAD

Factors contributing to the onset and maintenance of sad.

Recently, a great deal of research has been conducted on factors that are related to the onset and maintenance of SAD. A brief overview of the research conducted on a large array of constructs is provided here. A notable area of recent research involves examining the role of adverse social experiences in the development of SAD. Bjornsson and colleagues 14 examined “social trauma,” events which involve humiliation and rejection and would not necessarily be categorized as Criterion A traumatic stressors, because they were not life-threatening. They found that social trauma was highly prevalent among individuals with SAD and that nearly one third of their sample experienced clinically significant post-traumatic stress disorder (PTSD) symptoms related to their social trauma. This work highlights the potential clinical importance of understanding socially traumatic events and the possible limitations of our traditional definitions of what constitutes a trauma. Brühl and colleagues 15 examined adverse childhood experiences and found that adverse experiences that were social in nature were not uniquely associated with SAD relative to other mood or anxiety disorders. Although contrary to expectations, this study was limited by being both retrospective and cross-sectional. This limitation is frequently observed in research that tries to elucidate the relationship between adverse early experiences and later psychopathology. Longitudinal research studies that are designed to examine not only which adverse experiences (e.g., social trauma) are associated with SAD onset but how they relate to SAD onset and whether they are uniquely associated with SAD onset (or more broadly associated with onset of psychopathology) are needed.

A great deal of research has been conducted on cognitions, emotional states, and behaviors that may serve as maintenance factors for SAD. Most of these studies were conducted with relatively small, non-clinical samples in laboratories. Oren-Yagoda, Schwartz, and Aderka 16 analyzed data from a daily experience study and found that individuals with social anxiety experience higher levels of envy than those without social anxiety and that experiences of envy (particularly in social contexts) uniquely predicted subsequent anxiety. Arch and colleagues 17 found that individuals with SAD had more frequent off-task thoughts and rated these thoughts as less controllable, more personal, and associated with worse mood than those without social anxiety. Bailey and colleagues 18 found that perseverative cognition was highly prevalent among individuals with SAD and was associated with lower heart rate variability following a negative social interaction. Goodman and colleagues 19 found that although individuals with SAD derived less pleasure from social interactions than those without, they still derived positive emotions from social interactions and experienced more positive emotions than they did from non-social situations. Maleki and colleagues 20 found that affective theory of mind (i.e., ability to infer others’ emotional states and feelings) but not cognitive theory of mind (i.e., ability to infer others’ beliefs and intentions) was disrupted in individuals with SAD relative to healthy controls. Romano, Tran, and Moscovitch 21 found evidence of context-specific memory impairments among individuals with SAD for social situations with positive outcomes.

Additionally, new research has been conducted on the role of behavioral avoidance 22 , safety behaviors 23 , attentional bias to threatening faces 24 , gaze avoidance 25 , delays in attention disengagement from social threat 26 , rumination 27 , 28 , blushing 29 , inattention, intolerance of uncertainty 30 , and fearful attachment style 31 .

Although the studies described above utilized different methods and focus on different cognitions, emotional states, and behaviors, they all provide interesting examples of how theoretically driven research can help elucidate key maintenance factors in SAD. The identification of such factors is essential for researchers to develop novel treatments and refine and enhance our existing treatments to ensure that they are effectively and efficiently targeting key mechanisms through which symptom reduction can occur. These studies also underscore the need for treatment studies to include measures (both self-report and more objective measures) of hypothesized mechanisms in their assessment batteries. Greater inclusion of such measures in treatment trials would also help overcome some of the limitations caused by relying on small, clinical analogue samples for mechanism research. It is important that future research seek to elucidate which of these constructs are unique to SAD versus, more broadly, characteristics of anxiety disorders or all mental disorders, as such knowledge is useful for treatment development (particularly the development of treatments for comorbid conditions and transdiagnostic treatments).

Factors contributing to impairment in SAD

Several research investigations in recent years have continued to examine the high levels of impairment experienced by many individuals with SAD. A well-designed investigation by Vilaplana-Pérez and colleagues 32 utilized data from over 2 million individuals born in Sweden and presented some of the most compelling and nuanced data to date regarding markedly worse educational performance across the life span for individuals with SAD relative to those without. The effect was particularly prominent for females. Impressively, their findings remained significant after controlling for psychiatric comorbidity and a host of genetic and social factors (utilizing data from siblings without SAD). Peyre and colleagues 33 utilized data from a large, nationally representative epidemiological survey to examine the relationship between age of onset of SAD and functioning. Interestingly, they found that a later age of onset was associated with greater psychiatric comorbidity and diminished quality of life. Krygsman and Vaillancourt 34 analyzed data from over 700 adolescents and found that those who had a trajectory of high and increasing social anxiety symptoms from 5th to 12th grade were much more likely than those with moderate and low symptom trajectories to meet criteria for major depression and cannabis use disorder as adults. Investigations such as these leverage large samples and sophisticated analytical methods to improve our understanding of the factors that contribute to worse outcomes for individuals with SAD and help us figure out what subgroups might need additional intervention and the type of additional interventions they may need (e.g., educational interventions for female children and adolescents with SAD). Future research should also seek to disentangle the degree to which observed impairments result from SAD as opposed to conditions comorbid with SAD to improve our understanding of where prevention and intervention efforts are most needed.

Understanding SAD comorbidity

Numerous studies that seek to elucidate our understanding of SAD comorbidity have been conducted in recent years. Two notable studies examined the comorbidity between SAD and alcohol use disorder (AUD) and came to markedly different conclusions. Torvik and colleagues 35 analyzed a longitudinal dataset of 2,801 Norwegian pairs of twins and found that SAD had a direct effect on the development of AUD that was not found for other anxiety disorders. In contrast, Miloyan and van Doorn 36 analyzed data from two nationally representative longitudinal datasets and found that neither lifetime subclinical social fears nor full SAD was associated with later AUD.

Wong 37 applied sophisticated statistical methodology to data from outpatients with early psychosis and found a direct relationship between the severity of negative symptoms and ideas of reference with social anxiety symptoms that provides greater understanding for the frequently observed comorbidity between social anxiety and psychosis. Langer and colleagues 38 conducted network modeling to better understand the well-established comorbidity between SAD and major depressive disorder (MDD) and concluded that the comorbidity likely results from direct relationships between associated nodes as opposed to hallmark symptoms of the disorder. Frandsen and colleagues 39 examined the relationship between SAD and avoidant personality disorder (APD) and concluded that although SAD is associated with lower levels of interpersonal distress than APD, SAD is related with higher levels of phobic anxiety than APD. They also concluded that both disorders feature non-assertiveness as a core feature.

Recent research has also sought to understand the relationship between SAD and suicidality. Chung and colleagues 40 analyzed data from a large sample of individuals with SAD and found support for the interpersonal theory of suicide 41 , according to which perceived burdensomeness and thwarted belongingness were significantly predictive of acute suicidal ideation in individuals with SAD over and above the effects of depression. Duffy and colleagues 42 analyzed data from individuals with social anxiety and found that perceived burdensomeness but not thwarted belongingness was associated with previous two-week suicidal ideation over and above the effects of depression. It is plausible that the divergence in findings was related to markedly different time frames for the suicidality assessment between the two studies. Yarrington and colleagues 43 examined suicidality in a large and diverse sample of individuals with SAD who were seeking community employment services. The authors found that suicidal ideation and a history of suicide attempts were highly prevalent but that they were largely accounted for by depression comorbidity, which was experienced by virtually all of the sample.

Taken together, these studies have contributed to a deeper understanding of SAD comorbidity and, in several cases, elucidate the explanatory variables that account for the comorbidity. However, given some of the discrepant findings described above, additional research is needed to ascertain whether an accumulation of more data will point to more consistent findings in one direction or another (e.g., for the role of social anxiety in contributing to AUD onset). Research on SAD comorbidity is also critical to helping us better understand what clinical features are unique to SAD and which are shared by broader classes of disorders (e.g., anxiety disorders).

Recent advances in improving the assessment of SAD

Self-report measures.

Even though numerous valid and reliable self-report measures are available for SAD, two new measures that addressed gaps in the available measures have emerged in recent years. The Socially Anxious Rumination Questionnaire (SARQ) 44 is a 24-item measure assessing negative thoughts before (pre-event) and after (post-event) distressing social situations. The initial validation study showed promising results for its utility assessing rumination both before and after social interactions, which has long been identified as a key maintaining factor for social anxiety symptoms. The Ryerson Social Anxiety Scale (RSAS) 45 is a 23-item measure developed to assess the distress and impairment associated with SAD, thus providing important information beyond most existing scales, which tend to focus on how severe and frequent the symptoms are without assessing broader impairment. A follow-up study has confirmed its validity and reliability in a clinical sample of patients with social anxiety 46 . These studies highlight the growing trend in clinical research and practice to move beyond simply assessing symptoms but to also assess maintaining factors that may be promising targets for treatment as well as to assess impairment and quality of life, which are considered patient-centered outcomes.

In addition, further validation efforts on existing scales have occurred in recent years. For example, a large-scale psychometric investigation provided further validation of the Multidimensional Social Anxiety Response Inventory – 21 (MSARI-21) 47 ; the validity and reliability of the Self-Rated Beliefs in Social Anxiety (SBSA) scale were confirmed in a clinical sample 37 ; an investigation in Brazil provided compelling data that translations of the widely used Liebowitz Social Anxiety Scale are valid and reliable tools for assessing social anxiety in global Spanish- and Portuguese-speaking individuals 48 ; and data from large clinical and non-clinical samples confirmed that the computerized adaptive version and static short-form version of the widely utilized Social Interaction Anxiety Scale (SIAS) and Social Phobia Scale (SPS) are as valid and reliable as the original, lengthier versions 49 . Investigations such as these are critical to ensure that we have valid and reliable tools that assess all relevant aspects of social anxiety and do so in a manner that is culturally appropriate and brief for use in busy clinical and research settings.

Leveraging technological advances to improve assessment of SAD

Clinical interviews and self-report measures will undoubtedly continue to play a critical role in diagnosing SAD and measuring change in severity of SAD symptoms over time. However, several researchers are leveraging recent technological advances to develop novel methods for assessing the disorder. One such method involves using data passively gathered from an individual’s smartphone (e.g., social interactions, geospatial movement, and activity level) to predict symptom trajectories. Jacobson and Bhattacharya 50 recently found that such data could be utilized in personalized deep learning models to predict future symptoms of anxiety, as well as hour-to-hour changes in anxiety symptoms, among individuals with scores in the clinical range of well-validated self-report measures for SAD and generalized anxiety disorder (GAD). New methods for SAD assessment are also emerging from advances in machine learning, which refers to the use and development of computer systems that are able to learn and adapt without explicit instructions by using algorithms to analyze and draw inferences from patterns of data. A recent investigation by Demetriou and colleagues 51 utilized sophisticated machine-learning algorithms to correctly differentiate three clinical conditions related with social difficulty: SAD, autism spectrum disorder, and prodromal psychosis. These studies provide excellent examples of the new frontiers that are being explored with respect to the assessment of SAD and other highly prevalent and impairing disorders.

Recent advances in improving the treatment of SAD

Digital therapy.

As described above, CBT already has a solid base of evidence. Therefore, advances in the improvement of CBT for SAD in recent years have largely emphasized ways to increase access to treatment as well as understanding moderators and mediators of treatment outcome in order to inform treatment personalization and targeting of presumed mechanisms, respectively, both of which should improve outcomes.

One of the largest areas of contribution in the past few years has been in the area of employing digital therapy or virtual treatment for SAD. Digital therapy has the benefit of increasing access to care by reducing traditional barriers that can make coming into a clinic on a weekly basis difficult. However, findings from recent studies specifically evaluating digital therapy for SAD typically were limited by high attrition and weak comparison conditions. One study found that nine sessions of unguided internet-based therapy for SAD showed moderate effects on measures of social anxiety compared with waitlist control 52 . Although attrition from the study assessments was generally low, nearly half of the participants stopped completing sessions by session five out of nine, suggesting that most did not receive the full course of treatment. This is particularly notable given that the intervention did not introduce behavioral experiments or in vivo exposure until the last few sessions, so the intervention that most participants received was cognitive restructuring, thought to be efficacious but a less potent treatment ingredient than exposure for SAD. On the one hand, low treatment adherence/completion could be considered one of the many “failures” to retain patients through a prescribed course of CBT for SAD using self-guided, digital therapy. Indeed, attrition rates tend to be high for self-guided, app-based therapy 53 . On the other hand, patients showed significant and reliable improvement in SAD symptoms fairly rapidly, suggesting that perhaps patients self-discontinue treatment when they have achieved the benefits they aim to receive. Future studies with larger samples can elucidate dose–response relationships and the patterns that characterize treatment discontinuation and symptom improvement.

Similarly, Powell and colleagues 54 evaluated a web-based self-help intervention for social anxiety and compared it with a waitlist control group. Even with an inactive control group, the improvements in social anxiety symptoms and cost-effectiveness (one presumed strength of self-directed interventions) were small. A web-based, self-guided, seven-session SAD intervention showed large effects on social anxiety symptoms (in an open trial) but only among those who completed treatment 55 . Attrition was also high in this study; only 38% of participants got to the fourth module, when behavioral experiments began. Similar to the study by Kahlke and colleagues 52 , this suggests that most people received cognitive restructuring but discontinued the self-guided treatment before any behavioral experiments or exposures began.

Collectively, these recent findings in web-based and digital therapies for SAD point to several issues that need to be addressed. Despite their proliferation and the enthusiasm that they bring to researchers and clinicians in terms of potential accessibility and cost-effectiveness, web-based and digital therapies must be delivered in ways in which patients adhere to them and see significant benefit. One promising approach may be in the area of digital approaches that utilize personalized engagement strategies, such as text messaging. Indeed, a brief text messaging intervention that encouraged patients to eliminate safety behaviors outperformed a comparison text messaging condition that encouraged individuals to focus on the present moment on social anxiety symptoms 56 . Another promising approach may be to directly target this lack of engagement and low treatment seeking in this population using web-based approaches. Indeed, a randomized trial of brief online interventions to increase treatment engagement in individuals with social anxiety found that a single-session online intervention that provided education and motivational strategies to individuals self-reporting social anxiety symptoms resulted in a significant uptake of treatment services 57 . Although this study did not have the benefit of comparison with a control condition that received no intervention (i.e., the conditions both included some form of intervention and both showed uptake increases) and thus it is difficult to draw conclusions about how these rates of uptake may differ from those of the general population of individuals who self-report social anxiety symptoms in the context of enrolling in an online study for individuals with social anxiety, this is nonetheless a promising finding and is particularly notable given its brevity and thus potential scalability. Future research should focus on strategies to increase engagement in digital therapy for SAD. This may include personalized feedback, live coaching via videoconference or text messaging, or adapting interventions so they are personalized and tailored to the individual’s needs in real time. Of note, some studies prior to the window of this review (i.e., before 2019) have explored some of these engagement strategies, particularly coach- or therapist-assisted digital platforms, and have demonstrated that this is a promising strategy for engagement in digital therapy and is deserving of future research.

Novel approaches to delivering cognitive behavioral therapy for SAD

In addition to utilizing technology for the delivery of interventions for SAD, technology has recently been leveraged in novel ways. One study (a small pilot) evaluated the efficacy of ACT (a third-wave behavior therapy and efficacious treatment for SAD) delivered via videoconferencing for public speaking fear 58 . Virtual audiences were video recorded for homework exposures, a unique way to overcome the barrier that many CBT clinicians face in developing realistic, repeatable, individualized, and generalizable exposures for homework in this population with public speaking fear. Importantly, there were no statistically significant differences between those who received in vivo exposure versus those who received only the teleconference-based exposure, suggesting that this may be a viable alternative for patients with public speaking fears who do not have access to audiences or venues with which to practice public speaking exposures. However, these findings should be interpreted cautiously, given the small sample size. Replication with larger samples may be able to detect differences between two active conditions, which is difficult to do with small samples.

Other studies have recently used virtual reality to conduct exposure therapy for SAD 59 , 60 . Benbow and Anderson 59 used a group therapy modality and showed significant long-term reductions in threat probability and cost estimates (i.e., two prominent cognitive mediators in CBT for SAD). Geraets and colleagues 60 reported findings from a small, uncontrolled pilot study of virtual reality CBT for SAD and demonstrated improvements in SAD symptoms. These studies point to the exciting use of technology that should be explored further to develop ways to overcome some of the barriers to delivering in vivo exposure for patients with social anxiety, particularly those with performance-related anxiety that requires large groups of people to serve as an “audience” for exposure. However, randomized clinical trials with stringent control groups and larger sample sizes are needed to strengthen the rigor of this line of research.

Addressing comorbidity

SAD is highly comorbid with other psychopathology, specifically MDD 61 and substance use disorders (SUDs) 62 . As such, research that aims to better understand how treatment works for patients with these comorbidities, as well as how to improve upon treatments for patients with these comorbidities, can make a significant impact. A recent innovative study developed and evaluated a novel behavioral intervention that fully integrated treatment for SAD and AUD 63 . Patients were randomly assigned to a typical evidence-based intensive outpatient program for SUDs in a community-based SUD setting or to an intensive outpatient program that integrated CBT for SAD into the typical content and adapted the content to meet the specific needs of this population. The group therapy format common in SUD specialty clinics, which typically is seen as a barrier to successful engagement in addiction treatment for patients with SAD, was instead viewed as an opportunity to practice exposure in the context of alcohol recovery activities. Findings demonstrated superiority of the integrated treatment over the traditional treatment on social anxiety and some measures of alcohol use. These findings pave the way for integration of CBT for social anxiety into the settings where it is often prevalent but inadequately treated. Along similar lines, a recent study compared treatment for AUD only with an integrated treatment for SAD and AUD that included CBT and motivation interviewing 64 . Findings were somewhat similar: both conditions showed improvement in social anxiety and alcohol use, and the integrated treatment was superior to the control condition. However, no between-group effects were observed on alcohol use outcomes in this study. Taken together, these findings illustrate the importance of continuing to refine, improve, and augment these integrated interventions to simultaneously address this prevalent comorbidity.

In addition to this recent work, a research group has been exploring the role of depressive symptoms and MDD diagnoses on SAD treatment outcomes. One study found that, among those with a MDD diagnosis, change in social anxiety and depressive symptoms did not mediate change in each other but that change in social anxiety symptoms mediated change in depressive symptoms among those who had SAD and no MDD 65 . These findings suggest that change in social anxiety may lead to change in depressive symptoms among those with milder depressive symptoms, whereas this relationship was not observed at higher levels of depressive severity. Interestingly, and perhaps at odds with conventional wisdom, these researchers uncovered that depression was associated with better social anxiety outcomes among those with SAD who received treatment (broadly including individual CBT, internet-based CBT, group CBT, or pharmacology) 66 . The effect of greater depressive symptoms on better social anxiety outcomes was particularly large for those who had received individual CBT or internet-based CBT (but not for group CBT or pharmacology). These findings are particularly interesting as they call into question typical misconceptions that many community clinicians hold about CBT for anxiety disorders as being most appropriate, suitable, and effective for those without complex comorbidity. Indeed, patients with the common comorbidity of depression and SAD see significant benefit from a course of CBT for SAD.

Taken together, the last few years have generated only a few new articles specifically focused on treatment for SAD in the context of common comorbidities; but those few studies have yielded interesting findings with practical and direct clinical implications.

Mediators and moderators of treatment outcomes for those with SAD

In addition to the studies described above that have touched on mediators of treatment outcome and change in presumed mechanisms of change 59 , 65 and moderators/predictors of treatment outcome 66 , studies in recent years have attempted to understand for whom CBT for SAD works best (moderators) and the mechanisms by which change may occur during treatment (mediators). These fine-grained analyses provide clear prescriptive information for treatment selection based on individual difference characteristics and inform researchers about how to augment or modify treatments to directly address treatment targets that are the presumed mechanisms of change.

In a comparison of CBT with ACT for SAD, Sewart and colleagues 67 examined changes in negative and positive affect in both conditions and also evaluated whether baseline negative and positive affect moderated treatment outcomes (i.e., differentially affected treatment outcomes between conditions). Negative affect decreased in both conditions, positive affect increased in both conditions, and neither negative nor positive affect at baseline moderated social anxiety outcomes. These findings indicate that ACT and CBT have common mechanisms despite their distinct theoretical approaches.

Another interesting study evaluating mechanisms of change asked an age-old question among CBT clinical researchers: “Does changing cognitions change behavior, or does changing behavior change cognitions?” Using a sample of patients with SAD who received CBT, the study found that avoidance behavior predicted cognitions at subsequent timepoints but not vice versa 68 . This is a fascinating finding showing the direct pathway from behavior to cognition but not the opposite, which is the more traditional way that the cognitive model is presented to patients. However, it is in line with a solid body of work showing that exposure therapy without any cognitive strategies works directly on changing threatening associations with feared stimuli (see Hofmann 69 for a review).

Other interesting studies that aim to better understand how treatments work have observed that sudden gains in CBT and mindfulness-based interventions for social anxiety do not predict better outcomes by the end of treatment 70 , added to the existing literature supporting the importance of targeting social cost as a cognitive maintaining factor and finding that decreases in social cost estimates during exposure are associated with better social anxiety outcomes 71 , and identified that changes in shame mediated outcomes during exposure for SAD 72 . Additionally, a unique study auditing charts from real-world clinic patients found that typical treatment mediators observed in CBT for SAD in research settings also demonstrated significant mediation, including self-focused attention, negative social cognitions, and depressed mood 73 .

Finally, an interesting moderator study categorized participants as high or low in attentional biases at baseline 74 . These participants were randomly assigned to attention bias modification training (an alternative to CBT; see Hakamata et al. 75 for a review) or placebo. Findings revealed no pre- to post-treatment effects on attention biases or anxiety in the condition that received the active intervention, and baseline attention biases did not moderate treatment. This study adds to the growing body of literature observing either null or small effects for attention bias modification for SAD 76 , which is unfortunate given the enthusiasm and promise it showed about a decade ago when it was developed and began to be evaluated as a low-intensity treatment option for anxiety disorders. Moreover, the fact that the intervention did not change the presumed mechanism of change is concerning.

Future directions, recommendations, and conclusions

As illustrated above, SAD is an example of a fairly well-understood disorder with efficacious treatments. Yet, in a span of just three years, a tremendous amount of work has been done to improve our understanding and treatment of this prevalent and disabling disorder. Common themes and current directions in this work include (1) understanding the role of, assessing, and targeting cognitive and behavioral maintaining factors that contribute to the onset and maintenance of social anxiety; (2) understanding and treating common comorbidities with SAD, particularly AUD and MDD; and (3) evaluating strategies for increasing accessibility to evidence-based behavioral treatment for SAD through digital virtual and other technology-assisted modalities. Replication is needed for many of the etiological studies (with longitudinal designs when possible), and efforts are needed to improve engagement and retention in digital therapy programs. Nonetheless, much progress has been made in the understanding and treatment of SAD in the past few years. Future research should include randomized designs with larger samples that allow fully powered moderator and mediator analyses to truly identify factors that can optimize treatment outcomes for patients with SAD and common comorbidities.

Funding Statement

The authors declare that no grants were involved in supporting this work.

The peer reviewers who approve this article are:

• Katharina Kircanski , National Institutes of Mental Health, Bethesda, MD No competing interests were disclosed.
• Tomas Furmark , Department of Psychology, Uppsala University, Uppsala, Sweden No competing interests were disclosed.
• David M Clark , Department of Experimental Psychology, University of Oxford, Oxford, UK No competing interests were disclosed.
• Eleanor Leigh , Department of Experimental Psychology, University of Oxford, Oxford, UK No competing interests were disclosed.

• Research article
• Open access
• Published: 23 April 2019

Social anxiety increases visible anxiety signs during social encounters but does not impair performance

• Trevor Thompson   ORCID: orcid.org/0000-0001-9880-782X 1 ,
• Nejra Van Zalk 2 ,
• Christopher Marshall 3 ,
• Melanie Sargeant 4 &
• Brendon Stubbs 5  

BMC Psychology volume  7 , Article number:  24 ( 2019 ) Cite this article

22k Accesses

4 Citations

40 Altmetric

Metrics details

Preliminary evidence suggests that impairment of social performance in socially anxious individuals may be specific to selective aspects of performance and be more pronounced in females. This evidence is based primarily on contrasting results from studies using all-male or all-female samples or that differ in type of social behaviour assessed. However, methodological differences (e.g. statistical power, participant population) across these studies means it is difficult to determine whether behavioural or gender-specific effects are genuine or artefactual. The current study examined whether the link between social anxiety and social behaviour was dependent upon gender and the behavioural dimension assessed within the same study under methodologically homogenous conditions.

Ninety-three university students (45 males, 48 females) with a mean age of 25.6 years and varying in their level of social anxiety underwent an interaction and a speech task. The speech task involved giving a brief impromptu presentation in front of a small group of three people, while the interaction task involved “getting to know” an opposite-sex confederate. Independent raters assessed social performance on 5 key dimensions from Fydrich’s Social Performance Rating Scale.

Regression analysis revealed a significant moderate association of social anxiety with behavioral discomfort (e.g., fidgeting, trembling) for interaction and speech tasks, but no association with other performance dimensions (e.g., verbal fluency, quality of verbal expression). No sex differences were found.

Conclusions

These results suggest that the impairing effects of social anxiety within the non-clinical range may exacerbate overt behavioral agitation during high demand social challenges but have little impact on other observable aspects of performance quality.

Peer Review reports

Social anxiety disorder (SAD) is a common psychiatric disorder, with up to 1 in 8 people suffering from SAD at some point in their life [ 1 ]. SAD is linked to reduced quality of life, occupational underachievement and poor psychological well-being, and is highly comorbid with other disorders [ 2 ]. Mounting evidence suggests that social anxiety exists on a severity continuum [ 3 ], and that social anxiety that is not severe enough to warrant a diagnosis of SAD may still produce significant individual burden [ 4 ].

There is little evidence to suggest that social anxiety may negatively affect others’ perceptions of agreeableness or warmth [ 5 ]. However, if social anxiety impairs an individual’s ability to function effectively in common performance situations such as job interviews, presentations and other social challenges [ 6 ], this could cause or maintain feelings of failure and inadequacy and even affect career success [ 7 ]. Cognitive models [ 8 ] predict that social anxiety could impair social competence by increasing self-focused attention and consuming attentional resources necessary for effective communication. On the other hand, social anxiety can also lead to a willingness to engage in socially-facilitative behavior such as polite smiling, head nodding and avoiding interruption, which can facilitate interaction and lead to more favorable impression of another’s social behavior [ 9 ].

While socially anxious individuals reliably believe their social behavior is deficient, the existence of actual impairment has been the subject of a fair amount of debate [ 10 ]. Empirical studies that have examined the association between social anxiety and behavior in response to social challenge tasks in both clinical and non-clinical samples have produced inconsistent findings. Strahan and Conger [ 11 ], for example, compared the responses of 26 men with low social anxiety with 27 men reporting clinical levels of social anxiety on the Social Phobia and Anxiety Inventory in their response to a simulated job interview. Observer ratings of videotaped interviews indicated no group differences in overall social competence ratings. Rapee and Lim [ 12 ] found that, when asked to give a brief impromptu speech, a group of 28 individuals with SAD did not differ in observer ratings of overall performance relative to a group of 33 non-clinical controls. Similar null results have been reported in a non-clinical sample of males on overall impressions of social skill on an opposite-sex “getting to know you” task [ 13 ], and in a sample of 110 schoolchildren participating in a two-minute impromptu speech where observers rated video recordings for global impressions and “micro-behaviors” (e.g., clarity of speech, ‘looking at the camera’) [ 14 ].

However, a number of other studies have identified a link between social anxiety and impaired social behavior. Levitan et al. [ 15 ] found that patients with SAD were rated significantly more poorly on observer ratings of voice intonation and fluency during a three-minute speech compared to controls. Other studies have also found patients with SAD to be rated more poorly by observers on adequacy of eye contact and speech clarity [ 16 ] and as exhibiting more “negative social behaviors” (e.g. awkwardness) during conversations [ 17 , 18 ]. In a non-clinical study of 48 women, Thompson and Rapee [ 18 ] found individuals with high social anxiety to be rated more poorly during an opposite-sex “getting to know you” task on summed measures of molecular (e.g. voice quality, conversational skill) behaviors and on overall impression.

A recent review by Schneider and Turk [ 10 ] suggests that the apparently variable link between social anxiety and behavior is likely to be influenced by differences across studies in factors such as statistical power, sample characteristics and the type of behavioral assessments used. Assessment measures, for example, have ranged from global impression ratings to composite scores of molecular behaviors (e.g., smiling frequency, eye contact), and it may be that social anxiety impairs certain social behaviors but not others. There is some evidence that social anxiety may selectively exacerbate observable anxiety signs but have little impact on performance ‘quality’ (e.g. factors central to effective communication) [ 14 , 19 ]. Schneider and Turk [ 10 ] note, however, that it is difficult to identify a coherent pattern that identifies which aspects of performance may be impaired by social anxiety and which may not and this is additionally complicated by differences in study designs. Furthermore, where associations of social anxiety across multiple behavioral dimensions have been examined within the same study, where they are evaluated under the same conditions, these differences have rarely been compared statistically which limits the reliability of the current evidence for selective deficits in social behavior [ 20 ].

Norton [ 21 ] also notes that studies using exclusively female samples have often found stronger associations of social anxiety with behavioral deficits than studies with male samples, consistent with the argument that gender-role expectations may lead to more deleterious effects of social anxiety in women [ 22 ]. Again, however, it is impossible to determine with any certainty whether more pronounced effects of social anxiety in studies with females is attributable to moderating effects of gender or some other difference in study characteristics. Unfortunately, few studies have directly compared males and females, or different performance dimensions, within the same study where there is greater methodological homogeneity.

This study aimed to assess social behavior during social challenges in a non-clinical sample of individuals varying in their levels of social anxiety. We used speech and interaction tasks, as these represent different types of commonly-encountered social challenges. Performance was assessed by independent raters using Fydrich’s Social Performance Rating Scale, which consists of five separate dimensions of social competence. The aim of the study was to examine whether social anxiety is associated with impaired social behavior, and in particular: (1) whether impairment occurs only for specific dimensions of behavior, and (2) whether impairing effects are greater in females.

Participants

The sample consisted of 93 participants (45 males and 48 females) with a mean age of 25.6 years ( SD  = 7.7, Range = 18–53). Males ( M  = 26.5 years) and females ( M  = 24.7 years) did not differ significantly with respect to age, t (86)  = 1.12, p  = .26. Scores on the Social Phobia Scale were lower for males (M = 17.1, SD = 9.68) compared to females (M = 22.7, SD = 12.7), and this difference reached statistical significance, t (91) = 2.36, p  = .02.

The mean SPS score of the current sample was 20.0 ( SD  = 11.6, range = 2–48). Compared to McNeil et al.’s (1995) reference data, this is significantly lower than the mean SPS score of individuals with SAD, M  = 32.8, SD  = 14.8, t (57) = 5.86, p  < .001, but significantly higher than undergraduates, M  = 13.4, SD  = 9.6, t (144) = 3.69, p  < .001, and community volunteers, M  = 12.5, SD  = 11.5, t (141) = 3.70, p  < .001. The mean age of these comparison groups was higher (SAD sample M  = 36.5 years, community sample M  = 33.2 years, with age data not reported for undergraduates) than the current sample.

An exclusion criterion of previous acquaintance with the experimenters was implemented, as familiarity may have reduced the effectiveness of the social challenge tasks as anxiety inductions. A recruitment request was e-mailed to all students at Greenwich University which stated that “volunteers are sought to take part in a paid (£10) study which will involve filling in some questionnaires, engaging in a conversation task and talking to others about a set topic, giving your views”.

Anxiety and social behavior scales

Mattick and Clarke’s Social Phobia Scale (SPS) Footnote 1 was used to assess level of trait social anxiety. The SPS consists of 20 items rated on a five-point (0–4) scale, with higher scores indicating greater social anxiety. The scale has been shown to reliably assess social anxiety in both non-clinical and clinical populations [ 23 ]. The SPS has previously demonstrated good test-retest reliability, internal consistency and convergent validity [ 24 , 25 ] and exhibited high internal consistency (Cronbach’s α = .89) for the current data.

State anxiety was assessed in order to verify that the speech and interaction tasks resulted in increased anxiety relative to participants’ baseline anxiety. Baseline anxiety was assessed with a single self-report item that asked respondents to indicate their current anxiety on a scale of 1–10. State anxiety was also assessed immediately prior to the commencement of each task (participants had been provided with task details a few minutes earlier), and immediately after each task where participants were asked to rate the anxiety they had felt during the task itself. Single-item assessments of state anxiety have shown good reliability and convergent validity [ 26 ].

The Social Performance Rating Scale (SPRS) [ 27 ] was used to rate the participant on the following five dimensions: Gaze - adequacy of eye contact, Vocal Quality – warmth, clarity and enthusiasm demonstrated in verbal expression, Length – low level of monosyllabic speech/excessive talking, Discomfort – low levels of behavioral anxiety (e.g., fidgeting, trembling, postural tension), and Flow - verbal fluency (including the ability to incorporate information provided by the conversation partner smoothly into the interaction). The flow item was not used in the assessment of the speech task, as the rating descriptors for this component are specific to conversation. All SPRS items were rated on a 5-point scale and scored so that higher scores represented more effective social performance. Detailed descriptive anchors accompany each rating point to facilitate scoring; for example, Vocal Quality, “5 (Very Good) = Participant is warm and enthusiastic in verbal expression without sounding condescending or gushy”. The SPRS has shown excellent inter-rater reliability, internal consistency, convergent, discriminant and criterion validity [ 27 , 28 ]. Agreement across the three raters assessing the speech task was examined with an intraclass correlation (ICC). An absolute-agreement model was used [ 29 ], which is a stringent test requiring both high inter-rater correlations and minimal discrepancy in actual rating values to produce a high ICC. Analysis revealed ICC’s = .64–.86 for individual SPRS dimensions (all p’s < .001), suggesting good rater agreement [ 30 ]. Scores were therefore averaged across raters for each individual SPRS dimension for the speech task. Similar means (range: 3.4–3.8) and standard deviations (range: 0.7–1.1) were observed across SPRS components for both interaction and speech tasks.

Speech task

Participants were given 3 min to prepare a speech presenting a persuasive argument on their choice of one of the following topics: “sometimes it is ok to lie, discuss” or “can any crime be justified?”. Participants were told they would be presenting in front of a small audience and that they should try to keep going for 3 min although they could terminate the task at any point. Three confederates (one male and two female) comprised the “audience” for the speech task, with the same three-confederate audience used for each participant. The confederate audience had previously undertaken a number of trial sessions with several undergraduate volunteers acting as participants where they had practiced maintaining neutral facial expressions.

Interaction task

Participants were told that they would shortly be introduced to someone and that they would have 3 min to find out as much as they could about this person, although they could terminate the task at any time. The conversation partner was an experimental confederate, who was of the opposite-sex in order to maximize socially-evaluative challenge [ 6 ]. The same male confederate was used for each female participant, and the same female confederate was used for each male participant, with the one male and one female confederate taken from the pool of three confederates used in the speech task. Confederates had previously undertaken a number of trial sessions amongst each other and with undergraduate volunteers, where they practiced giving minimal responses, avoiding asking questions and maintaining neutral facial expressions [ 6 ]. Nobody other than the participant and the confederate was present during the interaction task when the experiment began.

To put participants in a relaxed state for a reliable assessment of baseline state anxiety, and to provide time for the experimenter to prepare the social challenge tasks, participants watched a 5-min relaxation video showing images of various seascapes accompanied by relaxing sounds. They then immediately completed the baseline state anxiety item along with the Social Phobia Scale and were randomized to undergo either the speech or interaction task first.

Participants were given details of the first social challenge task and reminded that they had the right to withdraw from the study at any point (no withdrawals occurred). Immediately prior to the social challenge task, participants completed the state anxiety item to assess anticipatory anxiety. Immediately following the task, participants again completed the state anxiety item, retrospectively indicating the anxiety they had experienced during the task. Participants were independently rated on their social performance by the audience of confederates (speech task) or the conversation partner (interaction task) using the SPRS, with ratings not disclosed to participants. This procedure was then repeated with the second social challenge task.

Statistical analysis plan

The association of social anxiety and sex with observer ratings was examined by conducting separate regression analyses on each SPRS dimension, with predictors of social anxiety, sex (− 1 = males, + 1 = females) and a Social Anxiety X Sex interaction term. Social anxiety was standardized but SPRS ratings were left unstandardized, so that the raw regression coefficient is interpreted as the mean change in rating points (on the 1–5 scale) following a one standard deviation increase in social anxiety. The interaction term was computed by cross-multiplication of sex and standardized social anxiety scores [ 31 ].

To determine whether regression coefficients of social anxiety and behavioral ratings differed significantly across the different SPRS dimensions, we tested the equality of these coefficients within a structural equation model. Predictors were the same as for the multiple regression analysis described above, and outcome variables were two SPRS dimensions (specified with correlated errors) whose coefficients were to be compared. We then imposed an equality constraint on the coefficient of social anxiety with each of two performance dimension coefficients. If a likelihood ratio test indicates a significant decrease in fit when an equality constraint is used, this indicates that the two coefficients are not equal [ 32 ]. Analyses were conducted in R using the lavaan [ 33 ] package .

Data screening

Regression residual plots for SPRS ratings revealed normality and homoscedasticity assumptions were met with no obvious outliers present. A negative skew of speech and interaction task times (due to a ceiling effect from the 3-min time limit) was observed, so p -values for analysis of task time data were computed from 10,000 bootstrapped samples.

Social challenge tasks: anxiety manipulation check

Consistent with the successful induction of anxiety, paired t-tests found significant increases from baseline anxiety for the speech task at pre-task ( t (92) =5.58, p  < .001) and during-task ( t (92) =9.92, p  < .001) periods, and for the interaction task at pre-task ( t (92) =5.84, p  < .001) and during-task periods ( t (92) =5.69, p  < .001) (see Table  1 for mean task anxiety scores at each assessment period). To check that anxiety was induced in both male and female participants, t-tests were repeated for each gender separately. For males, significant increases from baseline anxiety were uniformly found at pre-task ( t (44) =3.61, p  < .001) and during-task ( t (44) =5.63, p  < .001) in the speech task, and pre-task ( t (44) =2.52, p  = .015) and during-task ( t (44) =4.15, p  < .001) in the interaction task. This pattern of results was replicated for females, with significant increases from baseline anxiety observed at pre-task ( t (47) =4.49, p  < .001) and during-task ( t (47) =8.58, p  < .001) for the speech task, and pre-task ( t (47) =5.89, p  = .015) and during-task ( t (47) =4.03, p  < .001) for the interaction task.

Table 1 also reports correlations of social anxiety and gender with self-reported anxiety and shows social anxiety to be consistently moderately associated with increased anxiety response, and additionally that females generally reported greater anxiety compared to males.

Some participants terminated the social challenge tasks before the 3-min limit (speech M  = 127  s , interaction M  = 177  s ). As such, we computed the association between social anxiety and task time, as observers’ ratings might conceivably be affected by early task termination. No significant association was observed for either speech ( r  = −.02, p  = .88) or interaction ( r  = −.19, p  = .13) tasks.

Primary analysis

Separate regression analyses were performed on each SPRS dimension for the speech and interaction tasks resulting in 9 regression tests (4 SPRS speech dimensions, 5 SPRS interaction dimensions). To control type I error rate, we used an adjusted alpha criterion of α = .021 based on the Dubey-Armitage Parmar correction [ 34 ], which adjusts the conventional level of .05 based on the number of tests conducted (9) and the mean correlation between outcomes ( r  = .59 for SPRS ratings).

Speech task: social anxiety, sex and SPRS ratings

Table  2 shows the unstandardized ( B ) and standardized ( ß ) coefficients of social anxiety with observer ratings on each SPRS item resulting from the regression analysis of the speech task. These results show that social anxiety was a significant predictor of increased discomfort 2 ( B  = -0.28, ß  = -0.42 , p  < .001), but not of gaze, vocal quality or length. There were no significant sex (Table 3 ) or Social Anxiety X Sex interaction effects ( p  = .10–.96).

With respect to the magnitude of the association between social anxiety and SPRS discomfort, as SPRS ratings were left unstandardized, B represents the mean change in SPRS discomfort ratings on the 5-point scale for a one SD increase in social anxiety. As such, this indicates that a change from − 1 SD (low) to + 1 SD (high) social anxiety is associated with a 0.56-point increase in discomfort. Footnote 2

Interaction task: social anxiety, sex and SPRS ratings

For the interaction task, social anxiety was significantly associated with ratings on the discomfort dimension ( B  = -0.36, ß  = -.45, p  < .001), but not with other SPRS dimensions (Table 2 ). No significant sex (Table 3 ) or interaction effects ( p  = .09–.98) were observed. The unstandardized regression coefficient of B  = -0.36 for discomfort indicates that a change from − 1 SD (low) to + 1 SD (high) social anxiety is associated with a 0.72-point increase 2 in discomfort.

Comparison of regression coefficients of social anxiety across SPRS dimensions

A likelihood ratio test was used to compare the regression coefficient of social anxiety for SPRS discomfort with regression coefficients for the other SPRS dimensions. For the speech task, the coefficient for SPRS discomfort was significantly greater than all other SPRS dimensions (χ 2  = 6.56–17.65, all p ’s < .01). For the interaction task, the coefficient was significantly greater for SPRS discomfort compared to all other SPRS dimensions (χ 2  = 4.37–5.36, all p ’s < .05) except SPRS gaze (χ 2  = 1.31, p  = .25). Footnote 3

One of the primary findings from this study was that social anxiety was associated with higher observer ratings of behavioral discomfort (e.g., fidgeting, trembling, swallowing) during interaction and speech tasks, but not with other dimensions such as verbal fluency or quality of verbal expression.

Previous research investigating the link between social anxiety and social behavior has produced inconsistent results. It has been suggested that this inconsistency could be partially attributable to differences across studies in the dimension of social behavior assessed, with social anxiety potentially impairing only some behavioral dimensions; although no coherent pattern of which elements of social behavior may be affected has emerged [ 10 ]. The current results suggest that, at the non-clinical level at least, social anxiety may magnify the visible signs of anxiety but have little impact on other social behavior dimensions that were assessed here. These results are broadly consistent with Bögels et al. [ 19 ] who compared performance ratings for undergraduates low and high in social anxiety. They found that socially anxious participants received significantly more negative ratings on a “showing anxiety symptoms” factor, but not on a “skilled behavior” factor. Similarly, Cartwright-Hatton et al. [ 14 ] found that social anxiety scores were significantly associated with observer ratings of nervousness in schoolchildren based on a videotaped two-minute presentation, but not with “overall” impressions of performance (based on three items of ‘cleverness of speech’, friendliness and performance quality). It is difficult to determine from these previous studies if this is indicative of genuine selective effects on visible anxiety signs or simply chance variation, as no statistical comparison across dimensions was made. To our knowledge, the current study is the first to provide a statistical evaluation of these differences. The fact that social anxiety was significantly more strongly associated with behavioral discomfort than the vast majority of all other dimensions suggests that social anxiety in the non-clinical range is reliably associated with selective behavioral impairment and that this is confined to manifest and observable signs of discomfort.

It is important to note that not all previous studies are consistent with an effect of social anxiety confined only to overt signs of anxiety. Some studies have found poorer observer ratings of fluency and voice intonation during a speech [ 15 ] and vocal clarity and eye contact during a conversation task [ 16 ] for patients with SAD compared to controls. However, a tabulated summary of past research findings [ 10 ] seems to suggest that where the ‘performance’ aspects of social behavior are also affected, this generally appears to be in clinical samples. The most logical conclusion to draw from this is that high levels of social anxiety within the non-clinical range may primarily exacerbate visible anxiety signs with less impact on other performance aspects, but exhibit broader impairing effects at the clinical level; although it is important to point out this does not appear to have been systematically examined.

The link between social anxiety and discomfort ratings suggests that behavioral signs of anxiety are visible to others during social challenges. If those high in social anxiety engage in safety behaviors to mask their anxiety (e.g., attempting to disguise shaking) as evidence suggests [ 8 ], our findings indicate these may have limited effectiveness – at least within the range of social anxiety typically encountered in a non-clinical population. In terms of the magnitude of increased visible anxiety symptoms, those high in social anxiety (one standard deviation above the mean) were rated by observers as approximately half (speech task) to three-quarters (interaction task) of a point higher than those low in social anxiety (one standard deviation below the mean) on the five-point scale used. Determining whether this constitutes a “meaningful” difference is difficult, although the fact that this difference at least approaches a whole-point difference in the scale’s anchor-points (e.g., from “good” to “fair”) is suggestive of a meaningful discrepancy and one that can be demonstrably perceived by others. Overall, these findings clearly show that social anxiety is associated with observable effect on social behavior even in the non-clinical range. Given that a non-clinical sample represents the largest segment of the population, this indicates that social anxiety may have negative effects for a large number of individuals.

The fact that social anxiety failed to be associated with behavioral ratings other than for overt anxiety symptoms is perhaps surprising. Social anxiety scores were strongly correlated with increased anxiety response during social challenges, and the disruptive effect of state anxiety on working memory and the processing of external information including social cues is well supported both theoretically (e.g., via occupation of attentional resources) and empirically [ 8 , 35 ]. As such, aspects of social behavior expected to involve significant cognitive demands, such as the production of coherent and fluent verbal responses, would seem likely to be impaired. While the lack of association is perhaps unexpected, several possible explanations can be considered. First, the sheer frequency of anxious thoughts in the socially anxious during social challenges could lead to their automatization, so that they fail to consume significant attentional resources to cause cognitive interference [ 11 ]. Second, socially anxious individuals are more likely to employ socially facilitative coping strategies, such as overt expressions of enthusiasm or listening to others [ 9 ], and this may help compensate for any disruptive effects of anxiety and encourage more favourable impressions of overall social competence. Third, although social anxiety was associated with increased task anxiety for our non-clinical sample, the magnitude of anxiety response needed to produce significant impairment may only be apparent at the clinical level. It should be noted that these explanations for the pattern of effects observed are necessarily speculative and require empirical corroboration.

With respect to sex, while women reported greater anxiety during social challenges, no evidence was found that the link between social anxiety and behavior was more pronounced in females. One recent non-experimental study did report a negative association between social anxiety and self-assessment of social skill in females but not males [ 36 ]. The current results suggest that, if such a sex-specific effect on self-assessed social competence is reliable, this does not appear to translate to actual behaviour as rated by others. It is important to treat the lack of any sex-specific influence found here with caution, however, given that interaction effects typically require large sample sizes to detect small or even medium effects. Nevertheless, our findings do suggest that if any such sex-specific effect does exist, this effect is unlikely to be large.

Several limitations of the current study should be noted. First, we used a non-clinical sample, and even if social anxiety does operate on a continuum as is commonly believed [ 3 ], results may not generalize to clinical levels of social anxiety. Second, conclusions drawn on the link between social anxiety and social behavior are necessarily limited to the circumscribed set of parameters examined, i.e., molecular indicators of performance during brief social challenges. Findings cannot be automatically assumed to apply to other, perhaps less easily defined or quantifiable facets of performance [ 6 ] in more prolonged or situationally different social challenges. Similarly, we used relatively structured tasks with participants given clear instructions on what to do, with evidence suggesting that unstructured situations may cause greater difficulties for socially anxious people [ 18 ]. Third, we restricted our study to presentational and interactive scenarios and did not examine situations involving fears of being observed (e.g. eating or drinking) and our results may not generalize to these types of situations. Nevertheless, the tasks employed here are fairly indicative of those commonly encountered outside of the laboratory, with the behavioral indicators believed to represent important features of social competence [ 27 ].

Despite these limitations, the current findings have several implications. The fact that social anxiety appears to be most strongly linked to an increase in observable signs of anxiety suggests that techniques directed towards the management of overt anxiety symptoms for those high in social anxiety may be particularly effective for improving impressions of social competence in specific domains where this is likely to be important. Techniques that help the individual recognize their use of anxious behaviors (e.g., throat clearing, fidgeting) and practicing elimination of these in a safe environment [ 37 ] may be especially beneficial. Progressive muscle relaxation may also prove useful to reduce muscle rigidity and promote the appearance of a relaxed posture. If successful, these techniques may produce more successful outcomes in situations where reduced signs of anxiety might be considered favorable, such as job interviews or presentations. Such interventions might even contribute to a potential reduction in social anxiety. Specifically, one feature of cognitive models is that socially anxious people tend to excessively focus on and overestimate the occurrence of behavioural, cognitive and somatic responses (e.g. shaking and sweating), and this contributes to a negative mental image of how one appears to others during social encounters [ 38 ]. Controlling somatic symptoms which are one source of this attentional focus may promote more positive imagery of one’s projected social self, which has been shown to increase explicit self-esteem [ 39 ] and may act as a positive reinforcer of social encounters reducing safety behaviours such as avoidance. It is important to emphasise that we did not investigate such interventions within this study, so these interpretations are entirely speculative. Nevertheless, these processes do represent logical pathways for how techniques directed towards managing visible anxiety signs, that we found to be amplified in those with high social anxiety here, could be potentially beneficial. In addition, the fact that social anxiety was associated with increased observable discomfort in a non-clinical sample also suggests that such management techniques may have potentially widespread benefits to a large sector of the population vulnerable to anxiety in a range of commonly encountered and important social challenges. The apparent selective effect of social anxiety also underlines the need for future studies to include multidimensional assessments of social behavior to fully explicate the nature of the relationship between social anxiety and social behavior.

In conclusion, the current findings suggest that, the detrimental effects of social anxiety on social behavior within the non-clinical range may be confined to the exacerbation of observable, physical anxiety symptoms with little discernible impact on performance quality. These results underline the necessity of including multiple behavioral dimensions in additional studies and suggest that techniques directed towards the management of outwardly observable anxiety symptoms may be particularly beneficial for socially anxious individuals. Given the importance of everyday “performing” to successful social functioning, research should continue to examine how social anxiety impacts upon social behavior at both the clinical and non-clinical level.

We also administered Mattick and Clarke’s companion SIAS scale to provide psychometric data for a separate study. When we substituted the SPS with the SIAS in the current study, there was no impact on the pattern of results.

SPRS discomfort is scored such that lower ratings indicate poorer performance (i.e. greater discomfort).

We also reran these tests using only one SPRS outcome at a time. This was done as a consistency check to ensure that the results of the hypothesis testing in sections 3.4 and 3.5, which used a regression approach, were the same as those using an SEM approach. As expected, both techniques produced the same results (least squares and maximum likelihood estimators used in regression and SEM respectively produce identical estimates under the usual assumptions of regression).

Abbreviations

Intraclass Correlation

Social anxiety disorder

Standard Deviation

Social Interaction Anxiety Scale

Social Performance Rating Scale

Social Phobia Scale

Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62(6):593–602. https://doi.org/10.1001/archpsyc.62.6.593 .

Article   PubMed   Google Scholar  

Acarturk C, de Graaf R, van Straten A, Have MT, Cuijpers P. Social phobia and number of social fears, and their association with comorbidity, health-related quality of life and help seeking: a population-based study. Soc Psychiatry Psychiatr Epidemiol. 2008;43(4):273–9. https://doi.org/10.1007/s00127-008-0309-1 .

McNeil DR. Evolution of terminology and constructs in social anxiety and its disorders. In: Hofmann SG, DiBartolo PM, editors. Social anxiety: clinical, developmental, and social perspectives. London: Academic Press; 2014. p. 3–21.

Chapter   Google Scholar  

Fehm L, Beesdo K, Jacobi F, Fiedler A. Social anxiety disorder above and below the diagnostic threshold: prevalence, comorbidity and impairment in the general population. Soc Psychiatry Psychiatr Epidemiol. 2008;43(4):257–65. https://doi.org/10.1007/s00127-007-0299-4 .

Dijk C, van Emmerik AAP, Grasman RPPP. Social anxiety is related to dominance but not to affiliation as perceived by self and others: a real-life investigation into the psychobiological perspective on social anxiety. Personal Individ Differ. 2018;124:66–70. https://doi.org/10.1016/j.paid.2017.11.050 .

Article   Google Scholar  

Stravynski, A., Kyparissis, A., & Amado, D. (2010). Social phobia as a deficit in social skills. In S. G. Hofmann & P. M. DiBartolo (pp. 147-181). San Diego: Elsevier academic press. doi: https://doi.org/10.1037/t03750-000 .

Leary MR, Kowalski RM. Social anxiety. New York: Guilford Press; 1995.

Google Scholar  

Clark DM, Wells A. A cognitive model of social phobia. In: Heimberg RG, Liebowitz MR, Hope DA, editors. Social phobia: diagnosis, assessment and treatment. New York: Guilford Press; 1995. p. 69–93.

Leary MR. Social anxiousness: the construct and its measurement. J Pers Assess. 1983;47(1):66–75. https://doi.org/10.1207/s15327752jpa4701_8 .

Schneider BW, Turk CL. Examining the controversy surrounding social skills in social anxiety disorder: the state of the literature. In: Weeks JW, editor. The Wiley Blackwell handbook of social anxiety disorder; 2014. p. 366–87. https://doi.org/10.1002/9781118653920.ch17/summary .

Strahan E, Conger AJ. Social anxiety and its effects on performance and perception. J Anxiety Disord. 1998;12(4):293–305. https://doi.org/10.1016/S0887-6185(98)00016-4 .

Rapee RM, Lim L. Discrepancy between self- and observer ratings of performance in social phobics. J Abnorm Psychol. 1992;101(4):728–31. https://doi.org/10.1037/0021-843X.101.4.728 .

Clark JV, Arkowitz H. Social anxiety and self-evaluation of interpersonal performance. Psychol Rep. 1975;36(1):211–21. https://doi.org/10.2466/pr0.1975.36.1.211 .

Cartwright-Hatton S, Hodges L, Porter J. Social anxiety in childhood: the relationship with self and observer rated social skills. J Child Psychol Psychiatry. 2003;44(5):737–42. https://doi.org/10.1111/1469-7610.00159 .

Levitan MN, Falcone EM, Placido M, Krieger S, Pinheiro L, Crippa JA, et al. Public speaking in social phobia: a pilot study of self-ratings and observers’ ratings of social skills. J Clin Psychol. 2012;68(4):397–402. https://doi.org/10.1002/jclp.20868 .

Baker SR, Edelmann RJ. Is social phobia related to lack of social skills? Duration of skill-related behaviours and ratings of behavioural adequacy. Br J Clin Psychol. 2002;41:243–57. https://doi.org/10.1348/014466502760379118 .

Stopa L, Clark DM. Cognitive processes in social phobia. Behav Res Ther. 1993;31(3):255–67. https://doi.org/10.1016/0005-7967(93)90024-O .

Thompson S, Rapee RM. The effect of situational structure on the social performance of socially anxious and non-anxious participants. J Behav Ther Exp Psychiatry. 2002;33(2):91–102. https://doi.org/10.1016/S0005-7916(02)00021-6 .

Bögels SM, Rijsemus W, De Jong PJ. Self-focused attention and social anxiety: the effects of experimentally heightened self-awareness on fear, blushing, cognitions, and social skills. Cogn Ther Res. 2002;26(4):461–72. https://doi.org/10.1023/A:1016275700203 .

Gelman A, Stern H. The difference between “significant” and “not significant” is not itself statistically significant. Am Stat. 2006;60(4):328–31. https://doi.org/10.1198/000313006X152649 .

Norton PJ. Social anxiety and withdrawal. In: Nangle DW, Hansen DJ, Erdley CA, Norton PJ, editors. Practitioner’s guide to empirically based measures of social skills. NY: Springer; 2010. p. 167–78. https://doi.org/10.1007/978-1-4419-0609-0 .

McLean CP, Anderson ER. Brave men and timid women? A review of the gender differences in fear and anxiety. Clin Psychol Rev. 2009;29:496–505. https://doi.org/10.1016/j.cpr.2009.05.003 .

McNeil DW, Ries BJ, Turk CL. Behavioral assessment: self-report, physiology, and overt behavior. In: Heimberg RG, Liebowitz MR, Hope DA, Schneier FR, editors. Social phobia: diagnosis, assessment, and treatment. New York: Guilford Press; 1995. p. 202–31.

Osman A, Gutierrez PM, Barrios FX, Kopper BA, Chiros CE. The social phobia and social interaction anxiety scales: evaluation of psychometric properties. J Psychopathol Behav Assess. 1998;20:249–64. https://doi.org/10.1023/A:1023067302227 .

Mattick RP, Clarke JC. Development and validation of measures of social phobia scrutiny fear and social interaction anxiety. Behav Res Ther. 1998;36:455–70. https://doi.org/10.1016/S0005-7967(97)10031-6 .

Davey HM, Barratt AL, Butow PN, Deeks JJ. A one-item question with a Likert or visual analog scale adequately measured current anxiety. J Clin Epidemiol. 2007;60(4):356–60. https://doi.org/10.1016/j.jclinepi.2006.07.015 .

Fydrich T, Chambless DL, Perry KJ, Buergener F, Beazley MB. Behavioral assessment of social performance: a rating system for social phobia. Behav Res Ther. 1998;36:995–1010. https://doi.org/10.1016/S0005-7967(98)00069-2 .

Harb GC, Eng W, Zaider T, Heimberg RG. Behavioral assessment of public-speaking anxiety using a modified version of the social performance rating scale. Behav Res Ther. 2003;41(11):1373–80. https://doi.org/10.1016/S0005-7967(03)00158-X .

McGraw KO, Wong SP. Forming inferences about some intraclass correlation coefficients. Psychol Methods. 1996;1(1):30–46. https://doi.org/10.1037/1082-989X.1.1.30 .

Cicchetti DV. Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychol Assess. 1994;6(4):284–90. https://doi.org/10.1037/1040-3590.6.4.284 .

Aiken LS, West SG. Multiple regression: testing and interpreting interactions. London: Sage; 1991.

Kwan JL, Chan W. Comparing standardized coefficients in structural equation modeling: a model reparameterization approach. Behav Res Methods. 2011;43(3):730–45. https://doi.org/10.3758/s13428-011-0088-6 .

Rosseel Y. Lavaan: an R package for structural equation modeling. J Stat Softw. 2012;48(2):1–36. https://doi.org/10.18637/jss.v048.i02 .

Sankoh AJ, Huque MF, Dubey SD. Some comments on frequently used multiple endpoint adjustment methods in clinical trials. Stat Med. 1997;16(22):2529–42. https://doi.org/10.1002/(SICI)1097-0258(19971130)16:22<2529::AID-SIM692>3.0.CO;2-J .

Amir N, Bomyea J. Cognitive biases in social anxiety disorder. In: Hofmann SG, DiBartolo PM, editors. Social anxiety: clinical, developmental, and social perspectives. London: Academic Press; 2010.

Al-Ali MM, Singh AP, Smekal V. Social anxiety in relation to social skills, aggression, and stress among male and female commercial institute student. Education. 2011;132(2):351–61.

Wells A, Clark DM, Salkovskis P, Ludgate J, Hackmann A, Gelder M. Social phobia: the role of in-situation safety behaviors in maintaining anxiety and negative beliefs. Behav Ther. 1995;26(1):153–61. https://doi.org/10.1016/S0005-7894(05)80088-7 .

Rapee RM, Heimberg RG. A cognitive-behavioral model of anxiety in social phobia. Behav Res Ther. 1997;35(8):741–56.

Hulme N, Hirsch C, Stopa L. Images of the self and self-esteem: do positive self-images improve self-esteem in social anxiety. Cogn Behav Ther. 2012;41(2):163–73. https://doi.org/10.1080/16506073.2012.664557 .

Article   PubMed   PubMed Central   Google Scholar  

Download references

Acknowledgements

Our grateful appreciation goes to Marta Kaminska for help with data collection and for acting as an experimental confederate.

This work was supported by an internal grant awarded to the first author by the University of Greenwich. The funders had no role in any aspect of the study design, data collection, analysis or data or writing of the manuscript.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Author information

Authors and affiliations.

Department of Psychology, School of Health and Social Care, University of Greenwich, London, SE9 2UG, UK

Trevor Thompson

Faculty of Engineering, Dyson School of Design Engineering, Imperial College, London, SW7 2AZ, UK

Nejra Van Zalk

Start2Stop Addictions Treatment Centre, London, SW7 3HG, UK

Christopher Marshall

University of Greenwich, London, SE9 2UG, UK

Melanie Sargeant

Institute of Psychiatry, Psychology and Neuroscience, King’s College London, De Crespigny Park, London, SE5 8AF, UK

Brendon Stubbs

You can also search for this author in PubMed   Google Scholar

Contributions

TT was responsible for the study conceptualization, data analysis and writing of the manuscript. CM and MS were responsible for data collection and some writing contribution. NVZ and BS provided critical revision of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Trevor Thompson .

Ethics declarations

Ethics approval and consent to participate.

Ethical approval for the study was granted by the University Research Ethics Committee at the University of Greenwich and all procedures performed were in accordance with the 1964 Helsinki declaration. Written informed consent was obtained from all individual participants.

Consent for publication

Not applicable.

Competing interests

All authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License ( http://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Cite this article.

Thompson, T., Van Zalk, N., Marshall, C. et al. Social anxiety increases visible anxiety signs during social encounters but does not impair performance. BMC Psychol 7 , 24 (2019). https://doi.org/10.1186/s40359-019-0300-5

Download citation

Received : 11 December 2018

Accepted : 08 April 2019

Published : 23 April 2019

DOI : https://doi.org/10.1186/s40359-019-0300-5

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Social anxiety
• Social performance
• Social discomfort
• Sex differences

BMC Psychology

ISSN: 2050-7283

• General enquiries: [email protected]

RECENT ADVANCES IN SOCIAL ANXIETY RESEARCH

November 14, 2023

Dear Colleagues,

The National Social Anxiety Center (NSAC) provides information about relevant and current research in service of disseminating and promoting evidence-based treatment. This month’s summary is written by Annika Okamoto, PhD, A-CBT, representing NSAC Santa Barbara , and examines the 2023 article by Wolitzky-Taylor & LeBeau: Recent advances in the understanding and psychological treatment of social anxiety disorder.

The Wolitzky-Taylor and LeBeau’s article provides an overview of the current understanding and psychological treatment of social anxiety disorder (SAD). Key takeaways for clinicians include:

How is SAD best treated?

• The gold standard psychotherapy continues to be CBT in individual or group settings, but comparison studies of ACT and CBT show that they have comparable effects.
• Pharmacological interventions, particularly SSRIs, may also be helpful.

What other factors that coincide with SAD could be addressed in therapy to improve outcomes?

• Prior social trauma that involves humiliation and rejection,
• Envy (individuals with social anxiety experience higher levels of envy that predicts anxiety),
• Uncontrollable thoughts, rumination and perseverative cognition,
• Inattention and being more likely to feel distracted,
• Difficulties with inferring others’ emotional states and feelings, but not beliefs and intentions,
• Deriving less pleasure from social interactions (but still more than from nonsocial situations),
• Poor recall of positive social outcomes,
• Behavioral avoidance and safety behaviors,
• Intolerance of uncertainty and
• Delays in shifting attention away from social threats.

What do we know about SAD and other disorders?

• SAD contributes to worse educational performance across one’s life span.
• Later onset is associated with higher comorbidity and diminished quality of life.
• SAD symptom severity is correlated to depression and cannabis use disorder as adults.
• There is a direct relationship between the severity of negative psychotic symptoms and ideas of reference, and social anxiety. Meanwhile, the SAD in psychosis may be overlooked as less salient.
• Non-assertiveness seems to be a shared core for SAD and some other disorders, such as avoidant personality disorder.
• SAD research supports the interpersonal theory of suicide that posits that perceived burdensomeness and thwarted belongingness predict acute suicidal ideation (SI). In individuals with SAD, these factors predict acute SI better than effects of depression, and clinicians should evaluate them as risk factors.

What’s new in the assessment of SAD?

• There are two new measures that address the gaps left by other measures: 1) The Socially Anxious Rumination Questionnaire (SARQ) assessing rumination, and 2) the Ryerson Social Anxiety Scale (RSAS) measuring distress and impairment level.
• Machine-learning algorithms have shown promise in differential diagnosis of social difficulties such as SAD, autism spectrum disorder, and prodromal psychosis.

What does research evidence show about the treatment of SAD?

• “Does changing cognitions change behavior (traditional model), or does changing behavior change cognitions?” Research evidence points to behavior changing cognition. For example, exposure therapy works even without a focus on cognitive strategies.
• Mindfulness interventions and attention bias modification do not improve treatment outcomes.
• Decreases in social cost estimates during exposure are associated with better social anxiety outcomes. Targeting shame may contribute to progress.
• Self-guided digital therapies for SAD work but roughly 2/3rds drop out. People with SAD need individualized engagement strategies.
• Telehealth-based exposures can be as effective as in vivo exposures for public speaking.
• For people with SAD who use substances, clinicians should consider integrated treatments (i.e., exposure therapy + motivational interviewing); just targeting SAD alone is less effective.
• Changes in social anxiety symptoms may lead to improvement in depression if the depression symptoms are not too severe.
• People who have significant depression and SAD improve from SAD treatment more than non-depressed individuals. Thus, comorbidity does not equal treatment resistance.

Question for clinicians: Behavior impacts cognition and mood. What would your patient do differently if they were not anxious? Could you get their buy-in for a few weeks to do these things on a trial basis, to see how it affects their thoughts about themselves and the way they feel?

Wolitzky-Taylor, Kate, and LeBeau, Richard. Recent advances in the understanding and psychological treatment of social anxiety disorder . Faculty reviews , vol. 12, issue 8, April 2023. __

Annika Okamoto, PhD, A-CBT Representing NSAC Santa Barbara ( California Counseling Clinics )

This paper is in the following e-collection/theme issue:

Published on 19.4.2024 in Vol 11 (2024)

This is a member publication of National University of Singapore

Testing the Efficacy of a Brief, Self-Guided Mindfulness Ecological Momentary Intervention on Emotion Regulation and Self-Compassion in Social Anxiety Disorder: Randomized Controlled Trial

Authors of this article:

Original Paper

• Nur Hani Zainal 1, 2 , BS, MS, PhD   ; 
• Hui Han Tan 1 , BS, MS   ; 
• Ryan Yee Shiun Hong 1 , BS, MS, PhD   ; 
• Michelle Gayle Newman 3 , BS, MA, PhD  

1 Department of Psychology, National University of Singapore, Singapore, Singapore

2 Department of Health Policy, Harvard Medical School, Boston, MA, United States

3 Department of Psychology, The Pennsylvania State University, University Park, PA, United States

Corresponding Author:

Nur Hani Zainal, BS, MS, PhD

Department of Psychology

National University of Singapore

9 Arts Link

Singapore, 117572

Phone: 65 917 767 7088

Fax:65 6516 3749

Email: [email protected]

Background: Theories propose that brief, mobile, self-guided mindfulness ecological momentary interventions (MEMIs) could enhance emotion regulation (ER) and self-compassion. Such changes are posited to be mechanisms of change. However, rigorous tests of these theories have not been conducted.

Objective: In this assessor-blinded, parallel-group randomized controlled trial, we aimed to test these theories in social anxiety disorder (SAD).

Methods: Participants with SAD (defined as having a prerandomization cut-off score ≥20 on the Social Phobia Inventory self-report) were randomized to a 14-day fully self-guided MEMI (96/191, 50.3%) or self-monitoring app (95/191, 49.7%) arm. They completed web-based self-reports of 6 clinical outcome measures at prerandomization, 15-day postintervention (administered the day after the intervention ended), and 1-month follow-up time points. ER and self-compassion were assessed at preintervention and 7-day midintervention time points. Multilevel modeling determined the efficacy of MEMI on ER and self-compassion domains from pretrial to midintervention time points. Bootstrapped parallel multilevel mediation analysis examined the mediating role of pretrial to midintervention ER and self-compassion domains on the efficacy of MEMI on 6 clinical outcomes.

Results: Participants demonstrated strong compliance, with 78% (149/191) engaging in at least 80% of the MEMI and self-monitoring prompts. MEMI was more efficacious than the self-monitoring app in decreasing ER goal–directed behavior difficulties (between-group Cohen d =−0.24) and lack of emotional clarity (Cohen d =0.16) and increasing self-compassion social connectedness (Cohen d =0.19), nonidentification with emotions (Cohen d =0.16), and self-kindness (Cohen d =0.19) from pretrial to midintervention time points. The within-group effect sizes from pretrial to midintervention were larger in the MEMI arm than in the self-monitoring app arm (ER goal–directed behavior difficulties: Cohen d =−0.73 vs −0.29, lack of emotional clarity: Cohen d =−0.39 vs −0.21, self-compassion domains of social connectedness: Cohen d =0.45 vs 0.19, nonidentification with emotions: Cohen d =0.63 vs 0.48, and self-kindness: Cohen d =0.36 vs 0.10). Self-monitoring, but not MEMI, alleviated ER emotional awareness issues (between-group Cohen d =0.11 and within-group: Cohen d =−0.29 vs −0.13) and reduced self-compassion acknowledging shared human struggles (between-group Cohen d =0.26 and within-group: Cohen d =−0.23 vs 0.13). No ER and self-compassion domains were mediators of the effect of MEMI on SAD symptoms ( P =.07-<.99), generalized anxiety symptoms ( P =.16-.98), depression severity ( P =.20-.94), repetitive negative thinking ( P =.12-.96), and trait mindfulness ( P =.18-.99) from pretrial to postintervention time points. Similar nonsignificant mediation effects emerged for all of these clinical outcomes from pretrial to 1-month follow-up time points ( P =.11-.98).

Conclusions: Brief, fully self-guided, mobile MEMIs efficaciously increased specific self-compassion domains and decreased ER difficulties associated with goal pursuit and clarity of emotions from pretrial to midintervention time points. Higher-intensity MEMIs may be required to pinpoint the specific change mechanisms in ER and self-compassion domains of SAD.

Trial Registration: Open Science Framework (OSF) Registries; osf.io/m3kxz https://osf.io/m3kxz

Introduction

Emotion regulation (ER) has been defined as a deliberate action taken to modulate emotional states [ 1 ]. Situations can trigger emotional reactions, including short- and long-term emotional responses [ 2 ]. Persistent suboptimal behaviors that limit opportunities and quality of life are inherent to various emotional disorders, such as avoidance of feared situations in social anxiety disorder (SAD) [ 3 ], brooding in depression [ 4 ], and impulsivity in attention-deficit/hyperactivity disorder [ 5 ]. Elevating self-compassion could facilitate better ER by fostering discomfort tolerance [ 6 ]. Self-compassion is described as accepting one’s experience without judgment and showing oneself kindness [ 7 - 9 ]. Increased self-judgment might create anxiety and depression via emotional dysregulation. Such anxiety and depression could lead to avoidance as a way to manage negative affect [ 10 ]. Chronic avoidance of social situations and failure to optimally face challenges head-on can hinder one from building self-efficacy; decrease pleasure derived from recreational activities; and restrict work, school, and other life opportunities [ 11 - 13 ]. Therefore, developing efficacious interventions to improve ER and self-compassion is essential.

Mindfulness-based interventions (MBIs) might enhance ER and self-compassion simultaneously. The monitoring and acceptance theory proposed that MBIs could boost ER by increasing acceptance, curiosity, and equanimity, coinciding with better interpersonal functioning [ 14 ]. This theory suggests that enhancing various outcomes in MBIs relies on integrating attention monitoring and acceptance skills [ 15 ]. It proposes that acceptance functions as an ER skill that promotes nonreactivity, self-compassion, and receptiveness to current experiences. The omission of acceptance skills training could potentially negate the stress-buffering and social cognitive advantages of MBIs. Consistent with the monitoring and acceptance theory, meta-analyses have shown inverse relations among self-compassion and anxiety, depressive symptoms, and stress [ 16 ] and found that MBIs were superior to controls in improving mental health symptoms and self-compassion [ 17 ]. However, traditional in-person MBIs, such as mindfulness-based stress reduction [ 18 ] and mindfulness-based cognitive therapy [ 19 ], typically involve significant financial costs, time commitments, and potential travel requirements, often extending from 1 hour to 2.5 hours per week across 8 to 16 sessions with 6-hour retreat sessions [ 20 ]. Individuals with high levels of emotional dysregulation and self-compassion deficits tend to self-report shame, stigma, and other hindrances to treatment seeking [ 21 , 22 ]. Those experiencing these struggles may not seek in-person therapy or may not remain in treatment, fearing the negative emotions associated with the treatment process. Collectively, such research emphasizes the importance of developing scalable (eg, app based) and concise MBIs.

Concise and scalable MBIs, typically lasting up to 2 weeks [ 23 ], may enhance ER and self-compassion. For instance, enhanced ER skills and reduced stress levels were observed among smokers and nonsmokers after a 5-day MBI, compared to ER and stress observed before the intervention [ 24 ]. Similarly, a single-session MBI was associated with increased cognitive reappraisal and reduced emotional suppression (ER domains) among graduate students over 3 months [ 25 ]. However, the absence of a control arm in these studies precluded knowing whether the brief MBI would outperform an active control, an essential aspect for drawing causal inferences [ 26 ]. Another study, a 14-day MBI randomized controlled trial (RCT) [ 27 ], implied this possibility by showing that acceptance was critical in effectively reducing loneliness and enhancing social engagement in everyday experiences. Another RCT showed that a single-session MBI was linked to better ER in trauma-distressed people [ 28 ]. Relatedly, 5-session MBIs consistently outperformed controls, raising self-compassion and mindfulness among undergraduates [ 29 , 30 ]. Together, these studies suggested that brief mindfulness ecological momentary interventions (MEMIs) that repeatedly instructed mindfulness exercises and recorded symptom levels in real time could outperform an active control in enhancing ER and self-compassion over time in clinical samples.

Moreover, improved ER and self-regulation may be a theoretical change mechanism explaining why brief, fully self-guided mobile MEMIs conferred mental health benefits such as symptom reduction over time [ 31 , 32 ]. Such theories posit that brief, fully self-guided mobile MEMIs may enhance the capacity to observe internal reactions in emotionally charged situations, enabling individuals to recognize when they are caught in their emotions, to pause to regain composure before reacting, and to practice self-kindness. Indeed, 2 recent experiments of diverse mindfulness practices showed that state-level and trait-level self-compassion improvements were linked to increased self-guided mindfulness practices and quality of life across 14 days in nonclinical samples [ 33 ]. Relatedly, baseline higher trait observing and describing mindfulness facets predicted reduced anxiety and depression via enhanced ER across an internet-delivered 8-week MBI in healthy adults [ 34 ]. This finding might extend to brief, fully self-guided mobile MEMIs for clinical samples. The dearth of mediation analyses testing the mechanistic role of ER and self-compassion highlights the importance of conducting such studies, particularly among individuals with emotional disorders. Discovering treatment mediators (proxy change mechanisms) might aid with fine-tuning and refining existing brief, scalable MEMIs for clinical samples [ 35 ].

SAD presents as one potential clinical sample case in point. Fostering better ER and self-compassion as treatment targets via brief, fully self-guided, mobile MEMIs could be especially advantageous for individuals with SAD, as they tend to exhibit poor ER, such as excessive avoidance and difficulties applying cognitive modification techniques [ 36 ]. Teaching nonjudgment and nonreactivity skills in SAD via the MEMI might pave the way for fostering cognitive modification and other ER skills and promote active efforts to increase exposure to various interpersonal situations. Furthermore, self-compassion has shown inverse associations with overall SAD symptoms and specific cognitive processes, such as the fear of both negative and positive evaluation [ 37 ] and postevent ruminative processing [ 38 ]. Those with versus without SAD felt it was essential to control their emotions, did not believe in emotion malleability [ 39 ], and had trouble practicing acceptance of emotional responses [ 40 ]. Therefore, MEMIs instructing present-moment focus and acceptance of myriad emotional experiences could remedy these psychological rigidities by enhancing ER skills, thereby improving a wide array of responses to fear-inducing social situations. Construed clinically, these studies suggest that individuals with SAD have room for growth in terms of benefiting from brief, fully self-guided, mobile MEMIs to reduce various ER difficulties and enhance diverse self-compassion domains across time.

Such ER difficulties could pertain to nonacceptance of emotions (proclivity to exhibit nonaccepting responses to one’s distress) and emotional awareness issues (the inclination to focus upon and recognize emotions within oneself). They might also be linked to a lack of emotional clarity (lucidity regarding present-moment emotions) and goal-directed behavior difficulties (maintaining focus and achieving tasks during episodes of negative emotional states). Furthermore, ER difficulties could encompass impulse control issues (inability to sustain behavioral self-regulation amid negative emotional states) and strategy use problems (lack of conviction for having the ability to effectively regulate emotions when feeling upset [ 41 ]). The facets of self-compassion might also pertain to acknowledging shared human struggles (acknowledgment of collective human experience, recognizing that all individuals encounter failures, make errors, and navigate imperfect life trajectories) and social connectedness (a sense of being linked socially with other humans). Cultivating self-compassion also embraces mindfulness (a state of cognizant immersion in one’s immediate experience, characterized by clarity and equilibrium) and nonidentification with emotions (absence or lack of tendency to become ensnared in an exaggerated narrative about the adverse facets of one’s self or life experiences). In addition, self-compassion includes nonjudgment toward oneself (absence or lack of tendency to be overly self-critical) and self-kindness (practice of approaching oneself with support and understanding) [ 42 ].

On the basis of the theory and logic outlined, the aims of this study were 2-fold. First, we hypothesized that a 14-day MEMI would be superior to a self-monitoring control condition in reducing various domains of ER difficulties and enhancing self-compassion from pretrial to midintervention time points across 7 days (hypothesis 1). Specifically, we examined the 6 difficulties in ER and self-compassion domains mentioned above as pretrial to midintervention outcomes. Second, we hypothesized that the effect of the 14-day MEMI versus self-monitoring app on change in SAD-related outcomes from pretrial to postintervention time points (pre-post; hypothesis 2a) or prerandomization to 1-month follow-up (1MFU) time points (pre-1MFU; hypothesis 2b) would be mediated via a change in ER difficulties and self-compassion domains assessed from pretrial to midintervention time points. Specifically, the outcomes examined were SAD symptoms, generalized anxiety symptoms, depression severity, repetitive negative thinking, and trait mindfulness. Our study was an extension of a primary RCT, which showed that brief MEMI and self-monitoring app led to sustained changes in all of these clinical outcomes, with small-to-large effect sizes from pre-post and pre-1MFU time points. There were no differences between MEMI and self-monitoring on the main outcome changes [ 43 ].

Ethical Considerations

All study procedures were approved by the National University of Singapore (NUS) before participant recruitment and all participants provided informed consent (institutional review board #S-20-025). This RCT was preregistered on Open Science Framework [ 44 ]. All data were de-identified. Participants were reimbursed up to $30, 8 subject pool hour credits, or both, pro-rated based on their degree of participation.

Study Design

In this assessor-blinded RCT, we randomized individuals into 1 of 2 arms with a parallel design and 1:1 allocation ratio. Randomization was stratified according to age and sex. We used a mixed design of 2 groups (group: MEMI vs self-monitoring) by 3 (prerandomization, postintervention, and 1MFU postrandomization) time points to evaluate the efficacy of the 14-day MEMI compared to self-monitoring on ER difficulties and self-compassion domain outcomes. Random treatment assignment to the MEMI and self-monitoring arms was the between-participant factor, whereas time points served as the within-participant factor. The trial was advertised as a “digital mindfulness intervention study” by emailing NUS students via a listserve, posting advertisements across the campus and NUS-affiliated mental health clinics, and permitting recruitment from both the student body and the broader community.

Eligibility Criteria

Details of the study methods can be found in Multimedia Appendix 1 and in an earlier report by Zainal et al [ 43 ]. Eligible participants were required to self-report SAD with a Social Phobia Inventory (SPIN) score ≥20 [ 45 ], be aged at least 18 years, own a smartphone, and seek help for mental health issues. We recruited treatment-seeking individuals from the psychology participant pool and the local community, excluding those with self-reported suicidal ideation, mania, or psychosis. Eligible participants were recruited to this web-based trial on campus (before the COVID-19 pandemic) and on the web (during the pandemic) between September 1, 2019, and May 31, 2021.

Several reasons prompted us to choose the SPIN measure to screen for participants with probable SAD. Unlike the Social Phobia Diagnostic Questionnaire (SPDQ), the SPIN was already integrated as part of a larger battery of screening assessments in a busy psychological clinic and an undergraduate psychology research participant pool at NUS. This clinical assessment battery was based on a series of Diagnostic and Statistical Manual (DSM), Fifth Edition –text revised web-based assessment measures made accessible and recommended by the American Psychiatric Association [ 46 ]. Although the SPDQ has superior psychometric properties to diagnose SAD, given its structural concordance with the DSM , the briefer SPIN was the more pragmatic and operationally efficient choice for this study. Moreover, both the SPIN and SPDQ baseline scores were highly correlated ( r =0.89; P <.001). Furthermore, all eligible participants met criteria for probable SAD at baseline using the recommended SPDQ cut-off score of ≥7.38 that had optimal sensitivity (82%), specificity (85%), a positive predictive value (83%), and a negative predictive value (83%) with a clinical diagnosis of SAD with the Anxiety Disorder Interview Schedule-IV [ 47 , 48 ].

Participants

All participants provided voluntary informed consent. We randomized 191 participants into 2 groups: MEMI (96/191, 50.3%) and self-monitoring (95/191, 49.7%). Their average age was 21.84 (SD 3.37; range 18-53) years. Of the 191 participants, 41 (21.5%) identified as male, 149 (78%) as female, and 1 (1%) as other; 165 (86.4%) identified as Chinese and the remaining 26 (13.6%) identified as Indian, Malay, and other ethnicities; 167 (87.4%) were categorized as never married, whereas 24 (12.6%) were married, living with a partner, or in an intimate relationship but not living together; 145 (75.9%) had completed junior college as their highest level of education, whereas 46 (24.1%) held diplomas, university degrees, or graduate degrees; 139 (72.8%) were not employed, whereas 52 (27.2%) were engaged in part-time or full-time work; 178 (93.2%) were full-time students, whereas the remaining 13 (6.8%) were part-time students or nonstudents; 172 (90%) reported an annual income within the range of US $0 to US $7500, whereas 19 (9.9%) fell into higher income brackets; 11 (5.8%) had previously received a clinical diagnosis of anxiety or depressive disorder; and 10 (5.2%) were currently using psychotropic drugs (Table S1 in Multimedia Appendix 1 ). Eligible participants had a mean SPIN score of 35.65 (SD 13.18; range 20-67), with 31.9% (61/191) in the mild severity (score of 20-30), 31.9% (61/191) in the moderate severity (score of 31-40), 21.9% (42/191) in the severe (score of 41-50), and 14.1% (27/191) in the very severe (score of ≥51) categories [ 45 , 49 , 50 ]. Figure 1 presents the CONSORT-EHEALTH (Consolidated Standards of Reporting Trials of Electronic and Mobile Health Applications and Online Telehealth) [ 51 ] diagram ( Multimedia Appendix 2 ).

The MEMI and self-monitoring arms were codeveloped by the lead author (NHZ) and senior author (MGN; both PhD-level psychologists). The study procedures were tested with various research assistants who helped provide feedback and troubleshoot technical issues and were refined from August 1, 2018, to May 31, 2019. Weekly meetings were held during this period to optimize the study procedures and reach a consensus regarding ambiguous procedural aspects. Therefore, the team did not encounter technical glitches with the Personal Analytics Companion app [ 52 ] used to deliver the MEMI and self-monitoring app during the subsequent data collection phase.

At baseline, eligible participants completed a counterbalanced series of web-based self-report measures. Subsequently, we randomly assigned participants to either the MEMI or self-monitoring arm using the Excel (Microsoft Corp) randomization function integrated into Qualtrics (Qualtrics International Inc). The mindfulness or self-monitoring video was provided toward the end of the baseline visit after the completion of all pretrial assessments. Then, participants installed the Personal Analytics Companion app [ 52 ], which contained either the MEMI or self-monitoring tool, on their smartphones, with the experimenter demonstrating its features. Assessors (also called experimenters herein) were blinded to group assignment. Participants were told that they would receive an intervention and experimenters could not know if they were randomly assigned to the MEMI or self-monitoring arm. The owner of the Personal Analytics Companion software had no role in the development of the MEMI and self-monitoring app. Participants were informed that they would receive prompts at 5 different times each day (around 9 AM, noon, 3 PM, 6 PM, and 9 PM) during the following 14-day period. These prompts could be edited within an approximately 2-hour window (ie, 8 AM-10 AM, 11 AM-1 PM, 2 PM-4 PM, 5 PM-7 PM, and 8 PM-10 PM) to fit each participant’s schedule and did not substantially differ across participants. To ensure the validity of responses, participants were instructed to provide input on their current state of depression, anxiety, and mindfulness within 2 hours of receiving the MEMI or self-monitoring prompt. These prompts guided the participants in 1-minute mindfulness or self-monitoring activities based on their assigned group. The apps did not require the training of Bachelor of Arts—level coaches and were entirely self-guided. Following the 14-day treatment phase, participants received emails prompting them to complete the self-reported clinical outcome measures at postintervention and 1MFU time points.

RCT Arms of MEMI and Self-Monitoring

Participants assigned to the MEMI app viewed a standardized video presentation led by the principal investigator that described the evidence-based MBI protocol akin to mindfulness-based stress reduction [ 18 , 53 ]. This presentation familiarized MEMI participants with mindfulness, encouraging them to fully immerse themselves in their present moment and equipping them with skills encompassing open monitoring and attentiveness to present transient experiences. Proficiency in present-moment awareness may increase the richness of experiences, thereby amplifying emotional responsiveness [ 14 ]. Next, the video therapist skillfully demonstrated the paced, rhythmic diaphragmatic breathing technique and guided participants in its practice. Subsequently, the video therapist continued by imparting lessons on nonjudgmental acceptance (ie, allowing emotions to fluctuate and experiences to unfold without deliberately changing them), drawing from the mindfulness-based cognitive therapy principles [ 19 ]. These lessons fostered mindfulness-related attributes such as observation, nonreactivity, and nonjudgmental acceptance. Subsequently, each MEMI participant received a comprehensive explanation regarding the importance and benefits of consistent mindfulness practice.

The MEMI app prompted participants to practice mindfulness 5 times daily (at approximately 9 AM, noon, 3 PM, 6 PM, and 9 PM) for 14 days. During each prompt, MEMI participants received the following instructions:

Pay attention to your breathing. Breathe in a slow, steady, and rhythmic manner. Stay focused on the sensations of the air coming into your lungs and then letting it out. As you’re breathing, observe your experience as it is. Let go of judgments that do not serve you. Focus on the here and now. Attend to the small moments right now (eg, reading a chapter, having a cool glass of water), as that is where enjoyment, peace, and serenity in life happen.

Participants rated their state-level (ie, momentary) depression (“To what degree do you feel depressed right now?”), anxiety (“To what degree do you feel keyed up or on edge right now?”), and mindfulness (“To what extent are you experiencing the present moment fully?”) levels on a 9-point Likert scale (1= not at all to 9= extremely ) before and after receiving these instructions. Each MEMI prompt ended with encouragement to inculcate these skills in the long term as follows:

Remember that the cultivation of mindfulness is lifelong. The goal of therapy is to be your own therapist. Practice mindfulness between the prompts and after you have completed this study.

These MEMI techniques were proposed to work for SAD in the following ways. First, focused attention to present-moment activities would break the habit of ruminating on the past unproductively, such as brooding over social events in self-critical ways often observed in SAD. Furthermore, this activity was designed to reduce pathological worry about the future, which is a common comorbid symptom of SAD. Second, open monitoring skills helped participants flexibly experience positive and negative emotions without resistance and cultivate stronger discomfort tolerance. Third, mindful diaphragmatic breathing might work therapeutically by inducing relaxation, improving heart rate variability, and lowering blood pressure [ 54 ]. Fourth, acceptance exercises might enhance the capacity to tolerate and manage emotional states, specifically addressing affect intolerance or sensitivity, an etiological and maintenance factor of SAD, and other psychopathology [ 55 ]. Unlike cognitive behavioral therapy (CBT) apps, the MEMI app does not work by continually instructing exposure therapy designed to encourage immersion in fear-inducing social situations to help persons with SAD cope with their anxiety and enhance self-efficacy in these contexts. No instructions were given to persuade participants to create a list of feared and avoided situations and to gradually and consistently approach and engage with these situations. Collectively, the MEMI app focused primarily on teaching mindfulness principles and skills.

The experimenter was available to address questions and then administered the 6-item Credibility and Expectancy Questionnaire [ 56 ]. After participants grasped the rationale and techniques of mindfulness, they set up the MEMI app on their smartphones. Furthermore, participants were provided with a copy of the MEMI rationale handout and were encouraged to engage with it consistently ( Multimedia Appendix 1 ).

Self-Monitoring App

The video presentation created for self-monitoring participants commenced with the principal investigator (NHZ) explaining self-monitoring as an elevated awareness of one’s thoughts and emotions, with particular attention to one’s uncomfortable experiences. Following this, the video introduced the idea that monitoring thoughts and tracking associated distress could lead to healthier thinking. In essence, the self-monitoring video conveyed that self-monitoring (ie, focusing solely on distress) had the potential to alleviate anxiety. Self-monitoring was used in another set of studies that inspired the conceptual foundation of our self-monitoring app [ 57 , 58 ]. This adaptation sought to emulate the MEMI structure while intentionally excluding its presumed therapeutic elements, including acceptance, diaphragmatic breathing training, focusing on transient present moments, open monitoring, and a regimen of consistent mindfulness practices. Notably, it refrained from introducing the concept of mindfulness and abstained from guiding participants toward engaging with the present moment in ways that could influence their mood. In contrast to MEMI, the self-monitoring app encouraged participants to simply observe their thoughts and emotions and did not include emphasis on the need to accept these thoughts and feelings as they emerged. Furthermore, it did not include instructions for breathing retraining or the intention to elicit relaxation through abdominal breathing. Unlike MEMI, which promoted ongoing mindfulness practice, self-monitoring participants were not prompted to engage in self-monitoring between prompts and after the intervention. This self-monitoring approach was strategically designed to address potential credibility and expectancy effects, minimize the likelihood of regression to the mean, and mitigate the possibility of inflated effect sizes that could occur with a no-treatment or waitlist control group [ 59 ] ( Multimedia Appendix 1 ).

Instead of receiving lengthier messages to practice mindfulness continually as with the MEMI group, self-monitoring participants received the following brief instruction 5 times daily (at about 9 AM, noon, 3 PM, 6 PM, and 9 PM) for 14 days: “Notice your thoughts and how distressing they may be.” State-level depression, anxiety, and mindfulness were measured using the same 9-point Likert scale items before and after each self-monitoring prompt. Subsequently, similar to the MEMI group, experimenters administered the 6-item Credibility and Expectancy Questionnaire after confirming self-monitoring participants’ comprehension of the rationale and self-monitoring technique. Participants received compensation through either course extra credit hours or monetary rewards.

The self-monitoring app was designed to work in the following manner. By suggesting that solely self-monitoring and focusing on distressing thought patterns might remedy anxiety, the app controlled for treatment expectancy and credibility effects across arms. Furthermore, by eliminating active therapeutic ingredients in the MEMI app (eg, focused attention on the present moment, open monitoring, diaphragmatic breathing, and nonjudgmental acceptance), the self-monitoring app functioned as a placebo comparator to maximize the odds of attaining between-arm efficacy on clinical outcomes.

Pre- and Midintervention Measures

The Difficulties in Emotion Regulation Scale (DERS) [ 41 ] is a 36-item web-based self-report tool that evaluates difficulties in regulating emotions. Participants responded on a 5-point scale (1= almost never, 0%-10% to 5= almost always, 91%-100% ). The scale provides a total score (range 36-180), where higher scores reflect more ER difficulties. In addition, 6 ER difficulty facets were evaluated via the following DERS subscales: nonacceptance of emotions (acceptance), emotional awareness issues (awareness), lack of emotional clarity (clarity), goal-directed behavior difficulties (goals), impulse control problems (impulsivity), and strategy use problems (strategy and difficulty accessing ER skills) [ 41 ]. The DERS has shown strong internal consistency, excellent 2-month retest reliability [ 41 ], and good convergent and discriminant validity [ 60 ]. The internal consistency (Cronbach α values) was excellent at .95 at both prerandomization and 7-day midintervention time points.

Trait Self-Compassion

Participants rated on a 5-point Likert scale (1= almost never to 5= almost always ) the extent to which they identified with each statement on the Self-Compassion Scale (SCS) [ 8 ]. The SCS involves six distinct aspects assessed via the following subscales: (1) acknowledging shared human struggles (common humanity), (2) social connectedness (feeling a sense of being linked with other humans), (3) mindfulness (nonjudgmental awareness), (4) nonidentification with emotions (the absence of intense focus on negative emotions), (5) nonjudgment toward oneself (self-soothing tendencies during times of distress), and (6) self-kindness (showing warmth toward one’s imperfections) [ 8 ]. The SCS has shown strong internal consistency, predictive validity, convergent validity, discriminant validity, and retest reliability [ 61 - 63 ]. The Cronbach α values were good at 0.95 and 0.96 at prerandomization and 7-day midintervention time points, respectively.

SAD Symptoms

The 25-item SPDQ [ 48 ] assessed SAD fear and avoidance symptoms in different social situations, aligned with the DSM Fourth Edition criteria [ 64 ]. It has shown good retest reliability and strong internal consistency (Cronbach α=.96, .97, and .98 at prerandomization, postintervention, and 1MFU time points, respectively) [ 48 ]. Furthermore, it displayed robust convergent and discriminant validity, with high sensitivity (82%), specificity (85%) [ 48 ], and responsiveness to symptom reduction changes in clinical trials [ 13 ].

The 17-item SPIN [ 45 ] evaluated the SAD fear and avoidance symptoms in the past week, such as the fear of social embarrassment. Participants used a 5-point Likert scale (1= not at all to 5= extremely ) to rate the relevance of each statement to their past week’s experiences. Prior research has shown that the SPIN exhibited acceptable convergent validity with other established assessments of SAD [ 50 , 65 ]. The SPIN showed strong internal consistency in this study (Cronbach α=.93, .94, and .95 at prerandomization, postintervention, and 1MFU time points, respectively). Finally, a cut-off SPIN score ≥20 yielded excellent psychometric properties, with a sensitivity of 0.85, specificity of 0.86, positive predictive value of 0.85, negative predictive value of 0.85, and correct classification rate of 85% [ 50 ], when compared to the structured interview for the DSM Fourth Edition [ 66 , 67 ]. Excellent psychometric properties with a similar cut-off have been replicated in another Asian sample [ 68 ].

Generalized Anxiety Symptoms

The Generalized Anxiety Disorder Questionnaire–fourth edition (GADQ-IV) [ 69 ], comprising 14 items, evaluated the symptoms of generalized anxiety disorder (GAD) through a combination of dichotomous (“yes” or “no”) and continuous responses. The continuous responses included a 9-point Likert scale for assessing the interference and distress caused by GAD symptoms. The GADQ-IV showed strong internal consistency (Cronbach α=0.93, 0.93, and 0.94 at prerandomization, postintervention, and 1MFU time points, respectively) and high retest reliability [ 69 ]. Furthermore, it exhibited strong convergent and discriminant validity, and aligned well with the structured diagnostic assessments of GAD [ 70 , 71 ].

Depression Severity

The Beck Depression Inventory–second edition (BDI-II) [ 72 ], which is a 21-item scale, assesses depression symptom severity. Participants rated the severity of each symptom based on their experiences in the past 2 weeks, using a scale of 0 to 3 to indicate increasing severity. The BDI-II showed excellent internal consistency (Cronbach α=.93, .95, .95 at prerandomization, postintervention, and 1MFU time points, respectively), high retest reliability, and strong convergent and discriminant validity [ 73 , 74 ].

Trait Repetitive Negative Thinking

The 45-item Perseverative Cognitions Questionnaire (PCQ; PCQ-45) [ 75 ] assessed persistent, repetitive negative thinking tendencies related to worrisome, obsessive, and ruminative thoughts. Participants rated the items on a 6-point Likert scale (0= strongly disagree to 5= strongly agree ). The PCQ-45 included 6 domains: anticipating the worst, brooding over the past, future preparation, thoughts conflicting with the ideal self, uncontrollability, and seeking causes and meanings [ 75 ]. We calculated the total PCQ score by computing the average of all subscale scores. The scale showed high 2-week retest reliability and strong discriminant and convergent validity (Cronbach α=.96, .97, and .98 at prerandomization, postintervention, and 1MFU time points, respectively), and it demonstrated cross-cultural measurement invariance between the United States and Singapore [ 76 ].

Trait Mindfulness

The 39-item Five-Factor Mindfulness Questionnaire (FFMQ) [ 77 ] measured participants’ inclination to practice mindfulness across 5 domains: awareness of the consequences of actions, description, nonjudgment, nonreactivity to inner experiences, and observation. Participants rated these aspects on a 5-point Likert scale (1= never or very rarely true to 5= very often or always true ). The FFMQ total score showed strong convergent validity [ 78 ], differentiation from measures of unrelated factors (eg, psychological well-being) [ 77 ], and retest reliability [ 79 ]. The Cronbach α values were .90, .91, and .93 at prerandomization, postintervention, and 1MFU time points, respectively.

Statistical Analysis

Power analysis and missing data management.

On the basis of a Monte Carlo power analysis [ 80 , 81 ], this study had 100% power to detect a substantial group × time interaction with a small effect size of Cohen d =0.2. We performed intent-to-treat [ 82 ] analyses with the enrolled 191 participants by incorporating data from participants who did not meet the 7-day compliance check (completing ≥80% of the app prompts in the 2-week intervention phase). Participants demonstrated strong adherence, with 78% (149/191) responding to at least 80% (56/70) of the MEMI and self-monitoring prompts. A total of 2.03% of the data from the pre-1MFU time points were missing. To address this, we applied multiple imputation using the recommended predictive mean matching algorithm by pooling data from 100 imputed data sets, each with 10 iterations [ 83 ].

Hierarchical Linear Modeling

To test hypothesis 1, we used hierarchical linear modeling [ 84 ], also known as multilevel modeling, to account for data nonindependence resulting from the nesting of repeated observations (level 1) within participants (level 2). For each hierarchical linear modeling, group (intervention), time, and group × time interaction were fixed-effect predictors of improvement in the prerandomization to midintervention ER and self-compassion domains, and the intercept (ie, time-coded as 0 for preintervention and 1 for midintervention time points) was the single random-effect predictor (permitting participants to vary in their average-outcome values). We used fitted models to determine the estimated mean scores at each time point.

Parallel Structural Equation Modeling Mediation Analyses

To test hypothesis 2, we conducted parallel mediation analyses using the lavaan structural equation modeling R package (R Foundation for Statistical Computing) [ 85 ] to assess whether pretrial to midintervention ER and self-compassion domains mediated group effects on the 6 clinical outcomes from pre-post and pre-1MFU time points. Similar to the primary RCT study, the clinical outcomes examined were the alleviation of pre-post and pre-1MFU SAD symptoms, generalized anxiety symptoms, depression severity, repetitive negative thinking, and trait mindfulness. Although there were no significant between-group effects from pre-post and pre-1MFU time points [ 43 ], we conducted mediation tests because significant indirect effects can exist without significant between-group total or direct effects [ 86 ]. Furthermore, our primary efficacy paper still showed significant within-group effects on these 5 clinical outcomes from pre-post and pre-1MFU time points [ 43 ]. Moreover, as recommended, our mediation tests aligned with theoretical considerations [ 87 ] and statistical recommendations to measure mediators at the midpoint of the intervention. In order to establish them as potential mechanisms of change, mediators should temporally precede the outcome measures [ 88 ]. The associations between groups, pretrial to midintervention mediators, and pre-post or pre-1MFU outcomes can be described using 4 regression coefficients (or paths) [ 89 ]: the group effect on the pretrial to midintervention mediator (path a ), the effect of the pretrial to midintervention mediator on pre-post or pre-1MFU outcome (path b ), the total effect (path c ; the combination of the a and b paths), and the direct effect (path c ’; the group effect on the outcome, irrespective of the mediator). The product of paths a and b signifies the mediated (or indirect) effect, that is, the focal estimate when examining mediation. We assessed the significance of the mediated effect via a nonparametric bootstrap method, generating 2000 random samples to determine point estimates, SEs, and bias-corrected bootstrap 95% CIs for each indirect effect. The mediated effect was considered significant if the CI bounds (both upper and lower) did not include 0. To safeguard against type I errors, we conducted post hoc comparisons for statistically significant effects using the Simes Bonferroni correction method [ 90 ]. Cohen d effect sizes and their 95% CIs were calculated to ease the interpretation of parameter estimates for both study hypotheses, such that Cohen d values of 0.2, 0.5, and 0.8 denote small, moderate, and large effects, respectively [ 91 ].

Table S2 in Multimedia Appendix 1 displays descriptive statistics of the trait ER and self-compassion domain scores for participants in the MEMI and self-monitoring groups.

Between- and Within-Group Effects of Brief MEMI Versus Self-Monitoring on ER Domains

Significant between-group effects occurred from pretrial to midintervention time points on DERS domains, including lack of emotional clarity (Cohen d =0.16, 95% CI 0.02-0.31; P= .03) and difficulties engaging in goal-directed behavior (Cohen d =−0.24, 95% CI −0.39 to −0.10; P= .001; Tables 1 and 2 ). Regarding within-group effects, reductions in these difficulties were significantly stronger for the MEMI than self-monitoring groups from pretrial to midintervention time points (lack of emotional clarity: Cohen d =−0.39, 95% CI −0.54 to −0.25; P< .001 vs Cohen d =−0.21, 95% CI −0.35 to −0.06; P= .004 and goal-directed behavior difficulties: Cohen d =−0.73, 95% CI −0.88 to −0.58; P< .001 vs Cohen d =−0.29, 95% CI −0.44 to −0.15; P< .001). No significant between-group effects emerged from pretrial to midintervention time points on emotional awareness issues (Cohen d =0.11, 95% CI −0.03 to 0.26; P= .12), impulse control issues (Cohen d =0.11, 95% CI −0.03 to 0.26; P= .12), nonacceptance of emotions (Cohen d =−0.07, 95% CI −0.21 to 0.08; P= .36), and strategy use problems (Cohen d =−0.12, 95% CI −0.27 to 0.02; P= .10). Regarding within-group effects, both the MEMI and self-monitoring app significantly reduced impulse control issues (MEMI: Cohen d =−0.32, 95% CI −0.47 to −0.18; P= .006 and self-monitoring: Cohen d =−0.26, 95% CI −0.40 to −0.11; P< .001), nonacceptance of emotions (MEMI: Cohen d =−0.32, 95% CI −0.47 to −0.18; P< .001; self-monitoring: Cohen d =−0.24, 95% CI −0.38 to −0.09; P= .001), and strategy use problems (MEMI: Cohen d =−0.40, 95% CI −0.54 to −0.25; P< .001; self-monitoring: Cohen d =−0.21, 95% CI −0.35 to −0.06; P= .004) from pretrial to midintervention time points. A significant reduction in emotional awareness issues from pretrial to midintervention time points was observed in the self-monitoring group (Cohen d =−0.73, 95% CI −0.88 to −0.58; P< .001) but not in the MEMI group (Cohen d =−0.29, 95% CI −0.44 to 0.02; P= .07).

a DERS: Difficulties in Emotion Regulation Scale.

b SCS: Self-Compassion Scale.

a MEMI: mindfulness ecological momentary intervention.

b DERS: Difficulties in Emotion Regulation Scale.

c SCS: Self-Compassion Scale.

Between- and Within-Group Effects of Brief MEMI Versus Self-Monitoring on Self-Compassion Domains

Significant between-group effects occurred from pretrial to midintervention time points on SCS domains, including acknowledging shared human struggles (Cohen d =0.26, 95% CI 0.11-0.40; P< .001), social connectedness (Cohen d =0.19, 95% CI 0.04-0.33; P= .01), nonidentification with emotions (Cohen d =0.16, 95% CI 0.01-0.30; P= .03), and self-kindness (Cohen d =0.19, 95% CI 0.04-0.33; P= .009; Tables S3 and S4 in Multimedia Appendix 1 ). Regarding within-group effects, increases in these self-compassion domains were significantly larger for the MEMI than self-monitoring groups from pretrial to midintervention time points (acknowledging shared human struggles: Cohen d =0.13, 95% CI −0.02 to 0.27; P= .08 vs Cohen d =−0.23, 95% CI −0.38 to −0.08; P= .002); social connectedness: Cohen d =0.45, 95% CI 0.30-0.60; P< .001 vs Cohen d =0.19, 95% CI 0.04-0.33; P= .01; identification with emotions: Cohen d =0.63, 95% CI 0.48-0.78; P< .001 vs Cohen d =0.48, 95% CI 0.33-0.62; P< .001; and self-kindness: Cohen d =0.36, 95% CI 0.21-0.50; P< .001 vs Cohen d =0.10, 95% CI −0.05 to 0.24; P= .17). No significant between-group effects emerged from pretrial to midintervention time points on mindfulness (Cohen d <0.01, 95% CI −0.14 to 0.15; P= .97) and nonjudgment toward oneself (Cohen d =0.12, 95% CI −0.02 to 0.27; P= .10) domains. Neither the MEMI (Cohen d =0.03, 95% CI −0.12 to 0.17; P= .73) nor self-monitoring app (Cohen d =0.02, 95% CI −0.12 to 0.17; P= .77) significantly changed mindfulness from pretrial to midintervention time points. However, both the MEMI (Cohen d =0.61, 95% CI 0.46-0.76; P< .001) and self-monitoring app (Cohen d =0.53, 95% CI 0.38-0.68; P< .001) significantly increased nonjudgment toward oneself from pretrial to midintervention time points.

Testing ER and Self-Compassion as Mediators of Outcomes From Pre-Post Time Points

Neither ER nor self-compassion domains significantly mediated the effect of the treatment (brief MEMI vs self-monitoring) on pre- and postintervention SAD symptoms indexed by SPDQ and SPIN scores, generalized anxiety symptoms (GADQ-IV score), depression severity (BDI-II score), trait repetitive negative thinking (PCQ score), and trait mindfulness (FFMQ score; Table S3 in Multimedia Appendix 1 ). Therefore, hypothesis 2a was not supported.

Testing ER and Self-Compassion as Mediators of Outcomes From Pre-1MFU Time Points

Neither ER nor self-compassion domains significantly mediated the effect of the treatment (brief MEMI vs self-monitoring) on pre-1MFU SAD symptoms indexed by SPDQ and SPIN scores, generalized anxiety symptoms (GADQ-IV score), depression severity (BDI-II score), trait repetitive negative thinking (PCQ score), and trait mindfulness (FFMQ score; Table S4 in Multimedia Appendix 1 ). Therefore, hypothesis 2b was not supported. In a sensitivity analysis of single-mediator models, we found that mediation effects were still not statistically significant when the overall scores of DERS or SCS were tested as a single mediator for all 6 clinical outcomes.

Comparison With Prior Work

This study determined the efficacy of a 14-day, fully self-guided, mobile MEMI on ER difficulties and self-compassion domains from pretrial to midintervention time points and whether such changes mediated 6 clinical outcomes from pre-post and pre-1MFU time points. Extending a prior report of primary outcomes [ 43 ], this RCT determined which domains of ER difficulties and self-compassion a brief MEMI could impact from pretrial to midintervention time points. Encouragingly, the results showed that, similar to applied relaxation and self-compassion training [ 92 ], brief, fully self-guided, mobile MEMIs efficaciously increased specific self-compassion domains. The MEMI enhanced acknowledging shared human struggles (common humanity), social connectedness (sense of being linked with other people), nonidentification with emotions (not overassociating with one’s feelings), and self-kindness domains from pretrial to midintervention time points. Therefore, the MEMI was superior to self-monitoring in increasing the ER domains linked with goal-directed behavior pursuit and emotional clarity.

Why was the brief MEMI more efficacious than self-monitoring in enhancing social connectedness, nonidentification with emotions, self-kindness, acknowledging shared human struggles, reducing goal-directed behavior difficulties, and lack of emotional clarity from pretrial to midintervention time points? Consistent with the mindfulness-to-meaning theory [ 93 ], the MEMI might have helped cultivate nonjudgmental awareness of one’s and others’ experiences, steering clear of both rumination and the suppression of distressing emotions typical of SAD. By teaching the regulation of physiological anxiety and the practice of self-kindness during moments of confusion or insecurity, the brief MEMI might offer greater benefits than self-monitoring in managing processes related to SAD. This approach might encourage constructive engagement with experiences instead of getting stuck in the vicious cycle of avoidance and missing out on positive interactions that could disconfirm feared social outcomes and offer feelings of self-efficacy across diverse interpersonal contexts [ 94 , 95 ]. Furthermore, the emphasis on acceptance and constructive awareness by the MEMI instead of self-monitoring might have contributed to these findings. Prior dismantling studies of MBIs that evidenced stronger comparative efficacy of acceptance and monitoring over monitoring alone on clinically relevant outcomes have alluded to this possibility [ 96 ].

Moreover, it is noteworthy that both brief MEMI and self-monitoring reduced ER difficulties linked to impulse control issues, nonacceptance of emotional responses, strategy use problems, and nonjudgment toward oneself from pretrial to midintervention time points. These null between-group effects could be viewed as a comparison between awareness alone and awareness combined with acceptance, where acceptance did not contribute anything additional to generate differential comparative efficacy on these outcomes. Perhaps learning via both the MEMI and self-monitoring to calmly acknowledge current thoughts, emotions, and bodily sensations without judgment, regardless of intensity or discomfort, fostered the ability to endure distress and engage with experiences rather than avoid them [ 97 ]. Furthermore, these processes might have motivated actions and curbed the inclination to engage in self-sabotaging behaviors, including self-criticism and impulsive actions [ 98 , 99 ]. In addition, these alterations may have been due to placebo and related factors rather than the influence of “active ingredients” [ 100 ]. Without the MEMI showing superior efficacy over self-monitoring on these outcomes, we cannot determine whether the improvements exceeded the effects of the passage of time, expectancy effects, and related factors. Future RCTs could empirically evaluate the validity of these conjectures.

Notably, self-monitoring, but not the MEMI, led to an increased emotional awareness. This surprising finding could be attributed to repeated instructions by self-monitoring prompts to attend to thoughts and feelings with explicit instructions to notice how distressing they might be. Another explanation might be that self-monitoring could have enhanced emotional self-awareness and related mental health outcomes, such as psychological well-being [ 101 ].

Another counterintuitive finding was the absence of between- and within-group effectiveness of the MEMI on the mindfulness domain of the SCS from pretrial to midintervention time points. A potential account for these nonsignificant outcomes was that more frequent, intense, and longer-duration mindfulness practices were necessary. Consistent with this inference, a 2-week MEMI positively affected trait mindfulness from pre-1MFU but not from pre- to postintervention time points [ 102 ]. Alternatively, the mindfulness measure used in this study might lack sensitivity to detect a change in mindfulness from pretrial to midintervention time points.

In addition, contrary to our expectations, no examined pretrial to midintervention ER or self-compassion domains were mediators of the effect of MEMI on the pre-post and pre-1MFU SAD symptoms, generalized anxiety symptoms, depression severity, repetitive negative thinking, and trait mindfulness. Other studies on brief MEMIs similarly found that ER and self-compassion did not consistently mediate treatment effects on mental health outcomes in various populations, such as distressed college students [ 103 ], health care workers [ 104 ], and other nonclinical samples [ 105 ]. These nonsignificant mediation effects might suggest that ER and self-compassion domains were not the change mechanisms of brief, fully self-guided, mobile MEMIs, as previously suggested [ 31 , 32 ]. On the basis of prior mechanism-focused trials that delivered higher-intensity MBIs [ 106 - 110 ], longer-duration and more rigorous MEMIs may be necessary to observe ER and self-compassion domains as mediators of the effect of brief MEMI on various clinical outcomes in SAD. Higher-intensity MBIs may nurture stronger consolidation of mindfulness and related ER skills over long durations and across various contexts by modifying habitual responses to social and associated stressors in SAD [ 111 ]. Furthermore, it is possible that the ER and self-compassion measures used in this study were not sensitive enough to detect any existing true mediation effects. Alternatively, based on emerging evidence, other constructs, such as increased acceptance-attention, nonreactivity [ 112 ], and cognitive reappraisal [ 113 ], should be examined as mediators or proxy change mechanisms of brief MEMI on diverse SAD outcomes. Brief MEMIs could potentially enhance alternative mechanisms that boost mental well-being, coping skills, and related factors in individuals with SAD. These testable ideas await experimental evaluation.

Limitations and Strengths

This study had some limitations. First, all putative mediators and clinical outcome measures relied on web-based self-reports. Future RCTs should incorporate diverse measures of related constructs and evaluate crucial yet understudied factors believed to drive change, such as attentional bias toward threats [ 114 , 115 ], executive functioning [ 102 , 116 , 117 ], and physiological anxiety indices [ 118 , 119 ]. Second, we did not evaluate the sustainability of mindfulness skill use after the 2-week intervention duration. Third, we measured difficulties in ER and self-compassion domains at prerandomization and midintervention time points but not at posttreatment and 1MFU time points. This approach enabled us to evaluate the initial mechanism conditions and changes during the active treatment phase and test whether such conditions led to pre-1MFU changes in clinical outcomes [ 120 ]. Future investigations should delve into whether persistent mindfulness engagement, even without recurring MEMI instructions, can predict efficacy and change mechanisms at follow-up evaluations [ 121 ]. Fourth, the level of intensity in the brief MEMI may have been insufficient to produce substantial improvement in clinical outcomes via putative mediators. Finally, we used a clinical cut-off score on a self-report measure to determine a SAD diagnosis, instead of relying on a diagnostic interview.

Despite these limitations, this study had notable strengths, including a robust RCT design featuring an active comparator and high adherence rates, an uncommon achievement in the digital mental health intervention field, which often faces challenges such as high dropout and low use [ 122 ]. Furthermore, participants exhibited good compliance, with 78% (149/191) engaging in at least 80% of the MEMI and self-monitoring prompts. In addition, these measures have a well-established history of use and psychometric validation in previous trials [ 13 , 102 , 123 , 124 ]. Another strength of our study was the recruitment of a sample from Singapore, a Southeast Asian country. This method strengthened the potential for cross-cultural applicability, addressing a vital lacuna within the field of clinical psychology, which has traditionally concentrated on Western settings [ 76 , 125 ].

Conclusions

If replicated, our study has several clinical implications. Our observed advantages of a brief MEMI versus self-monitoring in decreasing isolation (sense of social disconnection) and overidentification (excessive attention to negative emotions) and increasing self-kindness (tenderness toward flaws in oneself), goal-directed behaviors, and emotional clarity in people with SAD indicated that established first-line treatments, such as CBT, could benefit from integrating self-compassion and mindfulness strategies. The finding that brief MEMI was not significantly different than self-monitoring in decreasing impulse control difficulties, nonacceptance of emotional responses, limited access to ER strategies, and self-judgment (excessive self-criticism) implies that the benefits of self-monitoring alone could be highlighted in the treatment to promote self-efficacy before moving on to graded exposure therapy and other CBT components for SAD. Higher-intensity brief, fully self-guided mobile MEMIs are likely necessary to identify which ER and self-compassion domains are change mechanisms [ 126 ]. In addition to identifying change mechanisms and establishing efficacy, it will be essential to investigate which individuals with SAD will derive the greatest benefit from the brief MEMI.

Acknowledgments

NHZ, HHT, RYH, and MGN take full responsibility for the data, accuracy of analyses and interpretation, and conduct of the research. This study was conducted in compliance with the American Psychological Association and Declaration of Helsinki ethical standards in treating human participants and was approved by the institutional review board. Informed consent was obtained from participants as per the National University of Singapore Institutional Review Board requirements. This randomized controlled trial received partial funding from the National Institute of Mental Health (R01 MH115128), the Pennsylvania State University Office of Research and Graduate Studies Dissertation Award, the Penn State Susan Welch-Nagle Family Graduate Fellowship, the National University of Singapore Development Grant, and the Association for Behavioral and Cognitive Therapies Leonard Krasner Student Dissertation Award.

Authors' Contributions

All authors have (1) made substantial contributions to the analysis and interpretation of the study and its findings, (2) drafted and revised the paper for intellectual content, and (3) given their final approval of the version to be submitted. The manuscript has been read and approved by all authors. NHZ was involved in the conceptualization, methodology (design and creation of models), funding acquisition, formal analysis, coding with the R software (R Foundation for Statistical Computing) for data analysis, visualization, and writing (original draft, reviewing, and editing) of this work. HHT assisted heavily with the data collection and various logistical aspects of the study. RYH and MGN assisted with the study conceptualization, design, data curation, validation, supervision, and writing (original draft, reviewing, and editing).

Conflicts of Interest

None declared.

Study methodology details.

CONSORT-EHEALTH (Consolidated Standards of Reporting Trials of Electronic and Mobile Health Applications and Online Telehealth) diagram.

• Gross JJ. Emotion regulation: affective, cognitive, and social consequences. Psychophysiology. May 2002;39(3):281-291. [ CrossRef ] [ Medline ]
• Lazarus RS. Progress on a cognitive-motivational-relational theory of emotion. Am Psychol. 1991;46(8):819-834. [ CrossRef ]
• Jazaieri H, Morrison AS, Goldin PR, Gross JJ. The role of emotion and emotion regulation in social anxiety disorder. Curr Psychiatry Rep. Jan 2015;17(1):531. [ CrossRef ] [ Medline ]
• Aldao A, Nolen-Hoeksema S, Schweizer S. Emotion-regulation strategies across psychopathology: a meta-analytic review. Clin Psychol Rev. Mar 2010;30(2):217-237. [ CrossRef ] [ Medline ]
• Christiansen H, Hirsch O, Albrecht B, Chavanon ML. Attention-deficit/hyperactivity disorder (ADHD) and emotion regulation over the life span. Curr Psychiatry Rep. Mar 02, 2019;21(3):17. [ CrossRef ] [ Medline ]
• Diedrich A, Grant M, Hofmann SG, Hiller W, Berking M. Self-compassion as an emotion regulation strategy in major depressive disorder. Behav Res Ther. Jul 2014;58:43-51. [ CrossRef ] [ Medline ]
• Gilbert P. The origins and nature of compassion focused therapy. Br J Clin Psychol. Mar 21, 2014;53(1):6-41. [ CrossRef ] [ Medline ]
• Neff KD. The development and validation of a scale to measure self-compassion. Self Identity. Jul 2003;2(3):223-250. [ CrossRef ]
• Neff K. Self-compassion: an alternative conceptualization of a healthy attitude toward oneself. Self Identity. Apr 2003;2(2):85-101. [ CrossRef ]
• Inwood E, Ferrari M. Mechanisms of change in the relationship between self-compassion, emotion regulation, and mental health: a systematic review. Appl Psychol Health Well Being. Jul 2018;10(2):215-235. [ CrossRef ] [ Medline ]
• Zainal NH, Newman MG. Relation between cognitive and behavioral strategies and future change in common mental health problems across 18 years. J Abnorm Psychol. May 2019;128(4):295-304. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Hong RY, Zainal NH, Ong XL. Longitudinal associations between academic competence-building and depression symptoms in early adolescence. Dev Psychopathol. Aug 12, 2022;35(4):2061-2072. [ CrossRef ]
• Zainal NH, Chan WW, Saxena AP, Taylor CB, Newman MG. Pilot randomized trial of self-guided virtual reality exposure therapy for social anxiety disorder. Behav Res Ther. Dec 2021;147:103984. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Lindsay EK, Creswell JD. Mindfulness, acceptance, and emotion regulation: perspectives from Monitor and Acceptance Theory (MAT). Curr Opin Psychol. Aug 2019;28:120-125. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Lindsay EK, Creswell JD. Mechanisms of mindfulness training: Monitor and Acceptance Theory (MAT). Clin Psychol Rev. Feb 2017;51:48-59. [ FREE Full text ] [ CrossRef ] [ Medline ]
• MacBeth A, Gumley A. Exploring compassion: a meta-analysis of the association between self-compassion and psychopathology. Clin Psychol Rev. Aug 2012;32(6):545-552. [ CrossRef ] [ Medline ]
• Wasson RS, Barratt C, O'Brien WH. Effects of mindfulness-based interventions on self-compassion in health care professionals: a meta-analysis. Mindfulness (N Y). 2020;11(8):1914-1934. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Kabat-Zinn J, Massion AO, Kristeller J, Peterson LG, Fletcher KE, Pbert L, et al. Effectiveness of a meditation-based stress reduction program in the treatment of anxiety disorders. Am J Psychiatry. Jul 1992;149(7):936-943. [ CrossRef ] [ Medline ]
• Segal Z, Williams JM, Williams M, Teasdale J. Mindfulness-Based Cognitive Therapy for Depression. New York, NY. Guilford Publications; 2012.
• Creswell JD. Mindfulness interventions. Annu Rev Psychol. Jan 03, 2017;68:491-516. [ CrossRef ] [ Medline ]
• Harvey LJ, White FA. Emotion self-stigma as a unique predictor of help-seeking intentions: a comparative analysis of early adolescents and young adults. Psychol Psychother. Sep 2023;96(3):762-777. [ CrossRef ] [ Medline ]
• Dschaak ZA, Spiker DA, Berney EC, Miller ME, Hammer JH. Collegian help seeking: the role of self-compassion and self-coldness. J Ment Health. Jun 2021;30(3):284-291. [ CrossRef ] [ Medline ]
• Schumer MC, Lindsay EK, Creswell JD. Brief mindfulness training for negative affectivity: a systematic review and meta-analysis. J Consult Clin Psychol. Jul 2018;86(7):569-583. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Tang YY, Tang R, Posner MI. Mindfulness meditation improves emotion regulation and reduces drug abuse. Drug Alcohol Depend. Jun 01, 2016;163 Suppl 1:S13-S18. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Bernstein EE, LeBlanc NJ, Bentley KH, Barreira PJ, McNally RJ. A single-session workshop to enhance emotional awareness and emotion regulation for graduate students: a pilot study. Cognit Behav Pract. Aug 2021;28(3):393-409. [ CrossRef ]
• Kazdin AE. Drawing causal inferences from randomized controlled trials in psychotherapy research. Psychother Res. Nov 2023;33(8):991-1003. [ CrossRef ] [ Medline ]
• Lindsay EK, Young S, Brown KW, Smyth JM, Creswell JD. Mindfulness training reduces loneliness and increases social contact in a randomized controlled trial. Proc Natl Acad Sci U S A. Feb 26, 2019;116(9):3488-3493. [ FREE Full text ] [ CrossRef ] [ Medline ]
• McLean CL, Ruork AK, Ramaiya MK, Fruzzetti AE. Feasibility and initial impact of single-session internet-delivered acceptance vs change skills for emotions for stress- and trauma-related problems: a randomized controlled trial. Behav Cogn Psychother. Sep 2023;51(5):443-458. [ CrossRef ] [ Medline ]
• Martínez-Rubio D, Navarrete J, Montero-Marin J. Feasibility, effectiveness, and mechanisms of a brief mindfulness- and compassion-based program to reduce stress in university students: a pilot randomized controlled trial. Int J Environ Res Public Health. Dec 23, 2021;19(1):154. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Bergen-Cico D, Possemato K, Cheon S. Examining the efficacy of a brief mindfulness-based stress reduction (Brief MBSR) program on psychological health. J Am Coll Health. 2013;61(6):348-360. [ CrossRef ] [ Medline ]
• Gratz KL, Tull MT. Emotion regulation as a mechanism of change in acceptance-and mindfulness-based treatments. In: Assessing Mindfulness and Acceptance: Illuminating the Processes of Change. Oakland, CA. New Harbinger Publications; 2010.
• Baer RA. Self-compassion as a mechanism of change in mindfulness- and acceptance-based treatments. In: Baer RA, editor. Assessing Mindfulness and Acceptance Processes in Clients: Illuminating the Theory and Practice of Change. Oakland, CA. New Harbinger Publications; 2010;135-153.
• Maloney S, Surawy C, Martin M, Montero-Marin J, Kuyken W. The state- and trait-level effects and candidate mechanisms of four mindfulness-based cognitive therapy (MBCT) practices: two exploratory studies. Mindfulness (N Y). Aug 24, 2023;14(9):2155-2171. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Ma Y, She Z, Siu AF, Zeng X, Liu X. Effectiveness of online mindfulness-based interventions on psychological distress and the mediating role of emotion regulation. Front Psychol. 2018;9:2090. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Kraemer HC, Kiernan M, Essex M, Kupfer DJ. How and why criteria defining moderators and mediators differ between the Baron & Kenny and MacArthur approaches. Health Psychol. Mar 2008;27(2S):S101-S108. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Werner KH, Goldin PR, Ball TM, Heimberg RG, Gross JJ. Assessing emotion regulation in social anxiety disorder: the emotion regulation interview. J Psychopathol Behav Assess. Apr 15, 2011;33(3):346-354. [ CrossRef ]
• Werner KH, Jazaieri H, Goldin PR, Ziv M, Heimberg RG, Gross JJ. Self-compassion and social anxiety disorder. Anxiety Stress Coping. 2012;25(5):543-558. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Blackie RA, Kocovski NL. Forgive and let go: effect of self-compassion on post-event processing in social anxiety. Mindfulness. Sep 15, 2017;9(2):654-663. [ CrossRef ]
• Goodman FR, Kashdan TB, İmamoğlu A. Valuing emotional control in social anxiety disorder: a multimethod study of emotion beliefs and emotion regulation. Emotion. Jun 2021;21(4):842-855. [ CrossRef ] [ Medline ]
• Dixon ML, Moodie CA, Goldin PR, Farb N, Heimberg RG, Gross JJ. Emotion regulation in social anxiety disorder: reappraisal and acceptance of negative self-beliefs. Biol Psychiatry Cogn Neurosci Neuroimaging. Jan 2020;5(1):119-129. [ CrossRef ] [ Medline ]
• Gratz KL, Roemer L. Multidimensional assessment of emotion regulation and dysregulation: development, factor structure, and initial validation of the difficulties in emotion regulation scale. J Psychopathol Behav Assess. Mar 2004;26(1):41-54. [ CrossRef ]
• Neff KD, Tóth-Király I. Self-Compassion Scale (SCS). In: Medvedev ON, Krägeloh CU, Siegert RJ, Singh NN, editors. Handbook of Assessment in Mindfulness Research. Cham, Switzerland. Springer International Publishing; 2022.
• Zainal NH, Tan HH, Hong RY, Newman MG. Is a brief mindfulness ecological momentary intervention more efficacious than a self-monitoring app for social anxiety disorder? a randomized controlled trial. J Anxiety Disord (forthcoming). Mar 24, 2024.:102858. [ CrossRef ]
• Zainal NH, Tan HH, Hong RY, Newman MG. Is a brief mindfulness app efficacious in enhancing emotion regulation and self-compassion in social anxiety disorder? a randomized controlled trial and mediation analysis. OSF Registries. 2023. URL: https://osf.io/m3kxz [accessed 2023-09-03]
• Connor KM, Davidson JR, Churchill LE, Sherwood A, Foa E, Weisler RH. Psychometric properties of the Social Phobia Inventory (SPIN). New self-rating scale. Br J Psychiatry. Apr 02, 2000;176(4):379-386. [ CrossRef ] [ Medline ]
• DSM-5-TR online assessment measures. American Psychiatric Association. URL: https://www.psychiatry.org/psychiatrists/practice/dsm/educational-resources/assessment-measures [accessed 2024-03-29]
• Di Nardo PA, Brown TA, Barlow DH. Anxiety Disorder Interview Schedule for DSM-IV: Lifetime version (ADIS-IV-L). New York, NY. Oxford University Press; 1994.
• Newman MG, Kachin KE, Zuellig AR, Constantino MJ, Cashman-McGrath L. The social phobia diagnostic questionnaire: preliminary validation of a new self-report diagnostic measure of social phobia. Psychol Med. May 2003;33(4):623-635. [ CrossRef ] [ Medline ]
• Antony MM, Coons MJ, McCabe RE, Ashbaugh A, Swinson RP. Psychometric properties of the social phobia inventory: further evaluation. Behav Res Ther. Aug 2006;44(8):1177-1185. [ CrossRef ] [ Medline ]
• Osório FL, Crippa JA, Loureiro SR. Evaluation of the psychometric properties of the Social Phobia Inventory in university students. Compr Psychiatry. 2010;51(6):630-640. [ CrossRef ] [ Medline ]
• Eysenbach G, CONSORT-EHEALTH Group. CONSORT-EHEALTH: improving and standardizing evaluation reports of web-based and mobile health interventions. J Med Internet Res. Dec 31, 2011;13(4):e126. [ FREE Full text ] [ CrossRef ] [ Medline ]
• PACO is an opensource, mobile, behavorial research platform. GitHub. URL: https://github.com/google/paco [accessed 2023-09-09]
• Kabat-Zinn J. Mindfulness-based stress reduction (MBSR). Construct Hum Sci. 2003;8(2):73-107. [ FREE Full text ]
• Heckenberg RA, Eddy P, Kent S, Wright BJ. Do workplace-based mindfulness meditation programs improve physiological indices of stress? a systematic review and meta-analysis. J Psychosom Res. Nov 2018;114:62-71. [ CrossRef ] [ Medline ]
• Kraemer KM, Luberto CM, Hall DL, Ngo LH, Yeh GY. A systematic review and meta-analysis of mindfulness- and acceptance-based interventions for affect intolerance/sensitivity. Behav Res Ther. Dec 2020;135:103746. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Devilly GJ, Borkovec TD. Psychometric properties of the credibility/expectancy questionnaire. J Behav Ther Exp Psychiatry. Jun 2000;31(2):73-86. [ CrossRef ] [ Medline ]
• LaFreniere LS, Newman MG. A brief ecological momentary intervention for generalized anxiety disorder: a randomized controlled trial of the worry outcome journal. Depress Anxiety. Sep 2016;33(9):829-839. [ CrossRef ] [ Medline ]
• LaFreniere LS, Newman MG. Exposing worry's deceit: percentage of untrue worries in generalized anxiety disorder treatment. Behav Ther. May 2020;51(3):413-423. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Cunningham JA, Kypri K, McCambridge J. Exploratory randomized controlled trial evaluating the impact of a waiting list control design. BMC Med Res Methodol. Dec 06, 2013;13:150. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Fowler JC, Charak R, Elhai JD, Allen JG, Frueh BC, Oldham JM. Construct validity and factor structure of the difficulties in emotion regulation scale among adults with severe mental illness. J Psychiatr Res. Nov 2014;58:175-180. [ CrossRef ] [ Medline ]
• Neff KD, McGehee P. Self-compassion and psychological resilience among adolescents and young adults. Self Identity. 2010;9(3):225-240. [ CrossRef ]
• Neff KD, Rude SS, Kirkpatrick KL. An examination of self-compassion in relation to positive psychological functioning and personality traits. J Res Pers. Aug 2007;41(4):908-916. [ CrossRef ]
• Neff KD. The self-compassion scale is a valid and theoretically coherent measure of self-compassion. Mindfulness. Dec 18, 2015;7:264-274. [ CrossRef ]
• Diagnostic and Statistical Manual of Mental Disorders DSM-IV. Washington, DC. American Psychiatric Association; 1994.
• Fogliati VJ, Terides MD, Gandy M, Staples LG, Johnston L, Karin E, et al. Psychometric properties of the Mini-Social Phobia Inventory (Mini-SPIN) in a large online treatment-seeking sample. Cogn Behav Ther. Apr 2016;45(3):236-257. [ CrossRef ] [ Medline ]
• First MB. User's Guide for the Structured Clinical Interview for DSM-IV Axis I Disorders SCID-I Clinician Version. Washington, DC. American Psychiatric Press; 1997.
• Crippa JA, de Lima Osório F, Del-Ben CM, Filho AS, da Silva Freitas MC, Loureiro SR. Comparability between telephone and face-to-face structured clinical interview for DSM-IV in assessing social anxiety disorder. Perspect Psychiatr Care. Oct 2008;44(4):241-247. [ CrossRef ] [ Medline ]
• Nagata T, Nakajima T, Teo AR, Yamada H, Yoshimura C. Psychometric properties of the Japanese version of the social phobia inventory. Psychiatry Clin Neurosci. Apr 2013;67(3):160-166. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Newman MG, Zuellig AR, Kachin KE, Constantino MJ, Przeworski A, Erickson T, et al. Preliminary reliability and validity of the generalized anxiety disorder questionnaire-IV: a revised self-report diagnostic measure of generalized anxiety disorder. Behav Ther. 2002;33(2):215-233. [ CrossRef ]
• Moore MT, Anderson NL, Barnes JM, Haigh EA, Fresco DM. Using the GAD-Q-IV to identify generalized anxiety disorder in psychiatric treatment seeking and primary care medical samples. J Anxiety Disord. Jan 2014;28(1):25-30. [ CrossRef ] [ Medline ]
• Zainal NH, Newman MG. Worry amplifies theory-of-mind reasoning for negatively valenced social stimuli in generalized anxiety disorder. J Affect Disord. Feb 2018;227:824-833. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Beck AT, Steer RA, Ball R, Ranieri W. Comparison of Beck Depression Inventories -IA and -II in psychiatric outpatients. J Pers Assess. Dec 1996;67(3):588-597. [ CrossRef ] [ Medline ]
• García-Batista ZE, Guerra-Peña K, Cano-Vindel A, Herrera-Martínez SX, Medrano LA. Validity and reliability of the Beck Depression Inventory (BDI-II) in general and hospital population of Dominican Republic. PLoS One. Jun 29, 2018;13(6):e0199750. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Huang SL, Hsieh CL, Wu RM, Lu WS. Test-retest reliability and minimal detectable change of the Beck Depression Inventory and the Taiwan Geriatric Depression Scale in patients with Parkinson's disease. PLoS One. Sep 25, 2017;12(9):e0184823. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Szkodny LE, Newman MG. Delineating characteristics of maladaptive repetitive thought: development and preliminary validation of the perseverative cognitions questionnaire. Assessment. Sep 2019;26(6):1084-1104. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Zainal NH, Newman MG, Hong RY. Cross-cultural and gender invariance of transdiagnostic processes in the United States and Singapore. Assessment. Mar 2021;28(2):485-502. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Baer RA, Smith GT, Lykins E, Button D, Krietemeyer J, Sauer S, et al. Construct validity of the five facet mindfulness questionnaire in meditating and nonmeditating samples. Assessment. Sep 2008;15(3):329-342. [ CrossRef ] [ Medline ]
• Goldberg SB, Wielgosz J, Dahl C, Schuyler B, MacCoon DS, Rosenkranz M, et al. Does the five facet mindfulness questionnaire measure what we think it does? construct validity evidence from an active controlled randomized clinical trial. Psychol Assess. Aug 2016;28(8):1009-1014. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Veehof MM, Ten Klooster PM, Taal E, Westerhof GJ, Bohlmeijer ET. Psychometric properties of the Dutch Five Facet Mindfulness Questionnaire (FFMQ) in patients with fibromyalgia. Clin Rheumatol. Aug 2011;30(8):1045-1054. [ FREE Full text ] [ CrossRef ] [ Medline ]
• powerlmm: power analysis for longitudinal multilevel models. The Comprehensive R Archive Network. URL: https://CRAN.R-project.org/package=powerlmm [accessed 2023-09-09]
• Arend MG, Schäfer T. Statistical power in two-level models: a tutorial based on Monte Carlo simulation. Psychol Methods. Feb 2019;24(1):1-19. [ CrossRef ] [ Medline ]
• McCoy CE. Understanding the intention-to-treat principle in randomized controlled trials. West J Emerg Med. Oct 2017;18(6):1075-1078. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Lee T, Shi D. A comparison of full information maximum likelihood and multiple imputation in structural equation modeling with missing data. Psychol Methods. Aug 2021;26(4):466-485. [ CrossRef ] [ Medline ]
• Grimm KJ, Ram N, Estabrook R. Growth Modeling: Structural Equation and Multilevel Modeling Approaches. New York City, NY. Guilford Press; 2016.
• Rosseel Y. Lavaan: an R package for structural equation modeling. J Stat Softw. 2012;48(2):1-36. [ CrossRef ]
• Zhao X, Lynch JGJ, Chen Q. Reconsidering Baron and Kenny: myths and truths about mediation analysis. J Consum Res. Aug 2010;37(2):197-206. [ CrossRef ]
• Stanton AL, Luecken LJ, MacKinnon DP, Thompson EH. Mechanisms in psychosocial interventions for adults living with cancer: opportunity for integration of theory, research, and practice. J Consult Clin Psychol. Apr 2013;81(2):318-335. [ CrossRef ] [ Medline ]
• Maxwell SE, Cole DA. Bias in cross-sectional analyses of longitudinal mediation. Psychol Methods. Mar 2007;12(1):23-44. [ CrossRef ] [ Medline ]
• Lachowicz MJ, Preacher KJ, Kelley K. A novel measure of effect size for mediation analysis. Psychol Methods. Jun 2018;23(2):244-261. [ CrossRef ] [ Medline ]
• Simes RJ. An improved Bonferroni procedure for multiple tests of significance. Biometrika. Dec 1986;73(3):751-754. [ CrossRef ]
• Feingold A. Confidence interval estimation for standardized effect sizes in multilevel and latent growth modeling. J Consult Clin Psychol. Feb 2015;83(1):157-168. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Teale Sapach MJ, Carleton RN. Self-compassion training for individuals with social anxiety disorder: a preliminary randomized controlled trial. Cogn Behav Ther. Jan 2023;52(1):18-37. [ CrossRef ] [ Medline ]
• Garland EL, Hanley AW, Goldin PR, Gross JJ. Testing the mindfulness-to-meaning theory: evidence for mindful positive emotion regulation from a reanalysis of longitudinal data. PLoS One. Dec 6, 2017;12(12):e0187727. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Guendelman S, Medeiros S, Rampes H. Mindfulness and emotion regulation: insights from neurobiological, psychological, and clinical studies. Front Psychol. Mar 06, 2017;8:220. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Curreri AJ, Farchione TJ, Sauer-Zavala S, Barlow DH. Mindful emotion awareness facilitates engagement with exposure therapy: an idiographic exploration using single case experimental design. Behav Modif. Jan 2022;46(1):36-62. [ CrossRef ] [ Medline ]
• Chin B, Lindsay EK, Greco CM, Brown KW, Smyth JM, Wright AG, et al. Psychological mechanisms driving stress resilience in mindfulness training: a randomized controlled trial. Health Psychol. Aug 2019;38(8):759-768. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Herscu O, Somer E, Federman A, Soffer-Dudek N. Mindfulness meditation and self-monitoring reduced maladaptive daydreaming symptoms: a randomized controlled trial of a brief self-guided web-based program. J Consult Clin Psychol. May 2023;91(5):285-300. [ CrossRef ] [ Medline ]
• Roemer L, Williston SK, Rollins LG. Mindfulness and emotion regulation. Curr Opin Psychol. Jun 2015;3:52-57. [ CrossRef ]
• Ruffault A, Czernichow S, Hagger MS, Ferrand M, Erichot N, Carette C, et al. The effects of mindfulness training on weight-loss and health-related behaviours in adults with overweight and obesity: a systematic review and meta-analysis. Obes Res Clin Pract. 2017;11(5 Suppl 1):90-111. [ CrossRef ] [ Medline ]
• Kazdin AE. Research Design in Clinical Psychology. London, UK. Pearson Education; 2016.
• Bakker D, Rickard N. Engagement in mobile phone app for self-monitoring of emotional wellbeing predicts changes in mental health: MoodPrism. J Affect Disord. Feb 2018;227:432-442. [ CrossRef ] [ Medline ]
• Zainal NH, Newman MG. A randomized controlled trial of a 14-day mindfulness ecological momentary intervention (MEMI) for generalized anxiety disorder. Eur Psychiatry. Jan 16, 2023;66(1):e12. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Sun S, Lin D, Goldberg S, Shen Z, Chen P, Qiao S, et al. A mindfulness-based mobile health (mHealth) intervention among psychologically distressed university students in quarantine during the COVID-19 pandemic: a randomized controlled trial. J Couns Psychol. Mar 2022;69(2):157-171. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Keng SL, Chin JW, Mammadova M, Teo I. Effects of mobile app-based mindfulness practice on healthcare workers: a randomized active controlled trial. Mindfulness (N Y). Sep 16, 2022;13(11):2691-2704. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Golden HL, Vosper J, Kingston J, Ellett L. The impact of mindfulness-based programmes on self-compassion in nonclinical populations: a systematic review and meta-analysis. Mindfulness. Sep 25, 2020;12(1):29-52. [ CrossRef ]
• Sanilevici M, Reuveni O, Lev-Ari S, Golland Y, Levit-Binnun N. Mindfulness-based stress reduction increases mental wellbeing and emotion regulation during the first wave of the COVID-19 pandemic: a synchronous online intervention study. Front Psychol. 2021;12:720965. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Buxton AE, Hofheinz C, Remmers C, Michalak J. When depressed-be mindful of and kind to yourself: self-compassion as mediator of change in a mindful depression treatment. Psychother Res. Feb 2024;34(2):182-194. [ CrossRef ] [ Medline ]
• Ștefan CA, Căpraru C, Szilágyi M. Investigating effects and mechanisms of a mindfulness-based stress reduction intervention in a sample of college students at risk for social anxiety. Mindfulness. Feb 14, 2018;9(5):1509-1521. [ CrossRef ]
• Sevel LS, Finn MT, Smith RM, Ryden AM, McKernan LC. Self-compassion in mindfulness-based stress reduction: an examination of prediction and mediation of intervention effects. Stress Health. Feb 2020;36(1):88-96. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Keng SL, Smoski MJ, Robins CJ, Ekblad AG, Brantley JG. Mechanisms of change in mindfulness-based stress reduction: self-compassion and mindfulness as mediators of intervention outcomes. J Cogn Psychother. Jan 01, 2012;26(3):270-280. [ CrossRef ]
• Goldin PR, Morrison AS, Jazaieri H, Heimberg RG, Gross JJ. Trajectories of social anxiety, cognitive reappraisal, and mindfulness during an RCT of CBGT versus MBSR for social anxiety disorder. Behav Res Ther. Oct 2017;97:1-13. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Gavrilova L, Zawadzki MJ. Examining how headspace impacts mindfulness mechanisms over an 8-week app-based mindfulness intervention. Mindfulness. Sep 22, 2023;14(9):2236-2249. [ CrossRef ]
• Schulte-Frankenfeld PM, Trautwein FM. App-based mindfulness meditation reduces perceived stress and improves self-regulation in working university students: a randomised controlled trial. Appl Psychol Health Well Being. Nov 2022;14(4):1151-1171. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Hsu KJ, Shumake J, Caffey K, Risom S, Labrada J, Smits JA, et al. Efficacy of attention bias modification training for depressed adults: a randomized clinical trial. Psychol Med. Mar 26, 2021;52(16):3865-3873. [ CrossRef ]
• Zainal NH, Hellberg SN, Kabel KE, Hotchkin CM, Baker AW. Cognitive behavioral therapy (CBT) plus attention bias modification (ABM) reduces anxiety sensitivity and depressive symptoms in panic disorder: a pilot randomized trial. Scand J Psychol. Aug 2023;64(4):390-400. [ CrossRef ] [ Medline ]
• Zainal NH, Newman MG. Mindfulness enhances cognitive functioning: a meta-analysis of 111 randomized controlled trials. Health Psychol Rev. Aug 30, 2023.:1-27. (forthcoming). [ CrossRef ] [ Medline ]
• Zainal NH, Newman MG. Executive functioning constructs in anxiety, obsessive-compulsive, post-traumatic stress, and related disorders. Curr Psychiatry Rep. Dec 2022;24(12):871-880. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Papenfuss I, Lommen MJ, Huisman M, Ostafin BD. Aversive response to uncertainty as a mediator for the effect of a mindfulness intervention on symptoms of anxiety. Int J Psychophysiol. Sep 2022;179:30-42. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Matos M, Duarte C, Duarte J, Pinto-Gouveia J, Petrocchi N, Gilbert P. Cultivating the compassionate self: an exploration of the mechanisms of change in compassionate mind training. Mindfulness. Oct 21, 2021;13(1):66-79. [ CrossRef ]
• Bauer DJ, Preacher KJ, Gil KM. Conceptualizing and testing random indirect effects and moderated mediation in multilevel models: new procedures and recommendations. Psychol Methods. Jun 2006;11(2):142-163. [ CrossRef ] [ Medline ]
• Kazdin AE. Understanding how and why psychotherapy leads to change. Psychother Res. Jul 22, 2009;19(4-5):418-428. [ CrossRef ] [ Medline ]
• Torous J, Lipschitz J, Ng M, Firth J. Dropout rates in clinical trials of smartphone apps for depressive symptoms: a systematic review and meta-analysis. J Affect Disord. Feb 15, 2020;263:413-419. [ CrossRef ] [ Medline ]
• Bieber M, Görgülü E, Schmidt D, Zabel K, Etyemez S, Friedrichs B, et al. Effects of body-oriented yoga: a RCT study for patients with major depressive disorder. Eur Arch Psychiatry Clin Neurosci. Oct 2021;271(7):1217-1229. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Samantaray NN, Nath B, Behera N, Mishra A, Singh P, Sudhir P. Brief cognitive behavior group therapy for social anxiety among medical students: a randomized placebo-controlled trial. Asian J Psychiatr. Jan 2021;55:102526. [ CrossRef ] [ Medline ]
• Van Doren N, Zainal NH, Newman MG. Cross-cultural and gender invariance of emotion regulation in the United States and India. J Affect Disord. Dec 01, 2021;295:1360-1370. [ FREE Full text ] [ CrossRef ] [ Medline ]
• Koszycki D, Thake J, Mavounza C, Daoust JP, Taljaard M, Bradwejn J. Preliminary investigation of a mindfulness-based intervention for social anxiety disorder that integrates compassion meditation and mindful exposure. J Altern Complement Med. May 2016;22(5):363-374. [ CrossRef ] [ Medline ]

Abbreviations

Edited by J Torous; submitted 16.10.23; peer-reviewed by L LaFreniere, N Van Doren; comments to author 27.11.23; revised version received 27.12.23; accepted 12.01.24; published 19.04.24.

©Nur Hani Zainal, Hui Han Tan, Ryan Yee Shiun Hong, Michelle Gayle Newman. Originally published in JMIR Mental Health (https://mental.jmir.org), 19.04.2024.

This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Mental Health, is properly cited. The complete bibliographic information, a link to the original publication on https://mental.jmir.org/, as well as this copyright and license information must be included.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 20 April 2024

Associations between autistic traits, depression, social anxiety and social rejection in autistic and non-autistic adults

• Emine Gurbuz 1 , 4 ,
• Deborah M. Riby 2 , 5 ,
• Mikle South 3 &
• Mary Hanley 2 , 5  

Scientific Reports volume  14 , Article number:  9065 ( 2024 ) Cite this article

Metrics details

• Human behaviour
• Neuroscience

Autistic people frequently experience negative judgements from non-autistic people, often fuelled by misconceptions that autistic people lack empathy. Understanding responses to negative social judgement among autistic people is crucial because of the potential negative impact on wellbeing and future interactions. We investigated the role of autistic traits, social anxiety, and depression on behavioural indices of social rejection in 20 autistic (AUT; 11 males) and 40 non-autistic (N-AUT; 21 males) university students. Participants completed the Social Judgement Task (SJT) where they predicted whether they were liked by another person, then received feedback on whether those evaluations were correct. Participants also completed an Age Judgement Task (AJT) where they estimated the age of the pictured person. The AUT group had lower positive expectation scores, meaning less tendency to predict being liked. Across the whole sample, higher social anxiety predicted greater tendency to anticipate rejection from others, not autistic traits. These findings suggest early experiences of rejection might lead to a negative self-bias in autistic people and emphasise the importance of using a transdiagnostic approach by showing that social anxiety rather than autistic traits is associated with expectation of social rejection.

Introduction

The capacity to perceive and evaluate social cues is a vital component of social interaction and can play a critical role in making judgements about the people we interact with including trustworthiness and approachability 1 , 2 . Neurotypical people make frequent judgments of the faces of social partners that guide social interactions, and equally those partners make judgements to guide their own behaviour – it is a two-way process 3 . Understanding how we respond to social judgements made by others is important for understanding real world social behaviours and experiences. When this judgement is positive, we experience social acceptance, but when this judgement is negative, we experience social rejection. Frequent feelings of social rejection can impact mental health, such as contributing to increased depression 4 and anxiety 5 . Studies of social rejection in neurotypical adults using behavioural measures have shown that people are more likely to predict being liked by others even when others are unfamiliar to us 6 , 7 , 8 , suggesting that we like others to have a positive opinion of us and it makes us feel good.

Exploring whether social rejection experiences are the same or different for autistic adults is important for understanding broader social experiences of the autistic community. Many autistic young adults have reported experiencing loneliness, social isolation, peer rejection, bullying, stigma, and lack of social support networks 9 , 10 , 11 , 12 , 13 , 14 . These negative social experiences can create a risk for poor mental health. Indeed, frequent experiences of social rejection and loneliness 15 contribute to high rates of depression and social anxiety in autistic individuals 16 , 17 . Given the moderating role of rejection sensitivity in developing depression and anxiety in neurotypical individuals 18 and early life events of rejection leading to development of depression via changing the neurobiological responses 4 , it is crucial to investigate the associations between social rejection and mental health, particularly anxiety and depression. In addition, stigma and misconceptions about autistic people are very common among non-autistic community, which can also impact mental health of autistic people as well as social connectedness with others 19 . For example, due to the stereotypical assumption of reduced social interest, non-autistic individuals might have lower expectations of inclusion and social reciprocity in their interactions with autistic people 20 . In turn, this leads to higher expectations of negative judgements in autistic adults, including being rejected by others.

There is limited research on what factors interplay with experiences of social rejection, more specifically expectation of social rejection from others. Previous research suggested that adverse life events including rejection and bullying might explain higher mental health challenges in autistic people 21 . For example, autism acceptance from society predicted depression, but not anxiety in autistic adults 20 . Another study using a longitudinal design tested the interplay between autistic traits, self-compassion, and symptoms of anxiety and depression in autistic and non-autistic adults 22 . One of the components of self-compassion is self-kindness which means to treat oneself with kindness and support rather than being self-critical or self-judgemental 23 . The results showed that low self-compassion in autistic adults predicted later anxiety and depression, not the other way around. These studies suggested that poor mental health in autistic people might result from adverse life events including lack of acceptance from others, self-blame and negative self-judgement. However, it is not clear what autistic individuals think about how they are perceived by others (e.g. likeability) and what contributes to greater expectation of rejection (higher expectations of being unliked by others). Considering high co-occurrences of anxiety and depression together with experiences of social rejections in autistic individuals, it is important to understand the relationships between autistic traits, social anxiety or depression in both autistic and non-autistic individuals transdiagnostically, regardless of diagnosis.

One paradigm that has been used to empirically study the experience of social rejection is the Social Judgement Task (SJT) 24 . In the SJT, participants send their profile picture to the researcher and are told that the picture will be judged by a panel of unfamiliar adults. When the participants come to the lab, they are presented with pictures of other people who they are told have already made a judgement about them. The participant's task is then to make a judgment on whether this person liked them or not. This is followed by feedback indicating either social acceptance or social rejection by the other person. However, nobody has actually seen their profile picture and all feedback is randomly generated. To make sure that this response is specific to expectation of social rejection and not to general tendency to say “No”, participants also complete a control task where they judge the age of the same faces/people. Previous fMRI research using this paradigm found that adults show differential neural responses such that ventral anterior cingulate cortex (vACC) was sensitive to unexpected negative social feedback (social rejection) while dorsal anterior cingulate cortex (dACC) was sensitive to positive social feedback (social acceptance) and non-social negative feedback 24 .

Despite evidence showing that some autistic individuals can experience social rejection more frequently than their non-autistic peers, little is known about the relationship between expectation of rejection and mental health in autistic and non-autistic people. This is the first study to investigate behavioural indices of social rejection in autistic and non-autistic young adults using the SJT and the role of mental health and autistic traits in expectation of social rejection in a sample including people with high autistic traits.

Study predictions

Non-autistic participants were expected to make comparatively more predictions of being liked while autistic participants were expected to make comparatively more predictions of being unliked/rejected.

We expected that higher autistic traits, social anxiety and depression symptoms would be associated with greater expectation of social rejection in all participants.

Clinical and demographic variables

The AUT group ( M  = 34.65) had higher AQ scores than the N-AUT group ( M  = 16.46) ( t (57) =  − 8.208, p  < 0.001, d  = 2.21) (See Table 1 ). The DASS-21 total scores and its subscales of depression, anxiety , and stress were calculated for each participant. The AUT group had significantly higher DASS-21 total scores compared to the N-AUT group, t (57) =  − 3.121, p  = 0.004, d  = 0.92. When each subscale was compared, the autistic students scored significantly higher on anxiety, t (57) =  − 3.058, p  = 0.005, d  = 0.90, and stress, t (57) =  − 4.035, p  < 0.001, d  = 1.04, but not on the depression subscale, t (57) =  − 1.603, p  = 0.120, d  = 0.47. Similarly, the AUT group reported significantly higher social anxiety scores measured with LSAS compared to N-AUT, t (56) =  − 6.254, p  < 0.001, d  = 1.90. Specifically, the LSAS subscales of fear, t (56) =  − 6.572, p  < 0.001, d  = 1.97, and avoidance, t (56) =  − 5.506, p  < 0.001, d  = 1.67, were significantly higher for autistic participants compared to their non-autistic peers.

Behavioural results

Social judgement task.

One-sample t -tests to check whether the expectation of rejection (negative expectation score) or acceptance (positive expectation score) differed significantly from chance (50%) found no expectation bias in either group (see Table 2 ; AUT; t (19) =  − 1.879, p  = 0.90, d  = 0.40, N-AUT; t (38) = 0.754, p  = 0.456, d  = 0.12). However, on average, the AUT group had lower positive expectation scores (lower tendency to predict being liked; group mean 44.25%) compared to the N-AUT group (group mean 51.45%) and this group difference was significant, t (57) = 2.038, p  = 0.046, d  = 0.54. The exact analysis with the responses from the AJT can be found in the Supplementary information ( S1 ). This analysis found no differences between “Yes'' and “No” responses in AJT in the AUT group suggesting that the tendency to give more “No” responses in the SJT was task specific and cannot be explained by autistic participants having lesser tendency to answer affirmatively to yes/no questions.

Individual differences in experiences of social rejection

To test the hypothesis that higher autistic traits, social anxiety and depression symptoms would be associated with higher expectation of social rejection in all participants, one-tailed Pearson correlations were performed to check the linear relationships between behavioural responses from the SJT and autistic traits, social anxiety, and depression scores across participants. As can be seen in Table 3 , negative expectation scores indicated by percentage of ‘No’ responses in the SJT were significantly correlated with autistic traits, r  = 0.356, p  = 0.006 (moderate positive correlation), social anxiety, r  = 0.507, p  < 0.001 (strong positive correlation), and depression scores, r  = 0.374, p  = 0.004 (moderate positive correlation). Participants with a higher negative expectation score (higher tendency to predict being rejected by others) were more likely to report higher autistic traits, higher social anxiety, and more depression symptoms.

Hierarchical regression was conducted with AQ, DASS-21 depression scores and LSAS as predictors of negative expectation scores. Tests to see if the data met the assumption of collinearity indicated that multicollinearity was not a concern (VIF > 1.000). The data met the assumption of independent errors (Durbin-Watson value = 2.159). The histogram and P-P plot of standardised residuals indicated that the data contained normally distributed errors. The overall regression model was significant, F (3, 57) = 7.669, p  < 0.001, R i  = 0.299. However, only social anxiety was a significant predictor, β  = 0.408, t (57) = 2.212, p  = 0.031. Thus, higher self-reported social anxiety significantly predicted higher tendency to expect being rejected by others. Adding depression symptoms and autistic traits did not improve the model as they were not significant predictors of negative expectation scores (DASS-21 depression subscale; β  = 0.361, t (57) = 1.791, p  = 0.079, AQ; β  = -0.014, t (57) =  − 0.075, p  = 0.941).

This study was the first to look at expectation of social rejection in autistic and non-autistic participants and its relation to autistic traits, social anxiety, and depression in a sample consisting of both autistic and non-autistic adults. The findings reveal some important new insights that are particularly important for understanding the experiences of autistic people. First, there was not a bias towards being liked than unliked by others in non-autistic participants, which was in contrast with previous findings 6 , 7 , 25 , 26 . However, when compared to autistic participants, non-autistic participants did predict being liked more, even though autistic participants did not make a strong prediction that others would unlike them. The lack of positive or negative self-evaluation bias in either group could be explained by individual differences in self-evaluation bias in the current sample, as discussed below.

Autistic participants expected to be liked less by others indicated by greater expectations of social rejection than non-autistic participants. These results imply that experiences of social rejection in autistic people might be quantitatively different than non-autistic individuals due to more common experiences of social rejection in autistic people. Given the prevalence of rejection and negative judgements outlined in the introduction, and that only 7% of autistic people feel accepted by society as an autistic person 20 , it is not surprising that autistic people anticipate more rejection, leading to potential increased social withdrawal and poorer mental health over time. Moreover, recent empirical evidence showed that non-autistic individuals made more negative judgements about autistic individuals 27 , 28 and lower interactional rapport was reported between autistic and non-autistic adults 29 . Such research might explain why autistic people approach new situations thinking that they will be rejected again, due to many years of negative experience till adulthood. Longitudinal research is required to understand how negative childhood experiences including peer rejection, stigma, and stress influence future social experiences and interactions with others (e.g. friendships) in autistic people.

Next, individual differences in expectation of social rejection were investigated. Behavioural negative expectation scores (the negative self-evaluation bias) were found to be associated with higher self-reported autistic traits, social anxiety, and depression symptoms across participants. Subsequent regression analyses found that only social anxiety predicted higher expectations of rejection across participants, irrespective of autistic traits, suggesting that individuals who have higher social anxiety might be more prone to expect being rejected, irrespective of whether or not they have a diagnosis of autism. This relationship between social anxiety and negative self-evaluation bias has also been shown in individuals with Social Anxiety Disorders 30 , 31 . As discussed above, the negative self-evaluation bias among socially anxious individuals could be due to learned experiences from frequent social rejection in the past. According to the Compassionate Brain Theory, lower self-compassion including negative self-perception in individuals with high autistic traits might be a result of frequent negative experiences throughout one’s life 22 . Moreover, social anxiety and negative self-perception (e.g. “I am undesirable”) may lead to withdrawal from social interactions resulting in further isolation or depression 4 , 32 .

Social anxiety is not only observed autistic people, and therefore it should be taken into consideration in understanding transdiagnostic negative social experiences, particularly social rejection. However, given the higher levels of social anxiety in autistic people 33 and apparent similarities in clinical presentations of psychiatric disorders in autistic and non-autistic people 34 , it is important to examine the relationship between social anxiety and social experiences of rejection and isolation in autistic populations as well as transdiagnostically. Another future direction for research is to compare a group of non-autistic participants matched on social anxiety to the autistic participants in order to understand whether the current patterns of associations are related to autistic traits or social anxiety per se.

There are several methodological considerations for the current study. The first methodological issue is what the social and age judgement tasks meant to autistic participants. For example, while the age judgement task was used intended as a control, non-social feedback condition, similarities in the task could lead autistic participants to find it equally social: all stimuli were still faces and both tasks were performed in front of an experimenter, meaning that responses to both tasks was provided essentially in a social situation. Future research may explore the validity of experimental paradigms for autistic people and not assume that tasks mean the same thing for autistic and non-autistic participants. The AUT sample was smaller than the N-AUT sample and autistic participants self-reported their autism diagnosis; larger, verified autism samples will provide more insight on individual and transdiagnostic group differences. The role of gender in responses to social rejection was not examined in the current study but is important for future research especially in the context of significant current interest in autistic females. That said, inclusion of non-binary genders, who might be more vulnerable to rejection and stigmatisation is necessary 35 . . Adding a state measure of social anxiety after the SJT could give more insight into experiences of immediate distress following social rejection in non-autistic and autistic participants and add understanding of individual differences and potential mechanisms underlying the experiences of social rejection. Information about co-occurring diagnosis for autistic people were obtained, however similar information was not collected for the non-autistic group, which is a limitation. Lastly, longitudinal data is needed to test whether social anxiety or expectation of rejection comes first, especially given the bidirectional relationship between the two as suggested in previous research 20 , 22 .

This was the first study to administer the Social Judgement Task to autistic young adults, using behavioural responses to probe experiences of social rejection. Autistic participants expected more rejection than their non-autistic peers, and social anxiety was a significant predictor of higher expectations to be rejected in all participants, irrespective of autism diagnosis. Social anxiety due to early negative experiences of rejection rather than difficulties associated with autism per se might explain why autistic people predict further rejection from others. Understanding the reasons behind higher expectations of social rejection in autistic individuals and whether it is associated with negative past experiences is needed to improve well-being of autistic individuals. Studying the long-term consequences of adverse childhood experiences, rejection, and stigma in autistic people is crucial to create an inclusive and welcoming society.

Participants

Forty non-autistic participants (N-AUT; 21 males, mean age = 22.83 years, SD = 4.13) and 20 autistic participants (AUT; 11 males, mean age = 23.58 years, SD = 4.27) participated in this study. Autistic participants were recruited via disability support services in the UK universities and they all had an official diagnosis of autism to be registered in support services. Non-autistic participants were recruited from UK universities via flyers distributed across the campus. Participants received £10 for their participation. The groups were matched in terms of chronological age, t (58) = 0.652, p  = 0.52, and gender, x 2 (1, N  = 60) = 0.033, p  = 0.86. Thirteen of the 20 autistic participants (65%) self-reported co-occurring diagnoses with eight (40%) experiencing co-occurring mental health issues. Autistic traits were measured using the Autism Quotient 36 .

Ethics and inclusion statement

Ethical approval was provided by Durham University Psychology Department Ethics Sub-committee prior to the commencement of the study. All methods were carried out in accordance with relevant guidelines and regulations. Informed consent was obtained from all participants for study participation and publication of their data in an online open-access publication. All information obtained during the study are kept confidential and the data presented in the tables and figures does not include any identifiable information.

Task descriptions and procedure

Adapted versions of Social and Age Judgement Tasks used by Gunther Moor, Crone and Van Der Molen 6 were administered using E-prime 3.0 software 37 . Participants were told the task focused on forming first impressions among peers from multiple universities. First, they were asked to send a profile picture (e.g. a neutral expression head and shoulder shot) to the researcher approximately two weeks before the experiment. Participants were then told that their picture would be sent to a panel of students from other universities who would form first impression of them based on the picture. When participants arrived for the study, they were informed that they would first estimate the judgements of those who had already seen their profile picture and made a judgement whether they liked them or not (the SJT). Then, they were told that it was their turn to make judgments about the same students, and this judgement would be about the persons’ age (the AJT). Therefore, all participants completed the SJT first and then the AJT. Participants completed 4 practice trials followed by 121 trials in total for each task, with a break after the 60th trial. The feedback was randomly created by the computer programme such that there were 50% ‘Yes’ and 50% ‘No’ feedback trials in both tasks (see below for further explanation). Behavioural data of percentage of ‘Yes’ and ‘No’ responses given by the participants were collected throughout the two tasks. At the end of the experiment, participants were asked whether they believed in the cover story, and they all confirmed that they did.

In both tasks, the same pictures were presented. Instead of taking our own pictures as done in the original study, we used pictures of young adults taken from the Chicago Face Database 38 . The stimuli consisted of 121 face images (measuring 3.9 × 4.5 cm) presented on the computer screen. The age (inter-rater reliability of the ratings; α = 0.896) and attractiveness (α = 0.998) of the pictures were all rated, and the pictures standardized by the researchers who developed the database. The stimuli consist of 60 female and 60 male faces with a neutral expression on a white background (102 Caucasian, 4 African American, 8 Asian, and 6 Hispanic).

Social judgement task (SJT)

The timeline of the SJT paradigm can be found in Fig. 1. Each trial started with a fixation point for 1000 ms. Then the picture (cue) was presented for 3000 ms and it stayed on the screen until the end of the trial. During the presentation of the picture, the participants responded ‘YES’ by pressing on ‘1’ to indicate if they thought the person on the picture liked them or ‘NO’ by pressing ‘3’ to indicate if they thought the person on the picture did not like them. If the participants did not respond within 3000 ms, the feedback ‘too late’ was given. The participants’ response was then shown on the left side of the picture. After a delay of 1000 ms, the participants were provided with feedback on the supposed response from the person in the picture, with a ‘YES’ (the person on the picture liked them— acceptance ) or ‘NO’ feedback (the person on the picture did not like them— rejection ) presented on the left side of the picture. The feedback remained on the screen for 2000 ms. The next trial began with the fixation cross.

Age judgement task (AJT)

The AJT was very similar to SJT except for the type of judgement participants needed to make; the age of the person presented in the picture. The participants were presented with the same face stimuli for 3000 ms but this time they had to guess whether the person in the picture was 21 years old or older. The participants responded ‘YES’ by pressing on ‘1’ to indicate that the person on the picture was 21 years or older or ‘NO’ by pressing ‘3’ to indicate that the person on the picture was younger than 21 years old. 1000 ms after the participant’s response, the ‘YES’ (the person on the picture is 21 years or older) or ‘NO’ feedback (the person on the picture is not 21 years or older) was provided on the screen for 2000 ms (See Fig.  1 ).

The timeline of SJT and AJT. Note. After a fixation point for 1000ms, the participant answers the question “Do I think this person liked me?” in SJT or “Is this person 21 years old or older?” in AJT. The response of the participant “YES” or “NO” is presented during the delay for 1000ms.

Self-report measures

Following the SJT and AJT, participants were asked to complete the Autism Quotient, Liebowitz Social Anxiety Scale, and Depression Anxiety Stress Scale-21 to measure their autistic traits, social anxiety, and depression respectively.

The Autism Quotient (AQ) is a 50-item self-report questionnaire to measure autistic traits in general population with IQ > 70 36 . The five subscales are social skill, attention switching, attention to detail, communication and imagination. Participants respond by choosing one of the four options for each item; ‘definitely agree’, ‘slightly agree’, ‘slightly disagree’, ‘definitely disagree’. Items are scored dichotomously (‘0’ or ‘1’) and the total scores range from 0 to 50, with higher scores indicating more autistic traits. In the present sample, the AQ demonstrated a high internal consistency (coefficient α = 0.930); AUT (α = 0.854); N-AUT (α = 0.872).

Liebowitz Social Anxiety Scale (LSAS) contains 24 items; 13 of which measure performance anxiety and 11 measure social interaction anxiety 39 . Participants rate each item for fear (0–3; none-severe) and avoidance (0–3; never-usually). In previous research, the self-report version of LSAS (LSAS-SR) has been reported to have good psychometric properties indicated by strong test–retest reliability ( α  = 0.83), internal consistency ( α  = 0.95), and convergent ( α  = 0.88 to 0.94) and discriminant validity 40 . When compared to the clinician-administered version (LSAS-CA), the LSAS-SR has been shown to be a valid measure with high internal consistency ( α  = 0.95) and identical subscale intercorrelations ( α  = 0.71–0.91) 41 and it showed high internal consistency in autistic adults ( α  = 96) 42 . One participant in the AUT group did not complete the questionnaire. For the current sample with 19 participants in the AUT group and 40 participants in the N-AUT group, internal consistency of LSAS total scores was α  = 0.959 ( avoidance subscale; α  = 0.918 and fear subscale; α  = 0.929).

Depression Anxiety Stress Scale-21 (DASS-21) 43 is a 21-item self-report measure of depression, anxiety and stress. On a 4-point Likert scale (from 1 =  did not apply to me at all to 4 =  applied to me very much or most of the time ), participants rate each item based on its applicability to their life experiences over the past week, with higher scores indicating more severe symptoms. When administered to autistic adults, the DASS-21 has been reported to show good reliability and validity, with Cronbach’s alpha of 0.89 for the depression subscale, 0.83, for the anxiety subscale, and 0.86 for the stress subscale 44 . In the current sample, one N-AUT participant with a DASS-21 score of 112 (three SD above the mean) was excluded from further analysis. The internal consistency of DASS-21 total scores with 20 autistic and 39 N-AUT participants was α  = 0.924 ( anxiety subscale α  = 0.834, depression subscale α  = 0.793 , anxiety subscale α  = 0.836).

Data analysis plan

Behavioural data consisted of the percentage of ‘Yes’ and ‘No’ responses given by the participants. In the SJT, if the participant made more than 50% of ‘Yes’ predictions, it would indicate a positive expectation score (that the participant expected to be liked) and if the participant made more than 50% of ‘No’ predictions, it would indicate a negative expectation score (that the participant expected not to be liked). One-sample t -tests were carried out with each group separately to check whether the expectation score differed significantly from 50% 45 . Group differences (AUT; N-AUT) in negative expectation scores were calculated by using independent samples t -test. Bonferroni correction was used for post hoc analyses and Huynh–Feldt corrections for violations of the assumptions of sphericity were used when necessary 46 .

Pearson bivariate correlations were calculated to check linearity between each predictor (AQ, LSAS, and DASS-21 depression scores) and the outcome variable (between negative expectation scores) across participants. After correction for multiple correlations and checking assumptions for multivariate regression, hierarchical regression analyses were performed to examine whether AQ, LSAS, and DASS-21 depression scores would predict behavioural responses to social rejection across participants. In the regression analysis, autistic traits from AQ were entered in Step 1, depression scores from DASS-21 were entered in Step 2 and social anxiety scores from LSAS were entered in Step 3. Analysis has been conducted using SPSS Statistics v27.

An a priori power analysis was conducted using G*Power version 3.1.9.7 47 to determine the minimum sample size required to test the study hypothesis. Results indicated the required sample size to achieve 80% power for detecting a medium effect, at a significance criterion of α = 0.05, was N  = 55 for linear multiple regression. Thus, the obtained sample size of N  = 58 is adequate to test the hypothesis.

Data availability

All data generated and/or analyzed during this study are available from the corresponding author (E.G.) on request.

Willis, J. & Todorov, A. First impressions: Making up your mind after a 100-Ms exposure to a face. Psychol. Sci. 17 , 592–598 (2006).

Article   PubMed   Google Scholar  

Todorov, A., Mende-Siedlecki, P. & Dotsch, R. Social judgments from faces. Curr. Opin. Neurobiol. 23 , 373–380 (2013).

Article   CAS   PubMed   Google Scholar  

Zebrowitz, L. A. First impressions from faces. Curr. Dir. Psychol. Sci. 26 , 237–242 (2017).

Article   PubMed   PubMed Central   Google Scholar  

Slavich, G. M., O’Donovan, A., Epel, E. S. & Kemeny, M. E. Black sheep get the blues: A psychobiological model of social rejection and depression. Neurosci. Biobehav. Rev. 35 , 39–45 (2010).

Baumeister, R. F. & Tice, D. M. Point-counterpoints: Anxiety and social exclusion. J. Soc. Clin. Psychol. 9 , 165–195 (1990).

Article   Google Scholar  

Gunther Moor, B., Crone, E. A. & van der Molen, M. W. The Heartbrake of social rejection: Heart rate deceleration in response to unexpected peer rejection. Psychol. Sci. 21 , 1326–1333 (2010).

Van Der Molen, M. et al. Fear of negative evaluation modulates electrocortical and behavioral responses when anticipating social evaluative feedback. Front. Hum. Neurosci. 7 , 936 (2014).

PubMed   PubMed Central   Google Scholar  

van der Veen, F. M., van der Molen, M. J. W., van der Molen, M. W. & Franken, I. H. A. Thumbs up or thumbs down? Effects of neuroticism and depressive symptoms on psychophysiological responses to social evaluation in healthy students. Cogn. Affect. Behav. Neurosci. 16 , 836–847 (2016).

Chown, N. & Beavan, N. Intellectually capable but socially excluded? A review of the literature and research on students with autism in further education. J. Furth. High. Educ. 36 , 477–493 (2012).

Van Hees, V., Moyson, T. & Roeyers, H. Higher education experiences of students with autism spectrum disorder: Challenges, benefits and support needs. J. Autism Dev. Disord. 45 , 1673–1688 (2015).

White, S. W., Ollendick, T. H. & Bray, B. C. College students on the autism spectrum: Prevalence and associated problems. Autism 15 , 683–701 (2011).

White, S. W. & Roberson-Nay, R. Anxiety, social deficits, and loneliness in youth with autism spectrum disorders. J. Autism Dev. Disord. 39 , 1006–1013 (2009).

DeNigris, D. et al. Bullying and identity development: Insights from autistic and non-autistic college students. J. Autism Dev. Disord. 48 , 666–678 (2018).

Pearson, A., Rose, K. & Rees, J. ‘I felt like I deserved it because I was autistic’: Understanding the impact of interpersonal victimisation in the lives of autistic people. Autism 27 , 500–511 (2023).

McLeod, J. D., Meanwell, E. & Hawbaker, A. The experiences of college students on the autism spectrum: A comparison to their neurotypical peers. J. Autism Dev. Disord. 49 , 2320–2336 (2019).

Jackson, S. L. J., Hart, L. & Volkmar, F. R. Preface: Special issue—college experiences for students with autism spectrum disorder. J. Autism Dev. Disord. 48 , 639 (2018).

Storch, E. A. et al. Peer victimization in youth with autism spectrum disorders and co-occurring anxiety: Relations with psychopathology and loneliness. J. Dev. Phys. Disabil. 24 , 575–590 (2012).

Gao, S., Assink, M., Cipriani, A. & Lin, K. Associations between rejection sensitivity and mental health outcomes: A meta-analytic review. Clin. Psychol. Rev. 57 , 59–74 (2017).

Turnock, A., Langley, K. & Jones, C. R. G. Understanding stigma in autism: A narrative review and theoretical model. Autism Adulthood 4 , 76–91 (2022).

Cage, E., Di Monaco, J. & Newell, V. Experiences of autism acceptance and mental health in autistic adults. J. Autism Dev. Disord. 48 , 473–484 (2018).

Lai, M.-C. Mental health challenges faced by autistic people. Nat. Hum. Behav. 7 , 1620–1637 (2023).

Galvin, J., Howes, A. & Richards, G. Longitudinal associations between autistic traits, self-compassion, anxiety and depression in autistic and non-autistic adults without intellectual disability. J. Autism Dev. Disord. https://doi.org/10.1007/s10803023-06157-6 (2023).

Neff, K. D. Self-compassion, self-esteem, and well-being. Soc. Pers. Psychol. Compass 5 , 1–12 (2011).

Somerville, L. H., Heatherton, T. F. & Kelley, W. M. Anterior cingulate cortex responds differentially to expectancy violation and social rejection. Nat. Neurosci. 9 , 1007–1008 (2006).

Gunther Moor, B., Bos, M. G. N., Crone, E. A. & van der Molen, M. W. Peer rejection cues induce cardiac slowing after transition into adolescence. Dev. Psychol. 50 , 947–955 (2014).

Van der Veen, F. M., van der Molen, M. W., Sahibdin, P. P. & Franken, I. H. A. The heart-break of social rejection versus the brain wave of social acceptance. Soc. Cogn. Affect. Neurosci. 9 , 1346–1351 (2014).

Grossman, R. B., Mertens, J. & Zane, E. Perceptions of self and other: Social judgments and gaze patterns to videos of adolescents with and without autism spectrum disorder. Autism 23 , 846–857 (2019).

Sasson, N. J. et al. Neurotypical peers are less willing to interact with those with autism based on thin slice judgments. Sci. Rep. 7 , 40700 (2017).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Crompton, C. J. et al. Neurotype-matching, but not being autistic, influences self and observer ratings of interpersonal rapport. Front. Psychol. 11 , 586171 (2020).

Harrewijn, A., van der Molen, M. J. W., van Vliet, I. M., Tissier, R. L. M. & Westenberg, P. M. Behavioral and EEG responses to social evaluation: A two generation family study on social anxiety. NeuroImage Clin. 17 , 549–562 (2018).

van der Molen, M. J. W., Harrewijn, A. & Westenberg, P. M. Will they like me? Neural and behavioral responses to social-evaluative peer feedback in socially and non-socially anxious females. Biol. Psychol. 135 , 18–28 (2018).

Downey, G. & Feldman, S. I. Implications of rejection sensitivity for intimate relationships. J. Pers. Soc. Psychol. 70 , 1327–1343 (1996).

Maddox, B. B. & White, S. W. Comorbid social anxiety disorder in adults with autism spectrum disorder. J. Autism Dev. Disord. 45 , 3949–3960 (2015).

Rodriguez-Seijas, C. et al. A transdiagnostic model of psychiatric symptom co- occurrence and autism spectrum disorder. Autism Res. 13 , 579–590 (2020).

Khudiakova, V. & Chasteen, A. L. The experiences of stigmatization and discrimination in autistic people of different genders and sexualities. J. Interpers. Relat. Intergroup Relat. Identity 15 , (2022).

Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J. & Clubley, E. The autism spectrum quotient (AQ): Evidence from asperger syndrome/high-functioning autism, males and females, scientists and mathematicians. J. Autism Dev. Disord. 31 , 5–17 (2001).

Psychology Software Tools, Inc. [E-Prime Go]. (2020). Retrieved from https://support.pstnet.com/

Ma, D. S., Correll, J. & Wittenbrink, B. The Chicago face database: A free stimulus set of faces and norming data. Behav. Res. Methods 47 , 1122–1135 (2015).

Liebowitz, M. R. Social phobia. Mod. Probl. Pharmacopsychiatry 22 , 141–173 (1987).

Baker, S. L., Heinrichs, N., Kim, H.-J. & Hofmann, S. G. The Liebowitz social anxiety scale as a self-report instrument: A preliminary psychometric analysis. Behav. Res. Ther. 40 , 701–715 (2002).

Fresco, D. M. et al. The Liebowitz social anxiety scale: A comparison of the psychometric properties of self-report and clinician-administered formats. Psychol. Med. 31 , 1025–1035 (2001).

Boulton, K. A. & Guastella, A. J. Social anxiety symptoms in autism spectrum disorder and social anxiety disorder: Considering the reliability of self-report instruments in adult cohorts. Autism Res. 14 , 2383–2392 (2021).

Lovibond, P. F. & Lovibond, S. H. The structure of negative emotional states: Comparison of the depression anxiety stress scales (DASS) with the beck depression and anxiety inventories. Behav. Res. Ther. 33 , 335–343 (1995).

Nah, Y.-H., Brewer, N., Young, R. L. & Flower, R. Brief report: Screening adults with autism spectrum disorder for anxiety and depression. J. Autism Dev. Disord. 48 , 1841–1846 (2018).

Kortink, E. D., Weeda, W. D., Crowley, M. J., Gunther Moor, B. & van der Molen, M. J. W. Community structure analysis of rejection sensitive personality profiles: A common neural response to social evaluative threat. Cogn. Affect. Behav. Neurosci. 18 , 581–595 (2018).

Vasey, M. W. & Thayer, J. F. The continuing problem of false positives in repeated measures ANOVA in psychophysiology: A multivariate solution. Psychophysiology 24 , 479–486 (1987).

Faul, F., Erdfelder, E., Buchner, A. & Lang, A.-G. Statistical power analyses using G*Power 3.1: Tests for correlation and regression analyses. Behav. Res. Methods 41 , 1149–1160 (2009).

Download references

The study was fully funded by a doctoral scholarship from Durham University to the first author.

Author information

Authors and affiliations.

Department of Psychology, University of Portsmouth, King Henry Building, King Henry I Street, Portsmouth, PO1 2DY, UK

Emine Gurbuz

Department of Psychology, Durham University, Durham, UK

Deborah M. Riby & Mary Hanley

Emory University School of Medicine and Emory Brain Health Center, Atlanta, USA

Mikle South

Centre for Interaction, Development and Diversity, University of Portsmouth, Portsmouth, UK

Centre for Neurodiversity & Development, Durham University, Durham, UK

You can also search for this author in PubMed   Google Scholar

Contributions

E.G. D.R and M.H. developed the research rationale and questions, then collaborated with M.S. (visiting research fellow at Durham University at the time of the research) who contributed to research aims and development and adjustment of the experimental tasks. E.G. conducted the data analysis and interpreted the findings with the input from D.B., M.S., and M.H. E.G. presented the results in both national and international conferences. E.G. drafted the manuscript and D.R. and M.H. reviewed the draft and contributed to revising the draft revisions and the final document. M.S. reviewed the final draft. All authors reviewed and approved the final version. This manuscript has been submitted solely to this journal and is not published, in press, or submitted elsewhere.

Corresponding author

Correspondence to Emine Gurbuz .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Supplementary information., rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Gurbuz, E., Riby, D.M., South, M. et al. Associations between autistic traits, depression, social anxiety and social rejection in autistic and non-autistic adults. Sci Rep 14 , 9065 (2024). https://doi.org/10.1038/s41598-024-59532-3

Download citation

Received : 17 August 2023

Accepted : 11 April 2024

Published : 20 April 2024

DOI : https://doi.org/10.1038/s41598-024-59532-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Ten years of researches on generalized anxiety disorder (GAD): a scientometric review

• Published: 11 April 2024

Cite this article

• Ying Zhou 1 , 2 ,
• Yulin Luo 2 ,
• Na Zhang 3 &
• Shen Liu   ORCID: orcid.org/0000-0002-6900-8831 2  

38 Accesses

Explore all metrics

Generalized anxiety disorders (GAD) is a chronic anxiety disorder characterized by autonomic excitability and hypervigilance. However, there was currently a lack of a quantitative synthesis of this time-varying science, as well as a measure of researchers’ networks and scientific productivity. Searching from the Web of Science Core Collection, PubMed, and Scopus on January 31st, 2024. The scientometric analysis was realized and the clinical research of GAD in recent ten years was explored. 9703 studies published from 2014 to 2023 were included, which aggregated into a well-structured network with credible clustering. It was worth studying the recent trend of productivity. Eleven clusters were identified by the co-citation reference network. The network structure was reasonable ( Q  = 0.5996) and the clustering reliability was high ( S  = 0.8378). The main trend of research is ‘’china’’, ‘’epidemic’’. These results can provide reference for the future development of funding agencies and research groups.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Similar content being viewed by others

Literature reviews as independent studies: guidelines for academic practice

Sascha Kraus, Matthias Breier, … João J. Ferreira

What is Qualitative in Research

Patrik Aspers & Ugo Corte

How to design bibliometric research: an overview and a framework proposal

Oğuzhan Öztürk, Rıdvan Kocaman & Dominik K. Kanbach

Data availability

The datasets generated and analyzed during the current study are not publicly available. The datasets are available from the corresponding author on reasonable request when the aim is to verify the published results.

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

Adams, J. (2012). Collaborations: The rise of research networks. Nature , 490 (7420), 335–336.

Article   PubMed   Google Scholar  

Alomari, N. A., Bedaiwi, S. K., Ghasib, A. M., Kabbarah, A. J., Alnefaie, S. A., Hariri, N., Altammar, M. A., Fadhel, A. M., & Altowairqi, F. M. (2022). Social anxiety disorder: Associated conditions and therapeutic approaches. Cureus , 14 (12), e32687.

PubMed   PubMed Central   Google Scholar  

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). American Psychiatric Association.

Anderson, L., Campbell-Sills, L., Ursano, R. J., Kessler, R. C., Sun, X., Heeringa, S. G., & Stein, M. B. (2019). Prospective associations of perceived unit cohesion with postdeployment mental health outcomes. Depression and Anxiety , 36 (6), 511–521.

Article   PubMed   PubMed Central   Google Scholar  

Annan Liu, Y., Lu, C., Gong, J., Sun, B. W., & Zhimei Jiang. (2023). Bibliometric analysis of research themes and trends of the co-occurrence of autism and ADHD. Neuropsychiatric Disease and Treatment , 19 , 985–1002.

Auerbach, R. P., Mortier, P., Bruffaerts, R., Alonso, J., Benjet, C., Cuijpers, P., Demyttenaere, K., Ebert D. D., Green, J. G., Hasking, P., Murray, E., Nock, M. K., Pinder-Amaker, S., Sampson, N. A., Stein, D. J., Vilagut, G., Zaslavsky, A. M., Kessler, R. C.; & WHO WMH-ICS Collaborators. (2018). WHO World Mental Health Surveys International College Student Project: Prevalence and distribution of mental disorders. Journal of Abnormal Psychology , 127 (7), 623–638. https://doi.org/10.1037/abn0000362

Bandelow, B., & Michaelis, S. (2015). Epidemiology of anxiety disorders in the 21st century. Dialogues in Clinical Neuroscience , 17 (3), 327–335.

Boeldt, D., McMahon, E., McFaul, M., & Greenleaf, W. (2019). Using virtual reality exposure therapy to enhance treatment of anxiety disorders: Identifying areas of clinical adoption and potential obstacles. Frontiers in Psychiatry , 10 , 773.

Cao, W., Fang, Z., Hou, G., Han, M., Xu, X., Dong, J., & Zheng, J. (2020). The psychological impact of the COVID–19 epidemic on college students in China. Psychiatry Research , 287 , 112934.

Chen, B., & Shin, S. (2021). Bibliometric analysis on research trend of accidental falls in older adults by using Citespace—focused on web of Science Core Collection (2010–2020). International Journal of Environmental Research and Public Health , 18 (4), 1663.

Chen, C. M. (2006). CiteSpace II: Detecting and visualizing emerging trends and transient patterns in scientific literature. Journal of the American Society for Information Science and Technology , 57 (3), 359–377.

Article   Google Scholar  

Chen, C. M., Ibekwe-Sanjuan, F., & Hou, J. (2010). The structure and dynamics of co-citation clusters: A multiple-perspective co-citation analysis. Journal of the American Society for Information Science and Technology , 61 (7), 1386–1409.

Cortese, S., Sabé, M., Chen, C., Perroud, N., & Solmi, M. (2022). Half a century of research on Attention-Deficit/Hyperactivity disorder: A scientometric study. Neuroscience and Biobehavioral Reviews , 140 , 104769.

Cuijpers, P., Sijbrandij, M., Koole, S., Huibers, M., Berking, M., & Andersson, G. (2014). Psychological treatment of GAD: A meta-analysis. Clinical Psychology Review , 34 (2), 130–140.

Flückiger, C., Del Re, A. C., Wampold, B. E., & Horvath, A. O. (2018). The alliance in adult psychotherapy: A meta-analytic synthesis. Psychotherapy , 55 (4), 316–340.

Gao, J., Zheng, P., Jia, Y., Chen, H., Mao, Y., Chen, S., Wang, Y., Fu, H., & Dai, J. (2020). Mental health problems and social media exposure during COVID–19 outbreak. PLoS One , 15(4), e0231924.

Giotakos, O. (2020). Neurobiology of emotional trauma. Psychiatriki , 31 (2), 162–171.

Goodwin, G. M., & Stein, D. J. (2021). Generalised anxiety disorder and depression: Contemporary treatment approaches. Advances in Therapy , 38 (Suppl 2), 45–51.

Grant, B., Hasin, D., Stinson, F. S., Dawson, D. A., Ruan, W. J., Goldstein, R. B., … Huan, B. (2005). Prevalence, correlates, co-morbidity, and comparative disability of DSM-IV generalized anxiety disorder in the USA: Results from the National Epidemiologic Survey on alcohol and related conditions. Psychological Medicine , 35 , 1747–1759.

Hasin, D. S., Sarvet, A. L., Meyers, J. L., Saha, T. D., Ruan, W. J., Stohl, M., & Grant, B. F. (2018). Epidemiology of adult DSM-5 major depressive disorder and its specifiers in the United States. JAMA Psychiatry , 75 (4), 336–346. https://doi.org/10.1001/jamapsychiatry.2017.4602

Herrera, M., Roberts, D. C., & Gulbahce, N. (2010). Mapping the evolution of scientific fields. Plos One , 5 (5), e10355.

Huang, Y., & Zhao, N. (2020). GAD, depressive symptoms and sleep quality during COVID–19 outbreak in China: A web-based cross-sectional survey. Psychiatry Research , 288 , 112954.

Jia, H., Guerin, R. J., Barile, J. P., Okun, A. H., McKnight-Eily, L., Blumberg, S. J., Njai, R., & Thompson, W. W. (2021). National and state trends in anxiety and depression severity scores among adults during the COVID–19 pandemic-United States, 2020–2021. MMWR Morbidity and Mortality Weekly Report , 70 (40), 1427–1432.

Koleck, T. A., Dreisbach, C., Bourne, P. E., & Bakken, S. (2019). Natural language processing of symptoms documented in free-text narratives of electronic health records: A systematic review. Journal of the American Medical Informatics Association: JAMIA , 26 (4), 364–379.

Kumar, V., Sattar, Y., Bseiso, A., Khan, S., & Rutkofsky, I. H. (2017). The effectiveness of internet-based cognitive behavioral therapy in treatment of psychiatric disorders. Cureus , 9(8), e1626.

Lai, J., Ma, S., Wang, Y., Cai, Z., Hu, J., Wei, N., & Hu, S. (2020). Factors associated with mental health outcomes among health care workers exposed to coronavirus disease 2019. JAMA Network Open , 3(3), e203976.

Li, J., Goerlandt, F., & Reniers, G. (2021). An overview of scientometric mapping for the safety science community:Methods,tools,and framework. Safety Science , 134 , 105093.

Li, J., Zhong, Y., Ma, Z., Wu, Y., Pang, M., Wang, C., & Ning, Z. (2020). Emotion reactivity-related brain network analysis in GAD: A task fMRI study. Bmc Psychiatry , 20 (1), 429.

Lin, C. Y., Liaw, S. Y., Chen, C. C., Pai, M. Y., & Chen, Y. M. (2017). A computer-based approach for analyzing consumer demands in electronic word-of-mouth. Electronic Markets , 27 , 225–242.

Liu, D., Che, S., & Zhu, W. (2022). Visualizing the Knowledge Domain of Academic Mobility Research from 2010 to 2020: A bibliometric analysis using CiteSpace. SAGE Open , 12 (1).

Liu, S., Sun, Y. P., Gao, X. L., & Sui, Y. (2019). Knowledge domain and emerging trends in Alzheimer’s disease: A scientometric review based on CiteSpace analysis. Neural Regeneration Research , 14 (9), 1643–1650.

Li, Y., Abdul-Rashid, S. H., & Raja, Ghazilla, R. A. (2022). Design methods for the elderly in web of Science, Scopus, and China National Knowledge Infrastructure Databases: A scientometric analysis in CiteSpace. Sustainability , 14 (5), 2545.

Li, Y. G., & Wu, H. Y. (2012). A clustering method based on k -means algorithm. Physics Procedia , 25 , 1104–1109.

Michael, J. A., Wang, M., Kaur, M., Fitzgerald, P. B., Fitzgibbon, B. M., & Hoy, K. E. (2021). EEG correlates of attentional control in anxiety disorders: A systematic review of error-related negativity and correct-response negativity findings. Journal of Affective Disorders , 291 , 140–153.

Micoulaud-Franchi, J. A., Jeunet, C., Pelissolo, A., & Ros, T. (2021). EEG neurofeedback for anxiety disorders and post-traumatic stress disorders: A blueprint for a promising brain-based therapy. Current Psychiatry Reports , 23 (12), 84.

Newman, M. E. (2001). The structure of scientific collaboration networks. Proceedings of the National Academy of Sciences of the United States of America , 98 (2), 404–409.

Noyes, R. Jr (2001). Comorbidity in GAD. The Psychiatric Clinics of North America , 24 (1), 41–55. https://doi.org/10.1016/s0193-953x(05)70205-7 .

Noyes, R., Jr, Clarkson, C., Crowe, R. R., Yates, W. R., & McChesney, C. M. (1987). A family study of GAD. American Journal of Psychiatry , 144 (8), 1019–1024.

PubMed   Google Scholar  

Pan, R. K., Sinha, S., Kaski, K., & Saramaki, J. (2012). The evolution of interdisciplinarity in physics research. Scientific Reports , 2 , 551.

Pigott, T. A. (2003). Anxiety disorders in women. Psychiatric Clinics of North America , 26 (3), 621–672.

Plummer, F., Manea, L., Trepel, D., & McMillan, D. (2016). Screening for anxiety disorders with the GAD-7 and GAD-2: A systematic review and diagnostic meta analysis. General Hospital Psychiatry , 39 , 24–31.

Porter, A. L., & Rafols, I. (2009). Is science becoming more interdisciplinary? Measuring and mapping six research fields over time. Scientometrics , 81 , 719–745.

Priya, A., Garg, S., & Tigga, N. P. (2020). Predicting anxiety, depression and stress in modern life using machine learning algorithms. Procedia Computer Science , 167 , 1258–1267.

Radhakrishnan, S., Erbis, S., Isaacs, J. A., & Kamarthi, S. (2017). Correction: Novel keyword co-occurrence network-based methods to foster systematic reviews of scientific literature. Plos One 12(9), e0185771.

Sabe, M., Chen, C., Perez, N., Solmi, M., Mucci, A., Galderisi, S., Strauss, G. P., & Kaiser, S. (2023). Thirty years of research on negative symptoms of schizophrenia: A scientometric analysis of hotspots, bursts, and research trends. Neuroscience and Biobehavioral Reviews , 144 , 104979.

Sabe, M., Pillinger, T., Kaiser, S., Chen, C., Taipale, H., Tanskanen, A., Tiihonen, J., Leucht, S., Correll, C. U., & Solmi, M. (2022). Half a century of research on antipsychotics and schizophrenia: A scientometric study of hotspots, nodes, bursts, and trends. Neuroscience and Biobehavioral Reviews , 136 , 104608.

Sabljić, V., Ružić, K., & Rakun, R. (2011). Venlafaxine Withdrawal syndrome. Psychiatria Danubina , 23 (1), 117–119.

Santomauro, D. F., Herrera, A., Shadid, J., et al. (2021). Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic. The Lancet , 398 , 10312.

Sanyaolu, A., Okorie, C., Marinkovic, A., Patidar, R., Younis, K., Desai, P., Hosein, Z., Padda, I., Mangat, J., & Altaf, M. (2020). Comorbidity and its impact on patients with COVID-19. SN Comprehensive Clinical Medicine , 2 (8), 1069–1076.

Sharma, C., Ferrao, Albertella, E. C., & Fontenelle, L. F. (2021). The impact of GAD in obsessive-compulsive disorder patients. Psychiatry Research, 300 , 113898.

Shen, Z., Li, G., Fang, J., Zhong, H., Wang, J., Sun, Y., & Shen, X. (2022). Aberrated multidimensional EEG characteristics in patients with GAD: A machine-learning based analysis framework. Sensors (Basel) , 22 (14), 5420.

Shi, L., Lu, Que, Y., Huang, Liu, M. S., Gong, Yuan, W., Sun, y. k., Shi, J., Bao, Y. P., & Lu, L. (2020). Prevalence of and risk factors associated with mental health symptoms among the general population in China During the Coronavirus Disease 2019 Pandemic. JAMA Network Open, 3 (7), E2014053.

Shimada-Sugimoto, M., Otowa, T., & Hettema, J. M. (2015). Genetics of anxiety disorders: Genetic epidemiological and molecular studies in humans. Psychiatry and Clinical Neurosciences , 69 (7), 388–401.

Spitzer, R. L., Kroenke, K., Williams, J. B., & Löwe, B. (2006). A brief measure for assessing GAD: The GAD-7. Archives of Internal Medicine , 166 (10), 1092–1097.

Strawn, J. R., Geracioti, L., Rajdev, N., Clemenza, K., & Levine, A. (2018). Pharmacotherapy for GAD in adult and pediatric patients: An evidence-based treatment review. Expert Opinion on Pharmacotherapy , 19 (10), 1057–1070.

Ströhle, A., Gensichen, J., & Domschke, K. (2018). The diagnosis and treatment of anxiety disorders. Deutsches Arzteblatt International , 155 (37), 611–620.

Tankard, C. F., Waldstein, S. R., Siegel, E. L., Holder, L. E., Lefkowitz, D., Anstett, F., & Katzel, L. I. (2003). Cerebral blood flow and anxiety in older men: An analysis of resting anterior asymmetry and prefrontal regions. Brain and Cognition , 52 (1), 70–78.

Terlizzi, E. P., & Schiller, J. S. (2021). Estimates of mental health symptomatology, by month of interview: United States, 2019. National Center for Health Statistics .

Tolin, D. F., Davies, C. D., Moskow, D. M., & Hofmann, S. G. (2020). Biofeedback and neurofeedback for anxiety disorders: A quantitative and qualitative systematic review. Advances in Experimental Medicine and Biology , 1191 , 265–289.

Uttam, C., & Apurva Shah. (2021). Topic modeling using latent dirichlet allocation. ACM Computing Surveys (CSUR) , 54 , 1–35.

Google Scholar  

Verônica de Medeiros Alves, Edilson Leite de Moura, Larissa Tenório Andrade Correia, &, Antonio, E., & Nardi (2017). Genetic polymorphisms and GAD: A systematic review. Medical Express, 4 (1), M170101.

Wittchen, H. U., & Hoyer, J. (2001). GAD: Nature and course. The Journal of Clinical Psychiatry , 62 (Suppl 11), 15–21.

Woelk, H., & Schläfke, S. (2010). A multi-center, double-blind, randomised study of the lavender oil preparation silexan in comparison to Lorazepam for GAD. Phytomedicine , 17 (2), 94–99.

Zara, G., Settanni, M., Zuffranieri, M., Veggi, S., & Castelli, L. (2021). The long psychological shadow of COVID-19 upon healthcare workers: A global concern for the action. The Journal of Affective Disorders , 294 , 220–226.

Zheng, R., Zhou, Y., Fu, Y., Xiang, Q., Cheng, F., Chen, H., Xu, H., Fu, L., Wu, X., Feng, M., Ye, L., Tian, Y., Deng, R., Liu, S., Jiang, Y., Yu, C., & Li, J. (2021). Prevalence and associated factors of depression and anxiety among nurses during the outbreak of COVID-19 in China: A cross-sectional study. International Journal of Nursing Studies , 114 , 103809.

Zhou, S. J., Zhang, L. G., Wang, L. L., Guo, Z. C., Wang, J. Q., Chen, J. C., & Chen, J. X. (2020). Prevalence and socio-demographic correlates of psychological health problems in Chinese adolescents during the outbreak of COVID–19. European Child & Adolescent Psychiatry , 29 (6), 749–758.

Download references

Acknowledgements

This study was supported by the Outstanding Youth Program of Philosophy and Social Sciences in Anhui Province (2022AH030089) and the Starting Fund for Scientific Research of High-Level Talents at Anhui Agricultural University (rc432206).

Author information

Authors and affiliations.

Department of Psychology, Institute of Education, China University of Geosciences, Wuhan, 430074, China

Department of Psychology, School of Humanities and Social Sciences, Anhui Agricultural University, Hefei, Anhui, 230036, China

Ying Zhou, Yulin Luo & Shen Liu

Department of Information Management, Anhui Vocational College of Police Officers, Hefei, Anhui, China

You can also search for this author in PubMed   Google Scholar

Contributions

Study design: Shen Liu, Na Zhang. Data collection, analysis and interpretation: Ying Zhou. Drafting the manuscript: Ying Zhou, Yulin Luo, Na Zhang, Shen Liu. Critical revision of the manuscript: Shen Liu, Na Zhang. Approval of the final version for publication: all co-authors.

Corresponding authors

Correspondence to Na Zhang or Shen Liu .

Ethics declarations

Compliance with ethical standards.

This is a review. This article does not contain any studies involving human participants performed by any of the authors.

Competing interests

The authors have no conflicts of interest to declare that are relevant to the content of this article. The authors declare that they have no known competing fnancial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions.

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Zhou, Y., Luo, Y., Zhang, N. et al. Ten years of researches on generalized anxiety disorder (GAD): a scientometric review. Curr Psychol (2024). https://doi.org/10.1007/s12144-024-05872-2

Download citation

Accepted : 09 March 2024

Published : 11 April 2024

DOI : https://doi.org/10.1007/s12144-024-05872-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Bibliometric
• Bibliometrix
• Find a journal
• Publish with us
• Track your research

• Open access
• Published: 10 April 2024

The burden of anxiety, depression, and stress, along with the prevalence of symptoms of PTSD, and perceptions of the drivers of psychological harms, as perceived by doctors and nurses working in ICUs in Nepal during the COVID-19 pandemic; a mixed method evaluation

• Shirish KC 1 ,
• Tiffany E. Gooden 2 ,
• Diptesh Aryal 1 ,
• Kanchan Koirala 1 ,
• Subekshya Luitel 1 ,
• Rashan Haniffa 3 , 4 ,
• Abi Beane 3 , 4 on behalf of

Collaboration for Research, Implementation, and Training in Critical Care in Asia and Africa (CCAA)

BMC Health Services Research volume  24 , Article number:  450 ( 2024 ) Cite this article

2463 Accesses

5 Altmetric

Metrics details

The COVID-19 pandemic resulted in significant physical and psychological impacts for survivors, and for the healthcare professionals caring for patients. Nurses and doctors in critical care faced longer working hours, increased burden of patients, and limited resources, all in the context of personal social isolation and uncertainties regarding cross-infection. We evaluated the burden of anxiety, depression, stress, post-traumatic stress disorder (PTSD), and alcohol dependence among doctors and nurses working in intensive care units (ICUs) in Nepal and explored the individual and social drivers for these impacts.

We conducted a mixed-methods study in Nepal, using an online survey to assess psychological well-being and semi-structured interviews to explore perceptions as to the drivers of anxiety, stress, and depression. Participants were recruited from existing national critical care professional organisations in Nepal and using a snowball technique. The online survey comprised of validated assessment tools for anxiety, depression, stress, PTSD, and alcohol dependence; all tools were analysed using published guidelines. Interviews were analysed using rapid appraisal techniques, and themes regarding the drivers for psychological distress were explored.

134 respondents (113 nurses, 21 doctors) completed the online survey. Twenty-eight (21%) participants experienced moderate to severe symptoms of depression; 67 (50%) experienced moderate or severe symptoms of anxiety; 114 (85%) had scores indicative of moderate to high levels of stress; 46 out of 100 reported symptoms of PTSD. Compared to doctors, nurses experienced more severe symptoms of depression, anxiety, and PTSD, whereas doctors experienced higher levels of stress than nurses. Most (95%) participants had scores indicative of low risk of alcohol dependence. Twenty participants were followed up in interviews. Social stigmatism, physical and emotional safety, enforced role change and the absence of organisational support were perceived drivers for poor psychological well-being.

Nurses and doctors working in ICU during the COVID-19 pandemic sustained psychological impacts, manifesting as stress, anxiety, and for some, symptoms of PTSD. Nurses were more vulnerable. Individual characteristics and professional inequalities in healthcare may be potential modifiable factors for policy makers seeking to mitigate risks for healthcare providers.

Peer Review reports

Introduction

Between January 2020 and December 2021, the COVID-19 pandemic led to an estimated 18.2 million deaths [ 1 ]. Globally, healthcare systems were overwhelmed during the pandemic, with intensive care units (ICUs) receiving an unprecedented burden of patients [ 2 ]. In Nepal, the government first declared a lockdown on March 24, 2020, that lasted until July 21, 2020, and the second lockdown was announced on April 29, 2021, which was fully lifted on September 1, 2021 [ 3 ]. The first wave of the COVID-19 pandemic reached a peak of over 5000 cases a day in October 2020, and the second wave reached a peak of more than 9000 cases a day in May 2021, which was almost double [ 4 ]. Prior to the pandemic, Nepal reported a capacity of 1595 ICU beds across 194 hospitals and around 840 ventilators, equating to 2.8 ventilator-equipped ICU beds per 100,000 people [ 5 ]. To cope with the influx of COVID-19 patients, several existing postoperative wards and other high-dependency units of the hospitals were converted into improvised critical care units [ 6 ]. Globally, healthcare professionals (HCPs) and specifically those working in ICU and critical care services, arguably were at the frontline of the healthcare response. These HCPs faced the uncertainty of managing this new condition, extended working hours, limited personal protective equipment (PPE), and an increased risk of infection as they provide essential lifesaving interventions, including intubation and non-invasive respiratory management [ 7 , 8 ].

The impacts of the COVID-19 pandemic on the mental health and well-being of HCPs who worked during and after this global emergency are slowly becoming apparent. Research emerging from China, the USA, and Europe [ 9 ] describes a significant burden of psychological distress and symptoms synonymous with mental health conditions in HCPs. This is also evident from the limited studies that have been conducted in Nepal. For instance, one study conducted among 150 HCPs from outpatient clinics and inpatient wards caring for COVID-19 patients in Nepal reported that 38% of participants suffered from anxiety and/or depression [ 10 ]. Another Nepali study revealed that the prevalence of anxiety and depression among HCPs, including health assistants and support staff was 47% and 41%, respectively [ 11 ]. A larger online survey of 475 HCPs including pharmacists, paramedics and public health practitioners reported similar findings (42% had anxiety) and noted that nurses had a higher proportion of symptoms compared to other HCPs [ 12 ].. Whilst these studies, in conjunction with a meta-analysis, indicate that depression, anxiety, and post-traumatic disorder (PTSD) are highly prevalent among HCPs during the pandemic [ 9 , 10 , 11 , 12 , 13 ], fewer studies have explored the disparities between professionals’ roles, specifically among ICU workers, a group exposed to more advanced cases of COVID-19. Indeed a small study in Nepal comprising 96 nurses revealed that nurses who worked directly with COVID-19 patients experienced more severe symptoms of depression and anxiety [ 13 ]. The nature and characteristics of mental health symptoms appear to vary geographically, the HCPs’ role, their individual characteristics (age, gender) along with health system’s pre-existing resource capacity and ability to respond to increasing demand placed by events such as a pandemic. Understanding the mental health impact of ICU workers, any disparities between professional roles and drivers behind poor mental health in Nepal will help to identify what support is needed for ICU workers for pandemic preparedness; thus, providing important directions for investment in health systems strengthening.

We aimed to investigate the burden of anxiety, depression, stress, PTSD, and alcohol dependence among doctors and nurses in Nepal that worked in the ICU during the COVID-19 pandemic. We further sought to identify the factors driving the self-reported burden of psychological distress by exploring the lived experiences of these two different professional groups, and how these experiences impacted their psychological health and well-being.

Study design

We undertook a mixed-methods cross-sectional study [ 14 ] in Nepal with ICU doctors and nurses, combining an online questionnaire consisting of validated self-assessment tools combined with semi-structured interviews. The following self-reporting psychological assessment tools were used, given they have been used in previous studies in other settings and their widely validated in a variety of settings: Beck Anxiety Inventory (BAI) [ 15 ], Beck Depression Inventory (BDI) [ 16 ], Perceived Stress Scale (PSS) [ 17 ], PTSD Checklist for Diagnostic and Statistical Manual of Mental Disorders-5 (PCL-5) [ 18 ] and Alcohol Use Disorder identification Tool (AUDIT) [ 19 ]. BDI, BAI, and AUDIT have been validated in Nepal [ 20 , 21 , 22 ] and the PSS has been tested for reliability and correlation in Nepal [ 23 ]. Whilst the PCL-5 has not been validated in a Nepali setting, it was piloted (along with all other assessment tools used) with 20 people before the study commenced. Participants were given the flexibility to complete the questionnaire in either Nepali or English language. Despite this option, all participants opted to respond in English.

Ethics approval

was granted from the Nepal Health Research Council (approval number: 176/2021 P). All participants provided informed consent electronically before completing the online questionnaire. Participants from the qualitative component provided further informed verbal consent before the interview commenced.

In 2020, Nepal reported a capacity of 1595 ICU beds across 194 hospitals and around 840 ventilators, equating to 2.8 ventilator-equipped ICU beds per 100,000 people [ 5 ]. A year later, Nepal was under a state of health emergency, with patients being turned down due to a lack of ICU beds, oxygen, and ventilators [ 24 ].

Participants and recruitment

Doctors and nurses with experience in caring for COVID-19 patients in Nepalese ICUs were eligible for participation. Initially doctors registered with the Nepalese Society of Critical Care Medicine (NSCCM) [ 25 ] and nurses registered with the Critical Care Nurses Association of Nepal (CCNAN) [ 26 ] were contacted and invited to participate. Both organisations consist of voluntary memberships and represent the doctors and nurses working in a critical care setting in Nepal. At the time of recruitment, there were 187 doctors and 104 nurses registered at these organisations. This initial purposive sampling was augmented by snowballing techniques, whereby respondents were invited to forward the questionnaire link to other doctors or nurses working in ICUs [ 27 ]. Following completion of the questionnaire, respondents were invited to participate in a virtual interview. A convenience sample of 20 participants (a number which, based on the literature, was likely to provide saturation of findings [ 28 ]) was subsequently scheduled for an interview.

Study materials and data collection

The questionnaire was developed using an online survey platform (Google Forms) [ 29 ]. The questionnaire was piloted for readability and responder reliability with twenty HCPs based in Nepal, prior to roll out, who did not participate in the final analysis. Questionnaire content included socio-demographic information; age, sex, professional role and experience, degree of schooling, and home living arrangements; factors which had been identified as being important in the burden of psychological distress and impact on family life in similar research conducted during the previous SARS pandemic as well as the current COVID-19 event [ 30 ]. Participants could opt out of the study at any time. Participants could only complete the questionnaire once, and all survey responses were anonymous. Participants were signposted to healthcare services available to them should they be suffering from any distressing, mild, moderate or severe mental health symptoms. Invitations to participate in the questionnaire were sent out from 20th May 2021, and the questionnaire was closed to responses on 2nd October 2021.

The semi-structured interview topic guide was co-developed between doctors and nurses working in ICUs in Kathmandu. Co-design was used to ensure the sensitivity and appropriateness of the questions. None of the doctors and nurses involved in the codesign of the topic guide participated in the study proper. The qualitative component was aimed to augment the quantitative findings by providing an understanding of what social, organisational, and environmental factors were related to HCPs’ mental health. Topic guide questions focused on HCPs’ perceptions of their experiences of working during the pandemic and explored social, organisational, and environmental factors that may have influenced their self-reported burden and symptoms of psychological distress. These factors were selected from a review of the findings of the previously published meta-analysis and other studies conducted in Nepal [ 9 , 10 , 11 , 12 , 13 ]. The interview questions were piloted with five HCPs for interpretability and interviewer consistency. All interviews were conducted via video conferencing (Zoom) [ 31 ] between September 2021 and March 2022. Five ICU nurses with experience in conducting interviews and mixed methods research led the data collection following training on the topic guide. To ensure there was no prior relationship between the interviewer and the participant, interviewers were assigned to participants that worked in different ICUs than themselves and were not known to the interviewee. No one other than the interviewer and the participant was present for each interview, and interviews were conducted at the time chosen by the interviewee. Rapid assessment procedure (RAP) sheets were used for note-taking during the interviews [ 32 ]. Commonly used in rapid evaluations - designed to improve the rapidity and replicability of research during public health emergencies - RAP sheets help reduce the need for long-form transcription and encourage reflexivity for both interviewers and researchers, reduce interviewer bias, and enable validation of internal consistency with coding [ 33 ]. The RAP sheet contained the summary of questions from the topic guide, and the interviewers took notes of what the participants said regarding each question during the interview.

Data analysis

Descriptive statistics were used to describe participants’ demographics and professional profiles. Psychological health and well-being assessment tools from the questionnaire were analysed using published guidelines. For the BDI, each of the 21 items corresponding to a symptom of depression was summed for each participant to give a single total score [ 16 ]. With each item ranging from 0 to 3 points, a total score of 13 or less was considered minimal to no depression, 14 to 19 as mild depression, 20 to 28 as moderate depression, and 29 to 63 as severe depression [ 16 ]. Data is also presented separately for suicidality (question 9 from the BDI) whereby anyone that said they have thoughts about or plans to kill themselves is said to have experienced suicidality. The BAI scores reported included the 21 symptoms of anxiety that ranged between 0 and 63 points [ 15 ]. The values for each symptom were summed, and a total score of 0 to 7 was interpreted as a minimal level of anxiety, 8 to 15 as mild, 16 to 25 as moderate, and 26 to 63 as severe anxiety [ 15 ]. Scores on the PSS ranged from 0 to 40, with higher scores indicating higher perceptions of stress [ 17 ]: scores ranging from 0 to 13 were considered low descriptors of stress; 14 to 26 moderate; and 27 to 40 were considered higher levels of perceived stress. For alcohol use disorder reported using AUDIT [ 19 ], a score of 0 indicated no previous or current alcohol use; a score of 1 to 7 suggested low-risk consumption; 8 to 14 hazardous or harmful alcohol consumption; 15 or higher indicated the likelihood of alcohol dependence (moderate to severe alcohol use disorder). The PCL-5 included 20 items with a score range of 0 to 80 and a score of 33 or higher, indicating the presence of PTSD [ 18 ]. A sensitivity analysis was conducted for the BDI, BAI and AUDIT scores based on local validation studies whereby a score of 15 or lower from the BDI indicated no depression [ 20 ], 12 or lower from the BAI indicated no anxiety [ 21 ], and a score of 11 or above from the AUDIT indicated discriminate dependent drinkers [ 22 ].

RAP sheets, along with interviewer notes, were reviewed by the research team before analysis to ensure information was complete. SK, KK and AB used a constant comparative method, coding data following each round of interviews and then reflecting back on the summary of the codes together with the interviewers to promote the accuracy of findings and reduce recall and interviewer bias. In addition, emerging themes identified following each round of coding were used to guide subsequent interviews [ 34 ]. The broader research team met following each coding round to review the findings and reflexivity [ 35 ]. Categories and the subsequent themes (‘drivers’) were developed through the iterative process of interviewing, coding, analysing, and reviewing.

We invited 120 doctors and 341 nurses to participate. A total of 21 doctors and 113 nurses responded, all of which completed the BDI, BAI, PSS, and AUDIT questions; 100 completed the PCL-5 (16 doctors and 84 nurses). Nearly all nurses were female (99%, n  = 112), whereas most doctors were male (81%, n  = 17). The characteristics of respondents are described in Table  1 .

50% ( n  = 67) of respondents reported experiencing symptoms associated with moderate to severe anxiety, and a further 27% ( n  = 36) scored for mild anxiety as a result of working in the ICU during the COVID-19 pandemic (Table  2 ). Anxiety levels (and associated symptoms) were more pronounced in nurses than doctors, with 55% ( n  = 62) of the former scoring moderate to severe on the anxiety scale, compared to 24% ( n  = 6) of the latter. 21% ( n  = 28) of respondents described symptoms associated with moderate to severe depression, with a near-even split between nurses and doctors. Three-quarters of respondents ( n  = 114; 85%) had scores indicative of moderate to high levels of stress; this proportion was higher among doctors ( n  = 19; 91%) compared to nurses ( n  = 95; 84%). Of the 100 individuals that completed the PCL-5 assessment (16 doctors and 84 nurses), 45% ( n  = 46) reported a constellation of symptoms closely associated with PTSD, with a higher prevalence among nurses ( n  = 40; 47%) compared to doctors ( n  = 6; 38%).

Using cut-off scores from Nepali validation studies, 45 (34%) participants were experiencing mild, moderate or severe depressive symptoms, 80 (60%) were experiencing mild, moderate or severe anxiety symptoms, and 3 (2%) were considered discriminate dependent drinkers. These results are in line with our main analysis, including that a greater proportion of nurses were still found to suffer from depression and anxiety symptoms (supplementary Table 1 ).

Forty-six respondents to the online questionnaire volunteered to participate in the subsequent semi-structured interviews. Twenty participants were approached and consented to an interview: 16 were nurses (all female), and 4 were doctors (1 female, 3 male). On average, each interview resulted in 45 to 60 min of qualitative data. Saturation was met within the first 15 interviews, and findings were consistent between the coders and the research team. Analysis and synthesis of the interviews revealed nine themes, which, when codified, can be described as three key drivers of the psychological symptoms and impacts on mental well-being experienced by the interviewees: social stigmatism, physical and emotional safety, and organisational support. (Fig.  1 ). During the interviews, HCPs further described some of the coping strategies that they found helpful in mitigating the impacts experienced and may provide insights for future pandemic preparedness. These three themes, the drivers, and coping strategies, are explored below, along with quotes from the respondents.

Coding tree for the four main drivers for psychological distress

Social stigmatism

Interviewees described experiencing feelings of social stigmatisation as a result of interactions with their families, peers, as well as from the wider public. Examples of stigmatism experienced included physical avoidance from neighbours and community members when the HCP travelled to and from and around their home, especially when dwellings were in shared buildings and common areas.

“My house owner avoided talking and meeting me because I worked with COVID patients.” [N]. “I have an elderly family member, and I was afraid and worried [for them] when I came back from duty.” [N].

Interviewees described how rumours would spread within the community, notably related to concerns of risk of co-infection or cross-infection, either directly from parent to child or indirectly via friends and extended family. Some HCPs were asked or elected to stay away from their home so as to reduce the stigma to them and their family and in an attempt to reduce the risk of co-infection, particularly when they had vulnerable family members. Interviewees described how this self-selected or enforced separation and isolation resulted in feelings of rejection, physically and emotionally heightened feelings of stress and anxiety, alongside the threat to physical and emotional safety.

Physical and emotional safety

Increased workload and an enforced change in working pattern/ shift structures were experienced by all the HCPs interviewed. These longer overall working hours, increased duration of shift patterns, and enforced working rotas were perceived as resulting in a loss of physical and emotional safety by the interviewees. Feelings of loss of control, insomnia, or disruption to sleep patterns, alongside physical discomfort through sustained working in personal protective equipment, often in hot and humid temperatures. This physical and mental endurance contributed to feelings of emotional stress and anxiety.

“Shift frequency was increased, and I only got one night off in a week. Sometimes I had to work extra hours, which was very stressful.” [N]. “My sleep pattern had changed, I felt restless and was afraid about COVID” [D].

The change in shift structure and in working patterns meant for some HCPs enforced separation from family and friends whereby HCPs sought accommodation away from family or in temporary lodgings. This again resulted in isolation and additional strain on other family members so as to provide care for HCP’s dependents.

“I had to involve other family members to arrange for the medication and care of my grandmother” [N].

Increased working hours and changes in working patterns further had physical impacts; participants described skipping meals or having limited time to eat. The need to wear personal protective equipment (PPE), and indeed the risks to safety when PPE was not available, associated risks of non-availability of equipment, brought with it a risk to physical and emotional safety. HCPs interviewed reported skin lacerations, irritation, and discomfort whilst wearing equipment in hot, humid working environments.

“We had to frequently change the PPE and masks, which has caused skin problems that still exist.” [N].

Organisational support

Interviewees found the COVID-19 pandemic brought new and often enforced work responsibilities, some of which were associated with high levels of professional anxiety, stress, and uncertainty. A professionally challenging situation, even for those with many years of ICU working experience. HCPs faced emotionally challenging tasks such as dealing with end-of-life situations (particularly without relatives of the patient present) and having to comfort relatives over the phone, of which they received limited to no training or support on handling such situations.

“I went through an emotional breakdown while dealing with the end of the life situation of patients without the presence of family members in the COVID ICU… I felt sad when a young patient lost their lives” [D]. “Accommodation or isolation facilities should be provided by the hospital” [D]. “If incentives were provided in time and staff were provided with health insurance it would motivate us” [N].

Ever-changing role and responsibilities created anxiety for HCPs as to what care to deliver, and the rapidity and uncertainty of care were associated with feelings of vulnerability. Interviewees expressed how they wished there was a need for greater organisational support to better cope with the frequent updates and changes to practice. Furthermore, HCPs expressed concerns regarding a shortage of staff and the lack of mental health counselling and support, accommodation on-site at the hospital, and transportation to and from work.

“Mental health support or counselling facilities were not provided. It should be there… seniors and hospital staff should also talk to the staff to know the situation.” [N]. “Safety of healthcare workers should be the priority and nurse-patient ratio should be maintained to provide quality care to the patients… hospital should have recruited more staff.” [N].

Coping strategies

Participants described various ways in which they coped with the emotional, physical, social, and professional impacts of working through the pandemic. This included speaking with family and friends about the pressures they were under, taking up activities in their off time, such as gardening and reading, and using media entertainment such as music, movies, and shows. A few participants also mentioned that comparing the situation in Nepal to other countries (i.e., keeping up-to-date with the news) also helped them cope. Others mentioned that detachment from social media and more self-awareness through meditation helped.

“I ventilated my feelings with friends and family. Listening to soothing music also helped me cope with the stress.” [N]. “I coped by gardening with my sister in my home.” [N]. “I… watched the news that compared the death rates, which was low compared to others.” [D].

The COVID-19 pandemic’s impact on healthcare services and population health internationally is unprecedented in recent times. As healthcare professionals, policymakers, and researchers work to strengthen services in preparation for future pandemics now and mitigate the long-term impacts on individual and population health, understanding the impact on and perspectives of doctors and nurses at the frontline of care can provide important learning regarding the individuals characteristics and professional, social and economic drivers which may increase the risk of psychological impacts.

Mandated and enforced changes in role, specifically in working hours and shift patterns, were a key driver of psychological anxiety and distress. Within hospitals in Nepal, many departments were closed, and stay-at-home orders meant that outpatient or clinical services all but ceased. This resulted in an increased role and scope for critical care trained staff, and in contrast to other health systems (such as the UK) where healthcare staff were redeployed to ICU, there was a separation for ICU staff even from their professional peers working in other specialties. The increased scope and uncertainty of the HCP’s role, along with limited choice in redeployment in the ICU was another driver of poor mental health- and dominated nursing participants’ experiences. Interviewees described how these changes impacted not only themselves but the multigenerational families for whom many cared for. This enforcement of role change, and the related descriptions of the drivers for these impacts as experienced by participants in this study point not only to the differences in roles between nurses and doctors; but also highlights disparities in autonomy, advocacy for role change during international emergencies, and the implications of work on home and family life [ 36 ].

Giving staff choice to select shift patterns and ensuring the opportunity to have periods of rest to reconnect with family and have self-care is needed. Consultation and shared decision-making, even in times of restricted choice, are associated with improved perceptions of work from staff and may result in reducing psychological distress and promoting emotional safety, which is, in turn, associated with better outcomes for patients [ 37 , 38 ]. However, nurses in Nepal, as with many health systems, may have less opportunity for strategic and organisational decision making in response to public health emergencies. The impact of ongoing disparities between professionals and their agency to advocate for wellbeing and safety warrants further research.

Nurses were disproportionately burdened by both occurrence and severity of symptoms of anxiety and depression as a result of their work during the pandemic when compared to doctors.

Nearly half of all respondents had symptoms of anxiety and PTSD (again more prevalent in nurses), and the burden of anxiety symptoms was higher than the reported 22–33% from a recent umbrella review [ 39 ]. The burden of stress we report was also higher than a smaller study conducted in Nepal during the pandemic, which reported stress among 53.2% of healthcare professionals working in hospitals, primary health centres, pharmacies, and health posts in Nepal [ 40 ]; it was also higher than a meta-analysis of published studies exploring the incidence of both stress (57%) and PTSD (22%) among all cadres of healthcare workers [ 41 ]. One reason for the higher reported symptoms in our study may be the focus on ICU workers and their role in the management of end-of-life care. Indeed, our results for depression and anxiety are comparable to a study involving nurses working directly with COVID-19 in Nepal [ 13 ]. Studies conducted elsewhere in Asia have highlighted this positive relationship between ICU experiences and poor mental health [ 42 ].

Nurses in Nepal, as with many other countries, are more likely to be female, younger in age, and have less opportunity for graduate study; and have lower earning potential than physician colleagues [ 43 ]; all characteristics associated with increased risk of poorer mental health outcomes [ 44 ]. Exploration into the disparities of the psychological and health impacts of COVID-19 on different cadres of healthcare workers is emerging. A systematic review conducted in 2020, identified 27 studies which sought to explore the disparity in impacts of the pandemic on HCP’s psychological well-being. The findings from the review are in line with ours, indicating that the burden of symptoms for anxiety, depression, and PTSD is higher in nurses compared to doctors [ 45 ]. Notably only a few of these studies used validated tools for assessment of specific symptoms of anxiety, depression, or substance misuse [ 45 ]. Our study serves to strengthen the evidence of the vulnerability of nurses.

Nepal, like many other lower and middle-income countries in South and Southeast Asia, enforced large-scale lockdowns and restrictions of movement for all but essential healthcare and municipal staff [ 46 ]. As such, social stigmatism, physical and emotional safety, and organisational support were key drivers behind the elevated symptoms of psychological distress in ICU HCPs and may be a key determinant of differences between health systems internationally. Furthermore, the family responsibilities and social circumstances for nurses, contributed to their experiences of isolation, rejection, vulnerability, physical discomfort, and strain. These drivers mirrored those reported from Europe; and may reflect differences experienced by nurses as a result of their gender, and role norms of primary family carers within society [ 44 ].

Interviewees from both professional groups expressed concern at the absence of preparedness and support they felt from their employing institutions. This is notable given the ongoing investment in pandemic preparedness and the potential to make changes now to prepare for the next pandemic or public health emergency. Interventions such as resilience training, scenario-based simulation training, and group exercises based on psychoeducation and cognitive behavioural therapy (CBT) principles have proved effective in reducing anxiety, depression, stress, and PTSD among doctors and nurses while simultaneously improving their ability to work in unprecedented situations in other sectors [ 47 ]. Similar provisions may be valuable for ICU-based healthcare professionals and are deliverable online, making rollout potentially more feasible.

Strengths and limitations

A strength of this study is the exploration of participants’ perspectives on the drivers behind the burden of poor mental health described in ICU HCPs. This mixed methods approach offers insights into doctors’ and nurses’ unique individual, social and professional characteristics that may be associated with increased risk of distress. These differences and their potential for disparity in impacts on health and wellbeing should be of interest to policymakers and healthcare facility managers involved in future pandemic preparedness. However, the study has some limitations to acknowledge. Given the use of the snowball technique, we were able to ensure a high number of respondents, but as a consequence, we were unable to track the number of respondents that came from using this technique compared to those initially invited from the NSCCM and CCNAN. Therefore, a response rate and, subsequently, a non-response rate could not be reported. We did not collect information on the level of training in critical care that participants received; trained health professionals are likely to have additional skills in how to handle the potential stressful environment in critical care settings. Also, due to the lack of validation of the PCL-5 in Nepal, the results of this assessment tool should be interpreted with caution. The survey tools used for this study have not been validated in an online format. However, given these tools were self-reporting, and were piloted and administered in English, the online format is thought to have minimal impact on the results. Additionally, participants for the qualitative component were recruited based on convenience sampling; therefore, the diversity of the sample may not be optimised. We acknowledge that recall bias may be present in the participants during the interview, given they were recalling their experiences throughout the pandemic for up to 24 months prior to the interview; however, we hope the piloting of the interviews, the use of multiple researchers to code the data, and the constant comparative nature of the evaluation will mitigate this potential.

The COVID-19 pandemic negatively impacted the mental health of HCPs worldwide. This study strengthens existing evidence that nurses were (and may remain) at increased risk of both cross infection and may also be more vulnerable to psychological impacts including anxiety, depression and PTSD than their professional colleagues. In addition, critical care staff may be at even greater risk, due to the uniqueness of their role which includes prolonged periods of time with infected patients, frontline role in managing end of life care, and as described here, limited ability to advocate for changing role and working patterns during an emergency. Professional hierarchies, and social-economic and gender profiles unique to nurses, may be potential drivers for these disparities, and warrants further research. Learning from the ICU HCPs’ experiences during the COVID-19 pandemic may inform future preparedness strategies e to mitigate short and long-term mental illness among ICU HCPs in future pandemics.

Data availability

The interview guide is available in the Figshare repository,

https://doi.org/10.6084/m9.figshare.24247384.v1 .

The data supporting the conclusions of this article are available in the Figshare repository, https://doi.org/10.6084/m9.figshare.23999790.v1 .

Abbreviations

Coronavirus disease 2019

Intensive care unit

Healthcare professional

Personal protective equipment

Post-traumatic stress disorder

Nepalese Society of Critical Care Medicine

Critical Care Nurses Association of Nepal

Beck Anxiety Inventory

Beck Depression Inventory

Perceived Stress Scale

PTSD Checklist for Diagnostic and Statistical Manual of Mental Disorders-5

Alcohol Use Disorder Identification Tool

Rapid assessment procedure

Cognitive behavioural therapy

Wang H, Paulson KR, Pease SA, et al. Estimating excess mortality due to the COVID-19 pandemic: a systematic analysis of COVID-19-related mortality, 2020–21. Lancet. 2022;399(10334):1513–36.

Article   CAS   Google Scholar  

Greco M, De Corte T, Ercole A et al. Clinical and organizational factors associated with mortality during the peak of first COVID-19 wave: the global UNITE-COVID study. Intensive Care Med. 2022:1–16.

All Covid restrictions lifted in Valley [Internet]. kathmandupost.com. [Accessed 7 Feb 2024]. Available from: https://kathmandupost.com/valley/2022/03/05/all-covid-restrictions-lifted-in-valley .

Kharel P. Nepal’s fight against the second wave of COVID-19 pandemic. repositoryunescaporg [Internet]. 2021 [cited 2024 Feb 7]; Available from: https://hdl.handle.net/20.500.12870/5102 .

Neupane HC, Gauli B, Adhikari S, et al. Contextualizing critical care medicine in the face of COVID-19 pandemic. JNMA J Nepal Med Assoc. 2020;58(226):447.

PubMed   PubMed Central   Google Scholar  

Shrestha GS, Lamsal R, Tiwari P, Acharya SP. Anesthesiology and critical care response to COVID-19 in Resource-Limited settings. Anesthesiol Clin. 2021;39(2):285–92.

Article   PubMed   PubMed Central   Google Scholar  

Hölscher AH. Patient, surgeon, and health care worker safety during the COVID-19 pandemic. Ann Surg. 2021;274(5):681.

Article   PubMed   Google Scholar  

Gholami M, Fawad I, Shadan S, et al. COVID-19 and healthcare workers: a systematic review and meta-analysis. Int J Infect Dis. 2021;104:335–46.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Ghahramani S, Kasraei H, Hayati R et al. Health care workers’ mental health in the face of COVID-19: a systematic review and meta-analysis. Int J Psychiatry Clin Pract. 2022:1–10.

Gupta AK, Mehra A, Niraula A, et al. Prevalence of anxiety and depression among the healthcare workers in Nepal during the COVID-19 pandemic. Asian J Psychiatr. 2020;54:102260.

Adhikari SP, Rawal N, Shrestha DB et al. Prevalence of anxiety, depression, and perceived stigma in healthcare workers in Nepal during later phase of first wave of covid-19 pandemic: a web-based cross-sectional survey. Cureus. 2021, 13(6).

Khanal P, Devkota N, Dahal M, et al. Mental health impacts among health workers during COVID-19 in a low resource setting: a cross-sectional survey from Nepal. Global Health. 2020;16(1):1–12.

Google Scholar  

Tamrakar P, Pant SB, Acharya SP. Anxiety and depression among nurses in COVID and non-COVID intensive care units. Nurs Crit Care. 2021 Sep 28.

Fitzpatrick R, Boulton M. Qualitative methods for assessing health care. Qual Health Care. 1994;3(2):107.

Beck AT, Epstein N, Brown G, et al. An inventory for measuring clinical anxiety: psychometric properties. J Consult Clin Psychol. 1988;56(6):893.

Article   CAS   PubMed   Google Scholar  

Richter P, Werner J, Heerlein A, et al. On the validity of the Beck Depression Inventory. Psychopathology. 1998;31(3):160–8.

Cohen S, Kamarck T, Mermelstein R. Perceived stress scale. Measuring Stress: Guide Health Social Scientists. 1994;10(2):1–2.

Blevins CA, Weathers FW, Davis MT, et al. The posttraumatic stress disorder checklist for DSM-5 (PCL‐5): development and initial psychometric evaluation. J Trauma Stress. 2015;28(6):489–98.

Saunders JB, Aasland OG, Babor TF, et al. Development of the alcohol use disorders identification test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption-II. Addiction. 1993;88(6):791–804.

Kohrt B, Kunz RD, Koirala NR, Sharma VD, Nepal, Mahendra. Validation of a Nepali version of the Beck Depression Inventory. Nepal J Psychiatry. 2002;2:123–30.

Kohrt B, Kunz R, Koirala N, Campus M, Nepal. Validation of the Nepali version of beck anxiety inventory. Journal of Institute of Medicine [Internet]. 2007 Jan 21 [cited 2024 Feb 7]; Available from: https://www.semanticscholar.org/paper/Validation-of-the-Nepali-version-of-beck-anxiety-Kohrt-Kunz/148d 01852884f54d28967e105d57c177a3f0be36.

Pradhan B, Chappuis F, Baral D, Karki P, Rijal S, Hadengue A et al. The alcohol use disorders identification test (AUDIT): validation of a Nepali version for the detection of alcohol use disorders and hazardous drinking in medical settings. Substance Abuse Treatment, Prevention, and Policy [Internet]. 2012;7(1). Available from: https://substanceabusepolicy.biomedcentral.com/articles/ https://doi.org/10.1186/1747-597X-7-42 .

Sharma P, Devkota G. Mental health screening questionnaire: a study on reliability and correlation with perceived stress score. J Psychiatrists’ Association Nepal. 2019;8(2):4–8.

Article   Google Scholar  

Pandey BD, Morita K, Costello A. Twin crises in Nepal: covid-19 and climate change. 2022, 377.

Nepalese Society of Critical Care Medicine. NSCCM. 2022. [Accessed 16 August 2023]. Available from http://www.nsccm.org.np/ .

Critical Care Nurses Association of Nepal. Critical Care Nurses Association of Nepal. 2022. [Accessed 16 August 2023]. Available from https://ccnan.org.np/ .

Etikan I, Alkassim R, Abubakar S. Comparision of snowball sampling and sequential sampling technique. Biom Biostat Int J. 2016;3(1):55.

Saunders B, Sim J, Kingstone T, et al. Saturation in qualitative research: exploring its conceptualization and operationalization. Qual Quant. 2018;52(4):1893–907.

Google Inc. Google Forms. 2022. [Accessed 16 August 2023]. Available from https://www.google.co.uk/forms/about/ .

Zerbini G, Ebigbo A, Reicherts P et al. Psychosocial burden of healthcare professionals in times of COVID-19–a survey conducted at the University Hospital Augsburg. Ger Med Sci. 2020, 18.

Zoom Z. One platform to connect. 2022. [Accessed 16 August 2023]. Available from https://zoom.us/ .

Vindrola-Padros C, Chisnall G, Cooper S, et al. Carrying out rapid qualitative research during a pandemic: emerging lessons from COVID-19. Qual Health Res. 2020;30(14):2192–204.

Beebe J. Basic concepts and techniques of rapid appraisal. Hum Organ. 1995;54(1):42–51.

Althubaiti A. Information bias in health research: definition, pitfalls, and adjustment methods. J Multidiscip Healthc. 2016:211–7.

Macbeth D. On reflexivity in qualitative research: two readings, and a third. Qual Inq. 2001;7(1):35–68.

Jackson D, Anders R, Padula WV, et al. Vulnerability of nurse and physicians with COVID-19: monitoring and surveillance needed. J Clin Nurs. 2020;29(19–20):3584.

Bae SH. Intensive care nurse staffing and nurse outcomes: a systematic review. Nurs Crit Care. 2021;26(6):457–66.

Ejebu O-Z, Dall’Ora C, Griffiths P. Nurses’ experiences and preferences around shift patterns: a scoping review. PLoS ONE. 2021;16(8):e0256300.

Fernandez R, Sikhosana N, Green H, et al. Anxiety and depression among healthcare workers during the COVID-19 pandemic: a systematic umbrella review of the global evidence. BMJ Open. 2021;11(9):e054528.

Kafle K, Shrestha DB, Baniya A, et al. Psychological distress among health service providers during COVID-19 pandemic in Nepal. PLoS ONE. 2021;16(2):e0246784.

Marvaldi M, Mallet J, Dubertret C, et al. Anxiety, depression, trauma-related, and sleep disorders among healthcare workers during the COVID-19 pandemic: a systematic review and meta-analysis. Neurosci Biobehav Rev. 2021;126:252–64.

Thatrimontrichai A, Weber DJ, Apisarnthanarak A. Mental health among healthcare personnel during COVID-19 in Asia: a systematic review. J Formos Med Assoc. 2021;120(6):1296–304.

Prakash S, Yadav P, Yadav K. Perspectives of developing nursing education in Nepal. Nurs Care Open Access J. 2018;5(4):214–20.

Tiete J, Guatteri M, Lachaux A, et al. Mental health outcomes in healthcare workers in COVID-19 and non-COVID-19 care units: a cross-sectional survey in Belgium. Front Psychol. 2021;11:612241.

Kunz M, Strasser M, Hasan A. Impact of the coronavirus disease 2019 pandemic on healthcare workers: systematic comparison between nurses and medical doctors. Curr Opin Psychiatry. 2021;34(4):413.

Adhikari NK, Beane A, Devaprasad D et al. Impact of COVID-19 on non-COVID intensive care unit service utilization, case mix and outcomes: a registry-based analysis from India. Wellcome Open Res. 2021, 6.

Zaçe D, Hoxhaj I, Orfino A, et al. Interventions to address mental health issues in healthcare workers during infectious disease outbreaks: a systematic review. J Psychiatr Res. 2021;136:319–33.

Download references

Acknowledgements

We thank the volunteers who took the time to interview the participants: Radhika Maharjan, Dipika Khadka, Anita Bashyal, Samina Amatya, and Roshani Kafle. We also want to thank Dr. Rohini Nepal and Jugmaya Chaudhary of Rhythm Neuropsychiatry Hospital and Research Centre for their contribution to advising and reviewing the self-reporting psychological assessment tools used in the questionnaire. We would also like to thank Transcultural Psychosocial Organisation (TPO) Nepal and Dr. Nabaraj Koirala for the permission to use the Nepali-validated version of BDI I and BAI for the study. We additionally thank Nilu Dullewe, who helped in coding the qualitative data. For the ongoing mutual support for improvements in ICU care, we would also like to acknowledge and thank members of the CCAA.

CCAA members

Diptesh Aryal, Shirish KC, Kanchan Koirala, Subekshya Luitel, Rohini Nepal, Sushil Khanal, Hem R Paneru, Subha K Shreshta, Sanjay Lakhey, Samina Amatya, Kaveri Thapa, Radhika Maharjan, Roshani Kafle, Anita Bashyal, Reema Shrestha, Dipika Khadka and Nilu Dullewe.

This study was funded by a Wellcome Innovations Flagship Programme grant (Wellcome grant number: 215522/Z/19/Z). They had no role in the design, analysis, or reporting of this protocol.

Author information

Authors and affiliations.

Nepal Intensive Care Research Foundation, Kathmandu, Nepal

Shirish KC, Diptesh Aryal, Kanchan Koirala & Subekshya Luitel

Institute of Applied Health Research, University of Birmingham, Birmingham, UK

Tiffany E. Gooden

Centre for Inflammation Research, University of Edinburgh, Edinburgh, UK

Rashan Haniffa & Abi Beane

Mahidol Oxford Tropical Medicine Research Unit, Bangkok, Thailand

You can also search for this author in PubMed   Google Scholar

• Diptesh Aryal
• , Shirish KC
• , Kanchan Koirala
• , Subekshya Luitel
• , Rohini Nepal
• , Sushil Khanal
• , Hem R Paneru
• , Subha K Shreshta
• , Sanjay Lakhey
• , Samina Amatya
• , Kaveri Thapa
• , Radhika Maharjan
• , Roshani Kafle
• , Anita Bashyal
• , Reema Shrestha
• , Dipika Khadka
•  & Nilu Dullewe

Contributions

All authors conceptualised this study. SK, DA, AB, RH, and SL developed the protocol, study methods, and materials. KK and SL facilitated the data collection, supervised by SK and DA. Data were analysed by SK, AB, KK, and TEG. SK and TEG wrote the drafts of the manuscript, and all authors reviewed the manuscript and consented to it being submitted. AB is the senior author.

Corresponding author

Correspondence to Diptesh Aryal .

Ethics declarations

Ethics approval and consent to participate.

Ethics approval was granted from the Nepal Health Research Council (approval number: 176/2021 P). All participants provided informed consent electronically before completing the online questionnaire. Participants from the qualitative component provided further informed verbal consent before the interview commenced.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Members of the CCAA are listed in the acknowledgments

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

KC, S., Gooden, T.E., Aryal, D. et al. The burden of anxiety, depression, and stress, along with the prevalence of symptoms of PTSD, and perceptions of the drivers of psychological harms, as perceived by doctors and nurses working in ICUs in Nepal during the COVID-19 pandemic; a mixed method evaluation. BMC Health Serv Res 24 , 450 (2024). https://doi.org/10.1186/s12913-024-10724-7

Download citation

Received : 05 September 2023

Accepted : 14 February 2024

Published : 10 April 2024

DOI : https://doi.org/10.1186/s12913-024-10724-7

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Pandemic preparedness, psychological distress
• Healthcare professionals
• Critical care

BMC Health Services Research

ISSN: 1472-6963

• General enquiries: [email protected]

Anxiety Disorders

Find Your Local NAMI

Call the NAMI Helpline at

800-950-6264

Or text "HelpLine" to 62640

We all experience anxiety. For example, speaking in front of a group can make us anxious, but that anxiety also motivates us to prepare and practice. Driving in heavy traffic is another common source of anxiety, but it helps keep us alert and cautious to avoid accidents. However, when feelings of intense fear and distress become overwhelming and prevent us from doing everyday activities, an anxiety disorder may be the cause.

Anxiety disorders are the most common mental health concern in the United States. Over 40 million adults in the U.S. ( 19.1% ) have an anxiety disorder. Meanwhile, approximately 7% of children aged 3-17 experience issues with anxiety each year. Most people develop symptoms before age 21.

Anxiety disorders are a group of related conditions, each having unique symptoms. However, all anxiety disorders have one thing in common: persistent, excessive fear or worry in situations that are not threatening. People typically experience one or more of the following symptoms:

Emotional symptoms:

• Feelings of apprehension or dread
• Feeling tense or jumpy
• Restlessness or irritability
• Anticipating the worst and being watchful for signs of danger

Physical symptoms:

• Pounding or racing heart and shortness of breath
• Sweating, tremors and twitches
• Headaches, fatigue and insomnia
• Upset stomach, frequent urination or diarrhea

Types of Anxiety Disorders

There are many types of anxiety disorders, each with different symptoms. The most common types of anxiety disorders include:

Generalized Anxiety Disorder (GAD)

GAD produces chronic, exaggerated worrying about everyday life. This worrying can consume hours each day, making it hard to concentrate or finish daily tasks. A person with GAD may become exhausted by worry and experience headaches, tension or nausea.

Social Anxiety Disorder

More than shyness, this disorder causes intense fear about social interaction, often driven by irrational worries about humiliation (e.g. saying something stupid or not knowing what to say). Someone with social anxiety disorder may not take part in conversations, contribute to class discussions or offer their ideas, and may become isolated. Panic attacks are a common reaction to anticipated or forced social interaction.

Panic Disorder

This disorder is characterized by panic attacks and sudden feelings of terror sometimes striking repeatedly and without warning. Often mistaken for a heart attack, a panic attack causes powerful physical symptoms including chest pain, heart palpitations, dizziness, shortness of breath and stomach upset. Many people will go to desperate measures to avoid an attack, including social isolation.

We all tend to avoid certain things or situations that make us uncomfortable or even fearful. But for someone with a phobia, certain places, events or objects create powerful reactions of strong, irrational fear. Most people with specific phobias have several things that can trigger those reactions; to avoid panic, they will work hard to avoid their triggers. Depending on the type and number of triggers, attempts to control fear can take over a person’s life.

Other anxiety disorders include:

• Agoraphobia
• Selective mutism
• Separation anxiety disorder
• Substance/medication-induced anxiety disorder, involving intoxication or withdrawal or medication treatment

Scientists believe that many factors combine to cause anxiety disorders:

• Genetics.   Studies support the evidence that anxiety disorders “run in families,” as some families have a higher-than-average amount of anxiety disorders among relatives.
• Environment.  A stressful or traumatic event such as abuse, death of a loved one, violence or prolonged illness is often linked to the development of an anxiety disorder.

Physical symptoms of an anxiety disorder can be easily confused with other medical conditions, like heart disease or hyperthyroidism. Therefore, a doctor will likely perform an evaluation involving a physical examination, an interview and lab tests. After ruling out an underlying physical illness, a doctor may refer a person to a mental health professional for evaluation.

Using the Diagnostic and Statistical Manual of Mental Disorders (DSM) a mental health professional is able to identify the specific type of anxiety disorder causing symptoms as well as any other possible disorders that may be involved. Tackling all disorders through comprehensive treatment is the best recovery strategy.

Different anxiety disorders have their own distinct sets of symptoms. This means that each type of anxiety disorder also has its own treatment plan. But there are common types of treatment that are used. 

• Psychotherapy , including cognitive behavioral therapy
• Medications , including antianxiety medications and antidepressants
• Complementary health approaches , including stress and relaxation techniques

Related Conditions

Anxiety disorders can occur along with other mental health conditions, and they can often make related conditions worse. So, talk with a mental health care professional if you are experiencing anxiety and any of the following:

• Substance Use
• Attention Deficit Hyperactivity Disorder ( ADHD )
• Eating Disorders
• Trouble Sleeping

Reviewed December 2017

Advertisement

• Register for Free
• My account Orders Downloads Address Payment methods Account details
• About About Psychiatrist.com About JCP About PCC About CME Institute

Error: Search field were incomplete.

Weekly Mind Reader: Mindfulness, Problematic Social Media Use, and Anxiety

by Staff Writer April 19, 2024 at 9:53 AM UTC

Research suggests trait mindfulness-based interventions might help reduce anxiety levels in students with severe problematic social media use .

Mindfulness, Problematic Social Media Use, and Anxiety

The Primary Care Companion of CNS Disorders just published a study that investigates the relationship between problematic social media use (PSMU) and anxiety among Lebanese university students. The study also explored the potential moderating role of trait mindfulness.

The researchers surveyed 363 participants and assessed PSMU, trait mindfulness, and anxiety levels.

The survey results showed a strong positive association between PSMU and anxiety, including a negative link between trait mindfulness and anxiety.

Additionally, the researchers found that trait mindfulness eased the tension between PSMU and anxiety. The study showed that, at low mindfulness levels, an apparent link between higher PSMU and elevated anxiety levels. However, those levels fell with the implementation of moderate and high levels of mindfulness, despite similar levels of PSMU. These results suggest that trait mindfulness might buffer the negative effects of PSMU on anxiety symptoms among university students.

The study’s strengths include its focus on a specific outcome – anxiety – related to social media use, as well as its examination of a population that garnered limited attention in earlier research.

The results suggest the importance of addressing PSMU and its threat to the mental health of university students. School-based interventions that focus on mindfulness training might help mitigate anxiety symptoms associated with PSMU.

Overall, the paper highlights the need for a deeper dive into underlying the relationship between PSMU and anxiety. The researchers suggest that future research could employ longitudinal designs and multidimensional measures of mindfulness to further elucidate these relationships.

IN OTHER PSYCHIATRY AND NEUROLOGY NEWS

• Recent technological advances have allowed researchers to quantify antipsychotic-induced cervical dystonia severity using machine learning, potentially establishing a global standard for future AI devices in cervical dystonia assessment.
• Theta burst stimulation interventions could hold the key to addressing specific multiple sclerosis symptoms, notably fatigue and spasticity.
• The Primary Care Companion for CNS Disorders has published several case studies that explore various aspects of OCD .
• European and American regulators have found no evidence linking GLP-1 receptor agonists like Ozempic and Wegovy to thoughts of suicide .
• Research shows exercise reduces cardiovascular disease risk by reducing stress-related signals in the brain.

Severity of Antipsychotic-Induced Cervical Dystonia Assessed by the Algorithm-Based Rating System

Rater consensus data were compared with deviation angle data using AI-based deviation angle measurement technology. With the range of tilt angles found in the study, the authors propose a global standard for evaluating abnormal deviation severity in cervical dystonia for future d...

Toshiya Inada and others

Unlocking Therapeutic Potential: The Role of Theta Burst Stimulation in Multiple Sclerosis Management

Theta burst stimulation interventions may hold promise in addressing specific multiple sclerosis symptoms, notably fatigue and spasticity.

David F. Lo and others

Search Articles

Related articles, table of contents.

2nd April 2024 Newsletter

This week we’re releasing another two resources from our new Schema series: Defectiveness and Coping Styles And Responses (Schema Therapy) . Our research roundup examines the impact of early life experiences on eating disorders, and explores how negative expectations can increase social anxiety.

Psychology Tools Schema Series

Launched earlier this year, our schema series breaks down the foundational ideas of schema therapy in an approachable and straightforward way. It also addresses the most common early maladaptive schemas (EMS) with a range of problem-specific handouts.

With illustrations, client-friendly explanations and schema-specific examples, the handouts offer accessible guidance for those working with schema therapy in-session. They’re a great learning tool for anyone wanting to become more familiar with this approach.

Defectiveness

Individuals with defectiveness experience a pervasive sense of inadequacy and inferiority. They see themselves as fundamentally flawed and defective, leading to deep-seated shame and unworthiness. Research suggests that it is one of the most pronounced early maladaptive schemas across psychological disorders, and is readily transmitted between parents and their children.

Coping Styles And Responses (Schema Therapy)

Coping styles refer to the characteristic ways individuals manage their EMS. They develop in childhood and operate outside of awareness, helping individuals adapt to their EMS, the intense affect accompanying them, and the environments in which they were formed. Much like EMS, factors that may influence the emergence of coping styles include temperament, modeling, conditioning, and culture.

Research Summaries

What role do early life experiences play in eating disorders?

What role do early life experiences play in the development of eating disorders? To address this question, a recent study explored the relationship between perceived parenting, temperament, early maladaptive schemas, and eating disorder symptoms. The findings revealed a direct linear impact pathway: maladaptive parenting predicted temperament, which in turn predicted maladaptive schemas, which ultimately influenced eating disorder symptoms. These results not only support the schema therapy model, but also offer valuable insights for broad, tailored treatment approaches for eating disorders.

"When treating patients with eating disorders, psychologists and other health professionals should consider individual differences in perceived parenting style, temperament, and early maladaptive schema levels and how these may impact the way an eating disorder manifests… More specifically, understanding the interplay of factors that contribute to the development of an eating disorder can potentially counteract the sense of personal shame and inadequacy that is often associated with a diagnosis."

Joshua, P. R., Lewis, V., Simpson, S., Kelty, S. F., & Boer, D. P. (2024). What role do early life experiences play in eating disorders? The impact of parenting style, temperament and early maladaptive schemas. Clinical Psychology & Psychotherapy , 31(1), e2904.

The legacy of social anxiety-linked negative expectancy

For therapists working with clients experiencing social anxiety, understanding the link between negative expectations, anxiety during interactions, and post-event negative thinking is crucial. This study indicates that increased negative expectations before social events amplifies state anxiety during interactions, which predicts post-event negative thinking. By recognizing the mediating role of pre-event negative expectancies, therapists can tailor interventions to address this aspect of social anxiety and the impact it has on post-event processing. However, it also highlights the need for further research exploring factors that contribute to post-event negativity.

Basanovic, J., Kowal, L., Millward, S., & MacLeod, C. (2024). The legacy of social anxiety-linked negative expectancy: A pathway from pre-event negative expectancies to post-event negative thinking. Journal of Behavior Therapy and Experimental Psychiatry , 83, 101937.

Get sent more useful pieces like this!

Sign up to our monthly newsletter full of helpful ideas, tools, and tips for mental health professionals like you. You'll find out about our latest resources, and you can also read our reviews of the latest research.

It's completely free and you can unsubscribe at any time.

• For clinicians
• For students
• Resources at your fingertips
• Designed for effectiveness
• Resources by problem
• Translation Project
• Help center
• Try us for free
• Terms & conditions
• Privacy Policy
• Cookies Policy