Examining the role of social support in imposter phenomenon among medical students through mediation and moderation analysis of anxiety and stress Authors: Maryam Fatima, Manahil Qamar, Hira Irfan, Hurais Malik, Faraheem Batool, Jawaria Shahzad, Muhammad Shoaib Alam, Muhammad Usama Jamil, Alishba Jamil, Sumaira Mohiuddin, Muhammad Abdullah, Zuhair Ahmed, Abdul Rehman, Ayat ul Karam, Usaid Malik, Khabab Abbasher Hussien Mohamed Ahmed Published: 2025 ABSTRACT AbstractBackgroundImposter syndrome (IS) is characterized by individuals’ inability to internalize success and persistent feelings of being intellectual frauds despite evident achievements. Though not a clinically recognized disorder, IS significantly affects psychological well-being, particularly among medical professionals and students, leading to anxiety, stress, and burnout. In Pakistan, the relationship between IS, its psychological drivers, and protective factors remains underexplored. This study examines how perceived social support relates to IS, with a specific focus on the mediating and moderating roles of anxiety and stress.MethodsA multi-center cross-sectional study was conducted among 649 medical students from four medical institutions in Pakistan. Validated tools were used: the Clance IP Scale, Oslo Social Support Scale (OSSS-3), and the Anxiety and Stress Sub-Scale of DASS-21. Data were analyzed using hierarchical regression, correlation matrices, and mediation/moderation modeling via the PROCESS macro in SPSS v26.ResultsMean IS scores (M = 61.23, SD = 16.46) indicated frequent impostor feelings. Social support negatively correlated with IS (r = − 0.141, p < 0.001), anxiety (r = − 0.130, p = 0.001), and stress (r = − 0.132, p = 0.001). In regression, social support initially predicted IS (β = − 0.141, p < 0.001), but became non-significant when anxiety and stress were included. Anxiety (β = 0.351, p < 0.001) and stress (β = 0.670, p < 0.001) were significant predictors. Mediation analysis confirmed that both anxiety and stress significantly mediated the social support–IS relationship. Moderation analyses showed no significant interaction effects (p > 0.05).ConclusionSocial support reduces IP primarily by alleviating anxiety and stress, rather than through direct buffering. Targeted interventions that address both emotional distress and enhance supportive environments are essential for fostering psychological resilience in medical education.Keywords: Imposter syndrome, Imposter phenomenon, Social support, Anxiety, Stress, Medical students, Psychological well-being, Prevalence, Pakistan, Mediation, Moderation Introduction The imposter phenomenon (IP), also known as imposter syndrome (IS), was first defined by Clance and Imes in 19,781 [1]. It is characterized by a failure to internalize success and viewing oneself as an impostor, affecting individuals facing achievement tasks regardless of their success, status, or gender. IS manifests as an internal experience of intellectual fraud, where individuals attribute their academic or occupational achievements primarily to luck, error, or personal connections rather than their diligent efforts [2], thus perpetuating a vicious cycle that hinders their ability to develop an optimistic future perspective [3]. Despite not being a recognized clinical illness or psychiatric disease, IS significantly impacts psychological well-being and can lead to physical and professional damage, like burnout, early retirement, and increased suicide rates [1]. IS is prominent among highly successful individuals. According to a recent review in the United States, the prevalence of the syndrome ranges from 22% to 60% among physicians [4] and 15% among medical students [5]. However, within the Pakistani context—where medical education is highly competitive, expectations are immense, and mental health resources are often limited—the mechanisms underlying IP remain underexplored. Pakistan is no exception to these trends; a recent study from Pakistan reported a 62.6% prevalence among medical students [6]. Medical students are a high-risk group for IP, which has garnered increasing attention due to its profound and damaging implications on both individuals and health profession institutions [7]. The development of imposter sentiments may be influenced by several factors, including gender, perceived social expectations, poor coping strategies, low self-esteem, maladaptive perfectionism, poor sleep quality, anxiety, depression, burnout, and suicidal ideation [6]. Consequently, sociodemographic characteristics and mental health are critical factors associated with IS. Despite growing recognition of the IP, its underlying mechanisms remain poorly understood [8]. While literature highlights its psychological toll, there is a lack of research examining how psychosocial factors such as anxiety, stress, and social support interact to influence IP, particularly in medical students. Furthermore, there is limited consensus on whether social support exerts a direct protective effect or operates through intermediary variables like emotional distress. Most existing studies are cross-sectional and concentrated in high-income countries, revealing a geographic and contextual limitation in the evidence base. Additionally, the absence of IP from formal psychiatric classifications has contributed to fragmented research efforts and inconsistent methodologies. In Pakistan, this knowledge gap is even more pronounced. There is a conspicuous absence of empirical studies investigating the associations between IP and key psychological constructs—especially anxiety, stress, and perceived social support—among medical students. Given the high prevalence of mental health issues and academic stress in this population, this represents a practical problem with implications for student well-being and institutional outcomes. Therefore, the present study aims to (1) assess the prevalence of IP among Pakistani medical students, (2) evaluate its association with perceived social support, anxiety, and stress, and (3) explore whether anxiety and stress mediate or moderate the relationship between social support and IP. Methodology Study design This cross-sectional study was conducted at four prominent healthcare institutions: Fazaia Ruth Pfau Medical College in Karachi, Baqai Medical University in Karachi, Nishtar Medical University in Multan, and Bakhtawar Amin Medical College in Multan, from March 2024 to June 2024. Ethical approval was obtained from the institutional review and ethics boards of the respective institutions. We focused on the provinces of Punjab and Sindh, which boast the highest concentration of medical institutes among the country’s four distinct regions; moreover, including both government and private institutes. This decision was made to maximize the representativeness of our sample within the medical student population. A convenience sampling strategy was employed, using online distribution through institutional email lists and social media platforms (e.g., WhatsApp groups). This approach facilitated wide accessibility while maintaining voluntary and anonymous participation. Given the self-report nature of all variables, measures were taken to reduce common method bias. Anonymity and confidentiality assurances were provided to reduce social desirability bias; procedural separation was employed by organizing the questionnaire into clearly labeled sections to reduce item priming, and validated scales with strong psychometric properties were selected. Participants The study focused on medical students from the 1st to 5th year of the Bachelor of Medicine, Bachelor of Surgery (MBBS) program, including both pre-clinical and clinical years. Participants included medical students aged 18 years or above who provided consent to participate. Non-medical students, those who did not give consent, and participants requiring special communications were excluded from the study. The minimum sample size of 370 participants required for this study was determined using the WHO Sample Size Calculator 2.0. Parameters entered included an expected prevalence of 47.5% for Imposter Phenomenon (based on prior Pakistani data) [9], a 95% confidence level, and a 5% margin of error. However, we collected data from 649 students, all of whom provided consent and were interviewed, resulting in a cooperation rate of 100%. No imputation methods were utilized, and all responses were included in the analysis. Procedures Data collection involved a comprehensive questionnaire comprising four sections. Section I: socio-demographic information The first section of the questionnaire collected demographic information and background details from the participants. This section included questions about age, gender, residence, number of siblings, birth order, family system, year of study, and monthly household income. Section II: clance IP scale The Clance Impostor Phenomenon (IP) Scale was developed by Dr. Pauline Rose Clance in 1985 to measure the intensity of IS characteristics [10]. The scale consists of 20 items; each rated on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). It assesses individuals’ perceptions of their competence and the fear of being exposed as frauds despite evident success. In the context of this research, the Clance IP Scale is used to determine the prevalence and severity of IS among medical students. Scores are interpreted as follows: 40 or less indicates few IP characteristics, 41–60 indicates moderate IP experience, 61–80 indicates frequent imposter feelings, and above 80 indicates intense IP experience. Higher scores represent more severe IP experiences. The reliability of the Clance IP Scale has been demonstrated with a high Cronbach’s alpha score of 0.92. Section III: oslo social support scale (OSSS-3) The Oslo Social Support Scale (OSSS-3) was developed by Dalgard et al. in 2006 to measure social support through three items: the number of close confidants, the level of interest and concern shown by others, and the ease of obtaining practical help from neighbors [11]. The response options vary for each item, contributing to a total score ranging from 3 to 14. This scale is used in the current study to evaluate the level of social support available to medical students, which is hypothesized to influence the prevalence and intensity of IS. Interpretation of the OSSS-3 scores is as follows: 3–8 indicates poor social support, 9–11 indicates moderate social support, and 12–14 indicates strong social support. Higher scores reflect stronger social support networks. The reliability of the OSSS-3 has been validated with a Cronbach’s alpha score of 0.60, which is slightly below the conventional threshold of 0.70 but consistent with prior international validations of the OSSS-3 that have reported moderate internal consistency due to its brevity and multidimensional structure. Despite this limitation, the OSSS-3 is considered a reliable and efficient tool for large-scale surveys assessing perceived social support in community and student populations. Section IV: depression, anxiety, and stress scales (DASS-21) The DASS-21, developed by Lovibond and Lovibond in 1995, is a set of three self-report scales designed to measure the emotional states of depression, anxiety, and stress [12]. This shortened version of the original 42-item scale comprises 21 items, with each item rated on a 4-point Likert scale from 0 (Did not apply to me at all) to 3 (Applied to me very much or most of the time). For this study, the focus is on the anxiety and stress subscales, each containing 7 items. The DASS-21 is being used to assess the levels of anxiety and stress among medical students, which may mediate or moderate the relationship between social support and IS. Interpretation involves summing the scores for each subscale, with higher scores indicating greater severity of symptoms. Cutoff points categorize severity levels: normal (0–7 for anxiety, 0–14 for stress), mild (8–9 for anxiety, 15–18 for stress), moderate (10–14 for anxiety, 19–25 for stress), severe (15–19 for anxiety, 26–33 for stress), and extremely severe (20 + for anxiety, 34 + for stress). The DASS-21 has shown excellent reliability with a Cronbach’s alpha score of 0.89 for anxiety and 0.86 for stress. Context and usage in research These scales were chosen for their relevance and reliability in measuring the constructs of interest in this study. The Clance IP Scale provides a direct measure of IS, which is the primary outcome of interest. The DASS-21 assesses the mediating and moderating variables of stress and anxiety, which are critical for understanding the broader psychological context of IS. The OSSS-3 measures the perceived social support, a key independent variable hypothesized to impact IS. Collectively, these scales enable a comprehensive assessment of how social support, mediated or moderated by stress and anxiety, influences IS among medical students. The reliability of each scale, confirmed by high Cronbach’s alpha scores, ensures the robustness and accuracy of the measurements in this study. Outcomes The primary outcome of the study was the prevalence and severity of IS among medical students, assessed using the Clance Impostor Phenomenon (IP) Scale. The study also evaluated secondary outcomes, including the levels of social support, anxiety, and stress among medical students, measured using the Oslo Social Support Scale (OSSS-3) and the Depression, Anxiety, and Stress Scale (DASS-21). These outcomes were analyzed across different demographic variables such as gender, year of study, and socioeconomic background to determine their association with IS. Statistical analysis Data were analyzed using IBM-SPSS version 26.0 (IBM Corp., Armonk, NY, USA). The data analysis for the study was conducted in several stages to ensure a thorough understanding of the influence of social support on IP and the mediating or moderating role of Anxiety and Stress on it. Descriptive statistics Categorical variables, including socio-demographics (age, gender, residence, number of siblings, birth order, family system, year of study, and monthly household income), the categorization of IP scores, social support levels, and anxiety and stress scores, were reported as counts and percentages. Continuous variables, such as the mean scores of IS, social support, anxiety, and stress, were reported as mean and standard deviation (SD). Normality testing The normality of key continuous variables—IP scores, anxiety, stress, and social support scores—was assessed using both visual and statistical methods. Visual inspection involved histograms, Q-Q plots, and boxplots, while the Shapiro-Wilk test was applied due to its robustness for small to moderate sample sizes. Results indicated that several variables, including IP scores (p < 0.001), anxiety scores (p < 0.001), and stress scores (p < 0.001), deviated significantly from normality. Given these findings, non-parametric tests were selected to ensure statistical appropriateness and robustness. Correlation analysis To examine the relationships between the scales used (Clance IP Scale, DASS-21, and OSSS-3), Kendall’s Tau correlation was employed due to the ordinal nature of the Likert-scale-based variables and the presence of tied ranks in the dataset. Kendall’s Tau is particularly robust in handling ordinal data and small sample variations, providing a more conservative and reliable measure of association for non-parametric data. Correlation coefficients and significance levels were reported to identify significant associations between IS, social support, anxiety, and stress. Regression analysis Multiple regression analyses were conducted to explore the predictive relationships between social support (independent variable) and IS (dependent variable), with anxiety and stress included as covariates. Standardized regression coefficients (β) and R-squared (R²) values were reported to assess the strength and significance of these relationships. Mediation and moderation analysis Mediation analysis was performed to determine whether the relationship between social support and IS was mediated by anxiety and stress. The Model 4 of the PROCESS macro for SPSS (version 4.0) was used to test indirect effects, and bootstrapped 95% confidence intervals (5,000 samples) were used to test the significance of indirect and interaction effects. The Model 1 of the PROCESS macro for SPSS (version 4.0) was used to examine moderation. The moderation analysis was conducted to examine whether the strength of the relationship between social support and IS varied as a function of anxiety and stress levels. Interaction terms were included in the regression models, and significant interactions were further explored through simple slopes analysis. Despite the use of parametric techniques in multiple regression and PROCESS macro modeling, the large sample size (n = 649) provides sufficient power, and the Central Limit Theorem supports the use of these methods even under non-normality, particularly when bootstrapping techniques are employed, as was the case in our mediation and moderation analyses. Visualization and statistical significance Lateral stacked bar graphs were created to visualize the distribution of response options across the Clance IP Scale, anxiety, and stress scales. These visualizations helped in understanding the prevalence and severity of IS, anxiety, and stress among the participants. A p value of < 0.05 was considered statistically significant for all analyses. This threshold ensured that the findings were robust and less likely to be due to chance. Role of the funding source There was no funding source for this study. Ethical consideration The study obtained ethical approval from the Institutional Review Board (IRB) of FRPMC (Ref No: FRPMC-IRB-2024-55), BMU (BMU-IREB/09-2024/005), NMU (NMU-7121), and BAMDC (2277-24/E.C./BAM&DC). The consent section contained a debriefing statement in the Google Forms, promoting transparency and communication. They were fully informed about the study’s purpose, the voluntary nature of their participation, and the measures taken to protect their privacy and confidentiality. Participants were made aware of their right to withdraw from the study at any point without consequences. Data security measures were rigorously implemented to safeguard participant information. To ensure the privacy and confidentiality of participants, all data collected was de-identified and stored securely. Personal information such as names, contact details, or other identifiers was not linked to the collected data. Only the research team had access to the data, and any published results or reports would not contain any information that could identify individual participants. The data was stored on secure servers and password-protected computers. Results A total of 649 medical students were recruited and included in the analysis. Table 1 provides a detailed breakdown of their baseline characteristics. Participants were predominantly aged 18 to 25 years, with a slight female majority (56.7%). Most students resided in urban areas (84.3%) and belonged to nuclear families (77%). The majority had 1 to 4 siblings (75.3%), with 40.2% being the eldest child. Academic years were evenly distributed, with the first (24.5%) and fifth year (24.8%) having the highest representation. Socioeconomic status varied, with 41.4% of participants coming from households earning above 2 lakh per month. Table 1Baseline profile against imposter phenomenonSociodemographic variablesFreq (%)Few ICModerate ICFrequent ICIntense IC Age * 18 to 21 years336 (51.8)231411294322 to 25 years308 (47.5)481161103426 to 29 years5 (0.8)0230 Gender * Male281 (43.3)411089636Female368 (56.7)3015114641 Residence Rural102 (15.7)8413518Urban547 (84.3)6321820759 No. of Siblings None11 (1.7)05601 to 4489 (75.3)52200176615 to 8141 (21.7)185156169 to 128 (1.2)1340 Birth order of participants Elder261 (40.2)201129930Middle244 (37.6)32919031Younger144 (22.2)19565316 Family system Nuclear500 (77)5620318457Joint149 (23)15565820 Year of study ** 1st year159 (24.5)137356172nd year127 (19.6)94255213rd year71 (10.9)4372464th year131 (20.2)125545195th year161 (24.8)33526214 Monthly household income less than 50 thousand44 (6.8)41715850 thousand−1 lac142 (21.9)7655812Above 1 lac–2 lac194 (29.9)27756725Above 2 lacs269 (41.4)3310210232** Significant at the 0.01 level (2-tailed).* Significant at the 0.05 level (2-tailed).IC, impostor characteristics Baseline profile against imposter phenomenon ** Significant at the 0.01 level (2-tailed). * Significant at the 0.05 level (2-tailed). IC, impostor characteristics The mean Impostor Phenomenon Scale score was 61.23 (SD = 16.46), indicating a high frequency of imposter feelings. The mean anxiety score was 17.25 (SD = 11.37), while the mean stress score was 19.45 (SD = 9.99), reflecting significant levels of severe anxiety and moderate stress. The mean social support score was 8.63 (SD = 2.40), suggesting a low perceived quality of social support among participants. Regarding perceived social support, 44.1% of students reported having 3–5 close individuals for personal support, while 32% had more than five. Additionally, 33.4% of students felt that others showed some interest in their activities, while 19.3% perceived a high level of interest. However, a significant proportion (39.3%) felt uncertain or had little interest from others. Regarding practical help from neighbors, 38.7% found it possible, while 15.7% found it easy to obtain assistance; however, 35.4% found it difficult or very difficult. Cross-tabulation of the Impostor Phenomenon across Social Support Levels (Fig. 1) revealed that participants with poor social support had significantly higher levels of impostor feelings (118 reported frequent impostor feelings, and 38 reported intense impostor characteristics). Conversely, those with strong social support had lower impostor scores (20 reported frequent impostor feelings, and 8 reported intense impostor characteristics). Distribution of IP across PSS As shown in Table 2, IP demonstrated significant positive correlations with both anxiety (r = 0.536, p < 0.01) and stress (r = 0.585, p < 0.01), indicating that individuals experiencing higher levels of imposter feelings also tended to report greater anxiety and stress. A strong correlation was observed between anxiety and stress itself, suggesting a high degree of co-occurrence between these psychological states. In contrast, perceived social support was negatively correlated with all three variables: IP (r = – 0.141, p < 0.01), anxiety (r = – 0.130, p < 0.01), and stress (r = – 0.132, p < 0.01). These associations suggest that stronger perceived social support may serve a buffering role, potentially mitigating psychological distress and imposter-related feelings. Table 2Correlation matrix across scalesScalesI.II.III.IVImposter phenomena–0.536**0.585**− 0.141**Anxiety–0.705**− 0.130**Stress–− 0.132**PSS–*p ≤ 0.05, **p ≤ 0.01, PSS, perceived social support Correlation matrix across scales *p ≤ 0.05, **p ≤ 0.01, PSS, perceived social support A hierarchical regression analysis was conducted to determine the independent and combined effects of social support, anxiety, and stress on IS. In Model 1, social support alone accounted for 2% of the variance in IS (R2 = 0.020). However, adding anxiety and stress in Model 2 significantly improved the explained variance (ΔR2 = 0.356, p < 0.001). Notably, the effect of social support became non-significant (β = -0.056, p = 0.076), while anxiety (β = 0.351, p < 0.001) and stress (β = 0.670, p < 0.001) emerged as strong predictors. Mediation analyses (Table 3; Fig. 2) examined whether anxiety and stress mediated the relationship between social support and IS. Higher social support was associated with lower anxiety (β = − 0.618, p = 0.0009), and anxiety, in turn, was positively associated with IS (β = 0.763, p < 0.001). The indirect effect of social support on IS via anxiety was significant (β = − 0.471, 95% CI [–0 .786, − 0.165], R2 = 0.29, p < 0.001), indicating that social support reduces imposter feelings partly by lowering anxiety. Table 3Mediation analysis of the effect of social support on imposter syndrome with anxiety as a mediatorOutcome variablePredictorCoefficient (β)p value95% CI lower95% CI upperAnxietyConstant22.5846< 0.000119.337225.832Social support− 0.61790.0009− 0.9805− 0.2552Imposter syndromeConstant52.3634< 0.000147.831656.8953Social support− 0.49740.0302− 0.9472− 0.0477Anxiety0.7628< 0.00010.66790.8578Indirect effect on imposter syndromeAnxiety (mediator)− 0.4713–− 0.7864− 0.1648StressConstant24.189< 0.000121.33427.0439Social support− 0.54950.0008− 0.8683− 0.2307Imposter syndromeConstant46.6356< 0.000142.048651.2226Social support− 0.44730.0427− 0.8798− 0.0148Stress0.949< 0.00010.84521.0528Indirect effect on imposter syndromeStress (mediator)− 0.5215–− 0.8766− 0.1904 Mediation analysis of the effect of social support on imposter syndrome with anxiety as a mediator Mediation of social support on imposter syndrome via anxiety and stress Similarly, social support predicted lower stress (β = − 0.550, p = 0.0008), and stress was strongly associated with IS (β = 0.949, p < 0.001). The indirect effect via stress was significant (β = − 0.522, 95% CI [– 0. 877, − 0.190], R² = 0.346, p < 0.001). Overall, social support exerts both direct and indirect effects on IS by mitigating anxiety and stress, highlighting its protective role in medical students. Moderation analyses assessed whether anxiety or stress moderated the social support–IS relationship (Table 4). The model explained 29.3% of variance (R2 = 0.293, p < 0.001). Social support was negatively associated with IS but not statistically significant. Anxiety was a significant predictor; however, the interaction term was non-significant, indicating that anxiety does not moderate this relationship. Table 4Moderation analysis of the effect of social support on imposter syndrome with anxiety and stress as a moderatorVariableCoefficient (β)p value95% Confidence interval Anxiety (moderator) Constant50.7948< 0.00143.7457, 57.8440Social support− 0.31990.4079− 1.0786, 0.4387Anxiety0.8544< 0.0010.5253, 1.1835Social support × anxiety (Int_1)− 0.01060.5684− 0.0472, 0.0259 Stress (moderator) Constant46.2833< 0.00138.3827, 54.1839Social support− 0.40670.3525− 1.2650, 0.4517Stress0.9669< 0.0010.6240, 1.3099Social support × stress (Int_1)− 0.00210.9144− 0.0405, 0.0363 Moderation analysis of the effect of social support on imposter syndrome with anxiety and stress as a moderator The model accounted for 34.6% of variance (R2 = 0.346, p < 0.001). Stress significantly predicted IS, yet the interaction term between social support and stress was non-significant, suggesting no moderating effect. These findings indicate that while anxiety and stress mediate the effect of social support, they do not moderate it. Discussion The IP has emerged as a pervasive challenge within medical education, closely interwoven with the psychological burden of stress and anxiety. Across diverse cultural contexts, medical students report high prevalence rates of IP, with estimates up to 47% depending on the population studied [8]. From a mechanistic perspective, the medical training environment amplifies vulnerability to IP by fostering intense social comparison, continuous high-stakes evaluation, and an implicit culture of perfectionism. Previous studies suggest that medical students often compare themselves to idealized images of physicians and peers, reinforcing self-doubt [2]. External pressures, such as familial expectations and competitive academic environments, may exacerbate IP, making early interventions crucial [13]. These experiences are not merely isolated episodes of self-doubt but are strongly associated with heightened psychological distress, particularly anxiety and stress, which appear to function both as antecedents and consequences of IP [8]. However, resilience has been identified as a protective factor; targeted interventions, including cognitive-behavioral therapy and mentorship programs, are recommended to mitigate these effects [14]. Existing research consistently demonstrates a strong correlation between IP and psychological distress, with self-doubt and fear of being perceived as a fraud contributing to emotional burnout, depressive symptoms, and mental exhaustion [15, 16]. A Multilevel analysis in German schools demonstrates that both individual and classroom-level competitiveness significantly predict anxiety and stress among students [17]. Academic stress and performance anxiety, often rooted in high-achievement environments, create a psychological milieu conducive to impostor thoughts, especially in students who are high-performing but internally insecure [8]. The implications of these results are significant, as prolonged exposure to psychological distress can impair academic performance, hinder professional identity formation, and reduce long-term career satisfaction. To address these challenges, educational institutions should implement systemic reforms that foster resilience and psychological well-being among medical students. A key finding of this study is the critical role of social support in buffering the negative effects of IP. Prior research has established that individuals with poor social support are more susceptible to IP [18], reinforcing the importance of validation and belonging in mitigating self-doubt [19]. Several studies provide converging evidence on the significance of supportive environments. For instance, A study reports that medical students with strong family and peer networks were less likely to endorse IP despite high academic stress, suggesting that support functions as a resilience factor mitigating distress [20]. The institutional context further shapes the quality and accessibility of social support. Hierarchical cultures and competitive climates in medical schools may restrict opportunities for candid discussions about failure, inadvertently fostering isolation and impostor feelings. By contrast, structured mentorship, peer-support programs, and workshops normalizing impostor experiences have shown promise in reducing stigma and equipping students with strategies for self-compassion [16, 21]. Social support in these contexts not only reduces the psychological burden of IP but also enhances academic persistence and professional identity formation. Future research should explore specific mechanisms through which different forms of social support (e.g., peer vs. mentor) influence IP to inform targeted interventions that enhance resilience in academic settings. Social support appears to influence the development and intensity of the impostor phenomenon in medical students through multiple interconnected mechanisms. First, it exerts a stress-buffering effect, whereby sharing experiences of self-doubt with peers or mentors reframes challenges as common rather than personal deficiencies, weakening the cognitive basis of imposter feelings [22]. Second, supportive relationships facilitate adaptive emotion regulation, promoting strategies such as problem-solving and cognitive reframing while reducing reliance on avoidance or palliative coping, which in turn lowers anxiety that typically amplifies imposter fears [23]. Third, social support provides validation and reinforcement of competence, helping students align subjective self-assessments with objective performance, thereby correcting distorted self-perceptions and reducing anxiety-driven doubts about legitimacy [24]. Fourth, supportive networks mitigate social comparison stress inherent in competitive and hierarchical medical training environments by normalizing setbacks, encouraging collaboration, and decreasing maladaptive comparisons [25]. Finally, social connectedness may influence neurobiological stress regulation, as it is associated with lower cortisol levels and enhanced oxytocin signaling, thereby modulating physiological stress responses that otherwise sustain imposter feelings [26]. Collectively, these mechanisms illustrate how social support indirectly attenuates the impostor phenomenon by reducing psychological and physiological stressors. This underscores its protective role in promoting medical student well-being. Although the direct effect of social support on IP was statistically modest in our regression model, its role cannot be dismissed. The addition of anxiety and stress significantly increased the explained variance in IP, with both emerging as strong predictors. This shift underscores the complex, indirect pathway through which social support influences IP—primarily by mitigating psychological distress. Mediation analyses confirmed that higher social support was associated with lower levels of anxiety and stress, which in turn reduced the intensity of imposter feelings. This suggests that the practical utility of social support lies less in its direct predictive capacity and more in its buffering function against negative affective states that drive IP. These findings align with theoretical models that position social support as a resilience-enhancing factor within high-stress academic environments. Importantly, the modest direct effect size invites a broader discussion about unmeasured confounders such as maladaptive perfectionism, burnout, and personality traits, which may also contribute substantially to IP. Future interventions aimed at reducing IP among medical students should therefore adopt a multi-component strategy—enhancing social connectedness while simultaneously addressing individual psychological vulnerabilities. This holistic approach may yield more meaningful and sustainable reductions in imposter-related distress. Although the moderation analyses did not yield statistically significant interaction effects, several factors may explain this finding. It is possible that the OSSS-3’s brevity and limited internal consistency constrained its ability to capture the full spectrum of social support variability, limiting its power to detect interaction effects. Additionally, our sample’s relative homogeneity in terms of age, academic structure, and stress exposure may have narrowed the range of experiences, reducing effect sizes. Alternatively, the impact of stress and anxiety may function more strongly as mediators—influencing IS indirectly—rather than as moderators. This aligns with our mediation findings, where social support significantly reduced anxiety and stress, which in turn predicted lower IS scores. Future studies with more psychometrically robust instruments and broader sampling may better clarify these interaction dynamics. The present study underscores the pivotal role of social support in buffering the effects of anxiety and stress on the impostor phenomenon, providing actionable insights for medical education policy and practice. To address these psychological vulnerabilities, Pakistani medical schools should implement structured mentorship programs in which faculty and senior students provide personalized guidance, constructive feedback, and recognition of achievements, directly targeting competence-related self-doubt. Peer support networks can be formalized to facilitate shared experiences, normalize academic and professional challenges, and reduce maladaptive social comparisons, thereby alleviating stress and anxiety that mediate imposter feelings. Furthermore, faculty development initiatives should equip educators with skills to foster supportive learning environments, recognize early signs of psychological distress, and provide timely interventions. Integrating these evidence-based strategies into the curriculum and institutional support systems can enhance students’ resilience, improve mental well-being, reduce IP, and promote sustained academic and professional success. By explicitly linking interventions to the mechanisms identified in this study, these measures provide a targeted, theory-informed approach to cultivating a competent, confident, and psychologically healthy future medical workforce. This study has a few limitations that must be acknowledged. The cross-sectional design restricts causal inferences, and self-reported data may introduce response biases such as social desirability or recall bias. Additionally, we focused on the provinces of Punjab and Sindh, which host the highest concentration of medical colleges among Pakistan’s four provinces. To enhance diversity, we included both public and private sector institutions within these regions. While this sampling approach aimed to capture a broad spectrum of the medical student population, we acknowledge that limiting the study to two provinces may still not fully reflect the experiences of students in other regions, such as Khyber Pakhtunkhwa and Balochistan. Therefore, caution should be exercised when generalizing these findings to all medical students across Pakistan. We acknowledge that key variables such as perfectionism, personality traits, academic pressure, and burnout were not included in our model but may significantly influence IP. Their omission is a limitation and an avenue for future investigation. Lastly, anxiety and stress were treated as distinct constructs based on validated DASS-21 subscales; we did not conduct formal multicollinearity testing. Given their high correlation, future studies should evaluate multicollinearity (e.g., using VIF and tolerance) to ensure the stability and interpretability of regression and mediation estimates. Despite these limitations, this study offers several notable strengths that enhance its scientific and practical relevance. First, it is among the first multi-center studies in Pakistan to comprehensively investigate the interplay between social support, stress, anxiety, and IP among medical students, addressing a significant gap in the existing literature. Second, the study employed validated and widely recognized psychometric tools—including the Clance IP Scale, DASS-21, and OSSS-3—ensuring robust construct validity and internal consistency for most measures. Third, the large sample size (n = 649), drawn from four diverse medical institutions across two major provinces (Punjab and Sindh), improves the statistical power and contributes to the representativeness of the findings within the Pakistani context. Fourth, the use of advanced statistical modeling, including mediation and moderation analyses via the PROCESS macro, allowed for a nuanced examination of both direct and indirect pathways linking social support to IP, mediated through anxiety and stress. Lastly, this research provides actionable insights that could inform the development of targeted mental health and peer-support interventions in medical schools, making it both theoretically relevant and practically impactful. Future research should address these limitations by employing longitudinal designs to establish causal relationships between social support, IP, anxiety, and stress. While anonymity and confidentiality were maintained to reduce social desirability and recall bias, future studies may benefit from including multi-method data collection approaches (e.g., peer ratings, behavioral assessments, or institutional performance data) to validate self-reports. Expanding the sample to include medical students from diverse geographic and cultural backgrounds will enhance the generalizability of findings. Qualitative research could provide deeper insights into personal experiences of IP and the specific types of social support that are most effective. Additionally, interventional studies aimed at testing the efficacy of support programs in reducing impostor feelings and improving mental health outcomes would be valuable. Finally, integrating mental health education and support strategies into medical curricula should be explored as a proactive approach to mitigating IP and enhancing the well-being of future medical professionals. Conclusion This study demonstrates that social support indirectly mitigates IP in medical students by reducing anxiety and stress, highlighting the central role of negative affective states in perpetuating self-doubt. These findings underscore the need for medical institutions to move beyond generic well-being initiatives and implement targeted, evidence-based strategies. Practical steps include structured mentorship programs that provide guidance and competence validation, peer support networks that normalize challenges and reduce social comparison stress, and faculty training to identify and address early signs of anxiety and stress. By directly addressing the psychological mechanisms that sustain IP, such interventions can enhance resilience, foster confidence, and cultivate a mentally healthy and competent future medical workforce, particularly in resource-constrained contexts like Pakistan. Acknowledgements The authors have no acknowledgements to declare. Author contributions All authors read and approved the final manuscript. Funding The authors declare that no financial grants were obtained for the completion of this paper. Data availability The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The research adhered to ethical standards outlined in the Declaration of Helsinki of the World Medical Association. This study was reviewed and approved by the Institutional Review Board (IRB) of Fazaia Ruth Pfau Medical College (Ref No: FRPMC-IRB-2024-55), Baqai Medical University (BMU-IREB/09-2024/005), Nishtar Medical University (NMU-7121), and Bakhtawar Amin Medical and Dental College (2277-24/E.C./BAM&DC). Written informed consent was obtained from all participants prior to their inclusion in the study. Consent for publication All participants provided written informed consent for publication of their anonymized data. Clinical trail number Not Applicable, as this was an observational cross-sectional study, not a clinical trial. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.