Discussion

The present study demonstrated a high prevalence of social anxiety among medical students, with 201 of 507 participants (39.64%) meeting the predefined SIAS cutoff. The prevalence of social anxiety identified in the present study, 201 (39.64%), was substantially higher than the 3.6% current prevalence of anxiety disorders reported in the adult Indian general population by the National Mental Health Survey of India [3]. However, this comparison should be interpreted cautiously because the national estimate represented anxiety disorders collectively, whereas the present study specifically assessed social anxiety symptoms using the SIAS. The prevalence found in this study is consistent with previous studies suggesting that medical students represent a psychologically vulnerable group because of sustained academic demands, frequent assessments, and highly competitive learning environments [3,9]. This prevalence was substantially higher than the 7.8% reported by Honnekeri et al. among urban Indian undergraduate students using the SIAS and Social Phobia Scale [10] and the 9.6% prevalence reported by İzgiç et al. among Turkish university students [6]. However, it was lower than the 62.5% prevalence reported by Obadeji and Kumolalo among Nigerian undergraduate students [7]. These comparisons indicate that estimates of social anxiety among university students vary considerably across settings. Differences in the instruments used, cutoff scores, participant characteristics, academic disciplines, cultural attitudes toward social interaction, and timing of data collection may account for this variation [5-7]. However, direct comparisons should be interpreted cautiously because prevalence estimates vary according to study population, cultural context, screening instrument, cutoff score, and whether symptoms are identified through self-report screening or diagnostic interviews. The high prevalence observed in the present study is consistent with the broader evidence that students in demanding academic programmes experience a substantial burden of anxiety and psychological distress [9].

Several features of medical training may help explain the relatively high prevalence identified in this study. Medical students are repeatedly exposed to written examinations, oral viva examinations, bedside assessments, clinical case presentations, interactions with unfamiliar patients, and direct evaluation by faculty members and peers. These situations closely resemble the scrutiny-based and interpersonal circumstances that commonly provoke social anxiety. Sustained academic workload, competition, fear of committing errors, concern about professional competence, and limited time for rest may further increase vulnerability to psychological distress [4]. Anonymous self-administered screening may also identify students with significant symptoms who have not sought clinical help. Consequently, the present prevalence estimate represents screening-positive social anxiety and should not be interpreted as the prevalence of a formally diagnosed psychiatric disorder.

Low self-esteem was identified in 134 participants (26.43%). This estimate was comparable to the 24.1% prevalence reported by Alghamdi et al. among 1,099 medical students in Saudi Arabia [13]. The principal finding of this study was the strong association between low self-esteem and social anxiety. Participants with low self-esteem had more than fivefold higher odds of social anxiety in the unadjusted analysis (OR = 5.57), and this association remained statistically significant after adjustment for gender (adjusted OR = 5.76, 95% CI: 3.71-8.93). Obadeji and Kumolalo similarly reported that social anxiety scores were negatively correlated with self-esteem among undergraduate students [7]. İzgiç et al. also found significantly lower self-esteem among university students with social phobia than among those without social phobia [6]. In a college sample, Abdollahi and Talib reported that lower self-esteem was associated with greater social anxiety [8]. These studies support the present finding that impaired self-esteem is closely related to social anxiety across different student populations.

The continuous-score analysis further supported this association. Total SIAS and RSES scores showed a moderate-to-strong negative relationship in the present sample (Pearson’s r = −0.595 and Spearman’s ρ = −0.576; both p < 0.001). The magnitude of this relationship was stronger than the correlation reported by Ayed et al. among nursing students [12]. Differences in participant characteristics, scoring methods, symptom severity, academic setting, and sociocultural background may partly explain the variation in effect size. Importantly, the continuous relationship indicates that the association is not limited to students who cross a categorical cutoff; progressively lower self-esteem was associated with progressively higher social anxiety scores.

The relationship between the two constructs can be understood through cognitive and self-presentational models of social anxiety. Schlenker and Leary proposed that social anxiety develops when individuals are motivated to make a favourable impression but doubt their ability to do so successfully [5]. Students with low self-esteem may interpret themselves as socially inadequate, underestimate their communication ability, anticipate rejection or criticism, and monitor their behaviour excessively during interactions. This heightened self-focused attention can increase anticipatory anxiety and encourage avoidance. Although avoidance may produce short-term relief, it prevents corrective social experiences and may reinforce both negative self-beliefs and anxiety over time [5]. Therefore, low self-esteem may act as both a vulnerability factor and a maintaining correlate of social anxiety; however, the cross-sectional design of this study prevents determination of the temporal or causal direction.

Female participants had significantly higher odds of low self-esteem than male participants (OR = 2.31, 95% CI: 1.47-3.63), although gender was not independently associated with social anxiety. This finding is consistent with Alghamdi et al., who identified female gender as an independent predictor of low self-esteem among medical students [12]. Previous research has also shown that female medical students may report lower confidence and greater anxiety regarding competence despite academic performance comparable to that of male students [14,15,16]. Possible explanations include differences in self-appraisal, interpersonal expectations, perceived performance pressure, and sensitivity to external evaluation. Nevertheless, because socioeconomic status, age, academic year, academic performance, psychiatric history, and other potential confounders were not collected in the present study, the observed gender association should be interpreted cautiously.

Most participants reported onset of social anxiety- or low-self-esteem-related difficulties between 15 and 19 years of age. This finding is consistent with epidemiological evidence that social anxiety commonly begins during adolescence and early adulthood [2,15,9]. Adolescence involves major changes in identity, peer relationships, educational demands, and sensitivity to social evaluation. Early identification is particularly relevant in medical education because persistent social anxiety may interfere with classroom participation, oral examinations, communication training, clinical interaction, teamwork, and professional confidence.

The present study has several strengths. It included a relatively large sample of medical students from multiple MBBS admission batches, achieved an 84.5% response rate, and used established psychometric instruments. The relationship between social anxiety and self-esteem was examined using complementary categorical, correlational, and multivariable methods. The Social Interaction Anxiety Scale and Rosenberg Self-Esteem Scale also demonstrated excellent internal consistency in the present sample, with Cronbach’s alpha values of 0.923 and 0.870, respectively.

The study also has important limitations. Its cross-sectional design precludes causal inference. Questionnaire responses were self-reported and may therefore have been affected by recall, response, and social-desirability bias. The instruments were used for screening and cannot substitute for structured clinical assessment. The study was conducted at a single institution, limiting generalizability. Variables such as current age, socioeconomic status, academic year, academic performance, psychiatric history, place of residence, and social support were not collected, which restricted adjustment for potential confounding. Furthermore, the characteristics of the 93 non-respondents were unavailable, and non-response bias cannot be excluded. Future multicentre longitudinal studies incorporating broader demographic, academic, and clinical variables are needed to clarify the direction of the relationship and evaluate interventions aimed at strengthening self-esteem and reducing social anxiety.