Prevalence and associated factors of somatic symptoms among adolescents in Singapore: a cross-sectional study

Abstract

Background

Somatic symptoms are physical symptoms that often arise in response to emotional distress and can significantly impact well-being. Understanding the prevalence and interplay of these symptoms with anxiety and depression is crucial for understanding adolescent health outcomes in Singapore. This study aimed to determine the prevalence of somatic symptoms among adolescents in Singapore and examine their associations with demographic characteristics, life stressors, anxiety, and depression.

Methods

We conducted a cross-sectional online survey with 601 Singaporean adolescents aged 12 to 19, using the Patient Health Questionnaire Somatic, Anxiety, and Depressive Symptoms Scales (PHQ-SADS) to assess somatic symptoms, anxiety, and depression levels. Demographic information and experiences with stressful life events were also collected. Both descriptive and regression analyses were used to understand the relationships between participants’ sociodemographic factors, mental health symptoms, and the presence of somatic symptoms.

Results

32.3% of the adolescents reported clinically significant levels of somatic symptoms, with a higher prevalence observed among those who were females and older. While no significant associations were found between somatic symptoms and demographic factors, adolescents with anxiety or depressive symptoms presented significantly increased odds of reporting somatic symptoms. Specifically, the odds of reporting somatic symptoms were 2.91 times greater for those with anxiety (95%CI: 1.55–5.45, p < 0.001) and 6.54 times greater for those with depression (95%CI: 3.75–11.6, p < 0.001) than for those without these mental health concerns. Furthermore, those with somatic symptoms reported a greater number of stressful life events, with academic pressure emerging as the most prominent stressor.

Conclusions

This study highlights the prevalence of somatic symptoms among adolescents in Singapore, emphasising the interconnectedness of mental and physical health during this developmental stage. The strong associations between somatic symptoms, anxiety, depression, and life stressors underscore the need for a holistic approach to adolescent healthcare. Early identification and intervention strategies should focus on addressing mental health concerns, building resilience against stressors, and promoting healthy coping mechanisms to mitigate the burden of somatic symptoms and foster overall well-being in Singaporean adolescents.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12991-025-00582-w.

Background

Adolescence is a period marked by significant physical, emotional and social changes. For some, these changes can manifest as physical symptoms without a clear medical explanation, known as somatic symptoms, which often signal underlying stress and vulnerability [1, 2]. While somatic symptoms may exist independently of psychological factors, psychosomatic symptoms specifically arise when mental or emotional states contribute to this physical manifestation of distress. Common somatic symptoms include headaches, fatigue, muscle soreness, nausea, back pain, and abdominal pain, all of which can significantly impact daily functioning [3]. These symptoms place substantial strain on healthcare systems. Individuals with somatisation incur approximately twice the annual costs and utilisation for outpatient and inpatient services compared to those without such symptoms [4]. Beyond the healthcare costs, repeated and often unnecessary investigations and treatments can disrupt adolescents’ daily lives, highlighting the dual burden on individuals and healthcare systems [4, 5].

Globally, the prevalence of somatic symptoms among adolescents ranges widely from 10 to 15% for specific complaints in population surveys [6] to up to 50% in lifetime estimates and clinical settings, depending on measurement criteria and cultural context [7–9]. The World Health Organization’s cross-national study found that medically unexplained symptoms such as body aches and pain were the most common complaints in clinical settings across five countries [9]. Research from Europe and Asia associates somatic symptoms with gender, age, and school-related stress, with girls and older adolescents reporting more incidents of somatic symptoms [10–13]. Somatic symptoms are also closely associated with psychopathology, revealing comorbidity between somatisation and anxiety and depression problems [14, 15]. Psychosomatic symptoms have also been shown to rise with academic pressure, as demonstrated in a study across 36 countries, primarily in Europe [16]. These findings underscore the interplay between developmental stressors and sociocultural norms in how distress is expressed. Understanding these patterns in non-Western contexts is crucial, particularly in societies where emotional expression of distress may be more stigmatised or internalised.

In Singapore, earlier studies have alluded to the presence of somatic symptoms in children. For example, 12.2% of children aged 6–12 years were found to have internalising problems, including somatic complaints [17]. More recent research shows that adolescents with frequent primary care visits often present with acute physical complaints, despite underlying psychiatric issues like anxiety and depression [18]. According to a recent National Youth Mental Health Study, 33.8% of adolescents aged 15–19 reported severe anxiety, and 18.8% reported severe depression [19]. However, few local studies have examined how somatic symptoms are patterned across demographic and psychological factors, limiting our understanding of their specific risk profiles within Singaporean adolescents.

Studying somatic symptoms in the Singapore context is important not only due to the lack of local evidence but also because cultural values may shape how adolescents express distress. In high-performing, collectivist Asian societies like Singapore, where emotional disclosure may be discouraged and academic achievement heavily emphasised, adolescents may be more inclined to express psychological discomfort through somatic symptoms [20]. Consecutive epidemiological studies have also consistently shown that younger age groups report a higher prevalence of poor mental health, reinforcing the need for age-specific research focus [21, 22]. Existing studies on somatic symptoms from other countries may not generalise to Singapore’s multicultural and multilingual setting. Thus, local data are essential for early identification of potential psychological distress and to inform contextually relevant screening and intervention strategies for adolescents during this critical developmental stage [23].

This study aims to quantify the prevalence of somatic symptoms among adolescents in Singapore and examine their associations with demographic factors (age, gender, ethnicity, housing type as a proxy for socioeconomic status) and psychological factors (anxiety, depression and life stressors). We postulate that up to one in three adolescents in Singapore experience somatic symptoms, based on findings in similar Asian contexts [24]. This hypothesis aligns with recent national surveys reporting that about one in three youths in Singapore experience depression, anxiety, or stress symptoms [25–27]. While local data do not specifically report somatic symptoms, these mental health symptoms can serve as valuable proxies for the potential prevalence of somatic symptoms, given the established link between mental health difficulties and somatic complaints.

Methods

Study design

This cross-sectional community survey was conducted online in Singapore between 19 March 2024 and 1 April 2024. Participants aged 12–19 years were recruited through a survey vendor using an online panel. Parental consent was obtained separately for adolescents aged 12 to 17 years, before the adolescent commenced the online survey. For all participants, implied consent was established when they voluntarily proceeded with the survey, which took approximately 10 min to complete. A target sample size of 600 was determined to achieve a 5% margin of error with a 95% confidence level based on an estimated adolescent population of 420,000 in Singapore.

To ensure a nationally representative sample, demographic quotas for age, gender and race were employed based on a national population registry and allowing for a 5% variance. The inclusion criteria for the study were as follows: (1) Singaporean citizenship or permanent residency; (2) age between 12 and 19 years; and (3) ability to read and understand English.

Measures

Demographics

We collected data on age, gender, race and housing type. Age was measured in years and grouped into early (12–13), middle (14–16), and late adolescence (17–19), based on developmental stages described in the literature [28]. Race and housing type were classified using the standard categories established in the Singapore population census. Housing type was grouped into four categories based on cost, which serves as proxies for socioeconomic status (SES): 1–3 room HDB flats (small public housing), 4–5 room HDB or executive flats (large public housing), condominiums (small private housing), and landed properties (large private housing).

Patient health Questionnaire - Somatic, anxiety, and depressive symptoms scales

The PHQ-SADS is a composite measure of somatic, anxiety, and depressive symptoms. It consists of the PHQ-15 for somatic symptoms, the GAD-7 for anxiety symptoms, and the PHQ-9 for depressive symptoms [29]. Higher scores on these subscales indicate greater symptom severity. Following Kroenke et al. [29], we used a cut-off score of 10 or higher on any of these subscales (somatic, anxiety, depression) to indicate the presence of clinically significant symptoms for the corresponding symptom cluster. Anxiety and depression scales were selected due to their established association with somatic symptoms and their high prevalence in adolescence. These brief screening tools (PHQ-9 and GAD-7) were used to minimise participant burden and ensure comparability with other adolescent health studies.

As the PHQ-SADS has not been validated among Singaporean adolescents, we conducted face validation to ensure local relevance and understandability. This involved consultations with domain experts to assess the relevance and appropriateness of the adolescent-modified scales, followed by cognitive interviews with adolescents to test the comprehensibility of the language used. The items were modified based on expert judgment, and another round of cognitive interviews was conducted to pretest for item meaningfulness and adequacy of the scale response (refer to Supplementary Materials 1 for face validity findings). The adolescent-modified scales have good face validity and high levels of internal consistency. The Cronbach’s alpha values were 0.91 for both the somatic and depression scales and 0.92 for the anxiety scale.

Somatic symptoms

Somatic symptoms were assessed via the PHQ-13, an adolescent-modified version of the PHQ-15 somatic scale, which is the first component of the PHQ-SADS. This version was chosen because it excludes two items about menstruation and sexuality, making it more suitable for use with a broad range of adolescent ages [30]. The participants rated their experience of 13 somatic symptoms over the past four weeks on a 3-point scale (0 = not bothered at all, 1 = bothered a little, 2 = bothered a lot).

Anxiety symptoms

Anxiety symptom severity was assessed via the Generalised Anxiety Disorder 7-item scale, a component of the PHQ-SADS. The GAD-7 has demonstrated strong psychometric properties, including reliability and validity [31]. Its suitability for adolescent populations is well established, with strong correlations between GAD-7 scores and those obtained via the clinician-administered Paediatric Anxiety Rating Scale [32]. The participants rated their experience of anxiety symptoms over the past two weeks on a 4-point scale (0 = not at all, 1 = several days, 2 = more than half the days, 3 = nearly every day). A “prefer not to answer” response option was added to each item because of the sensitive nature of some of the questions.

Depressive symptoms

Depressive symptoms were measured via the adolescent-modified version of the PHQ-9, another component of the PHQ-SADS. This version of the PHQ-A has similar sensitivity and specificity to those of adult samples [33, 34]. It aligns with recommendations for adolescent assessment in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR) [35]. Similar to the GAD-7, a “prefer not to answer” response option was added to each question on the PHQ-9. The participants rated the frequency of depressive symptoms experienced over the past two weeks on a 4-point scale (0="not at all,” 1="several days,” 2="more than half the days”, 3="nearly every day”).

Stressful life events

The study used a life stressor indicator adapted from the Adolescent Stress Questionnaire [36] and previously employed in the National Youth Survey in Singapore [37]. This indicator comprises nine items rated on a 5-point scale (1=“not at all stressful”, 2=“a little stressful”, 3=“moderately stressful”, 4=“quite stressful”, 5=“very stressful”). A “prefer not to answer” response option was also added to each item. The means and standard deviations are reported for the top three stress items to highlight the most impactful stressors.

Statistical methods

Descriptive statistics were summarised using means and standard deviations when continuous, frequencies and proportions (%) when categorical, and they were stratified by demographic characteristics. The prevalence of somatic symptoms was calculated as the percentage of participants scoring 10 or higher on the PHQ-13, indicating significant somatic symptoms [29]. The same cut-off was applied to the anxiety (GAD-7) and depression (PHQ-9) scores to identify significant symptoms in each category [38]. For analyses involving individual somatic symptom responses from the PHQ-13, responses were dichotomised as “yes” (bothered a little or bothered a lot) or “no” (not bothered at all) to estimate the presence of specific symptoms.

Additionally, each stress item was dichotomised to indicate the presence or absence of significant stress. Responses of “3 = moderately stressful”, “4 = quite stressful”, and “5 = very stressful” were coded as “1” (indicating the presence of stress), whereas responses of “1 = not at all stressful” and “2 = a little stressful” were coded as “0” (indicating the absence of significant stress). The total stress count was then calculated for each respondent by summing the coded values across all stress items, resulting in a range of 0–9, representing the total number of significant stressors reported.

Univariable and multivariable logistic regression analyses were conducted to determine the associations between sociodemographic factors, anxiety, depression, and the presence of somatic symptoms. Odds ratios (ORs) and 95% confidence intervals (CIs) were reported to quantify the strength of associations. Missing data were handled using complete case analysis, with a final sample size of 546. While Little’s MCAR (missing completely at random) test suggested that the data were not missing completely at random (p = 0.0037), missingness was limited to non-forced items (anxiety, depression, and stressors), with no missing data in demographics or the main outcome. No significant associations were found between missingness and demographic variables. Multicollinearity among the predictor variables was assessed using the variance inflation factor (VIF). All the VIFs were less than 2 (mean VIF = 1.50), indicating no significant multicollinearity concerns [39]; thus, these variables were not excluded from the model. The significance level was defined as p < 0.05.

All analyses were performed using RStudio version 2024.04.0 + 735. This study’s reporting followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [40].

Ethical approval

This study was reviewed and approved by the SingHealth Centralised Institutional Review Board (CIRB reference number: 2023/2508).

Results

Descriptive characteristics of the study participants

A total of 601 adolescents completed the online survey, representing a 34% response rate. The mean age of participants was 15.9 years (SD = 2.2), with the majority aged between 14 and 19 years (82.6%). Table 1 presents the demographic characteristics of the entire sample. The majority of participants were male (51.2%), of Chinese ethnicity (72.7%) and resided in large public housing (4–5 room HDB and executive flats) (55.1%).

Table 1.

Sociodemographic characteristics of study participants (N = 601)

Characteristic	n	%
Age (mean, sd)	15.9	2.2
12–13	105	17.5
14–16	245	40.8
17–19	251	41.8
Gender
Female	292	48.6
Male	308	51.2
Others	1	0.2
Ethnicity
Chinese	437	72.7
Malay	109	18.1
Indian	44	7.3
Others	11	1.8
Housing type
1–3 room HDB1 flats	109	18.1
4–5 room HDB/Executive flats	331	55.1
Condominiums	139	23.1
Landed Properties	22	3.7

¹ HDB (Housing and Development Board) refers to Singapore’s public housing authority, which provides government-subsidised apartments for Singaporean residents.

Prevalence rates of somatic symptoms

The prevalence rate of somatic symptoms in the study sample was 32.3% (95% CI: 28.7–36.1), as determined by a PHQ-13 score of 10 or higher. The most prevalent somatic symptoms were “feeling tired or having little energy” (n = 395, 65.7%), “trouble sleeping” (n = 324, 53.9%), and “stomach pain” (n = 294, 48.9%). Chest pain and fainting spells had the lowest prevalence rates.

Table 2 presents the distribution of somatic symptoms according to sociodemographics and mental health-related factors. Older adolescents (aged 17–19 years) demonstrated a greater prevalence of somatic symptoms (37.1%) than their younger counterparts did (28.6% for those aged 12–13 years and 29.0% for those aged 14–16 years). Similarly, females were more likely to report somatic symptoms (34.9%) than males (29.9%), while a higher prevalence of somatic symptoms was observed among adolescents of Malay ethnicity (42.2%) than among Chinese adolescents (29.8%). Additionally, adolescents residing in private housing (condominiums and landed properties) reported a greater prevalence of somatic symptoms (36.0%) than those in public housing (30.5–32.1%).

Table 2.

Prevalence of somatic symptoms by socio-demographic and mental health-related factors

	Proportion with Somatic symptoms
Characteristic	n	%	95%CI
Age
12–13	30	28.6	20.8–37.9
14–16	71	29.0	23.7–35.0
17–19	93	37.1	31.3–43.2
Gender
Female	102	34.9	29.7–40.6
Male	92	29.9	25.0–35.2
Ethnicity
Chinese	130	29.8	25.6–34.2
Malay	46	42.2	33.3–51.6
Indian	13	29.6	18.2–44.2
Others	5	45.5	21.3–72.0
Housing type
1–3 room HDB flats	35	32.1	24.1–41.4
4–5 room HDB / Executive flats	101	30.5	25.8–35.7
Condominiums	50	36.0	28.5–44.2
Landed properties	8	36.4	19.7–57.0
Anxiety
No	106	22.5	18.9–26.4
Yes	77	70.0	60.8–77.8
Depression
No	81	18.8	15.4–22.8
Yes	96	69.1	60.9–76.1

In terms of mental health-related factors, adolescents reporting anxiety and depressive symptoms had higher proportions of somatic symptoms (70.0% and 69.1%, respectively) than those without these symptoms.

Anxiety and depressive symptoms

The overall prevalence of anxiety symptoms in our study sample was 18.3% (95% CI: 15.4–21.6), whereas the overall prevalence of depressive symptoms was 23.1% (95% CI: 19.9–26.7). Compared with their younger counterparts, older adolescents (aged 17–19) presented a greater prevalence of both anxiety and depressive symptoms. Specifically, 22.9% of the older adolescents reported anxiety symptoms, whereas 11.0% of the 12–13-year-olds and 16.6% of the 14–16-year-olds reported anxiety symptoms. Similarly, 30.0% of the older adolescents reported depressive symptoms, whereas 17.0% and 19.7% of the younger adolescents reported depressive symptoms.

Life stressors

Across participants, the highest-rated stressor was ‘studies’ (mean = 2.9, SD = 1.2), followed by ‘future uncertainty’ (mean = 2.5, SD = 1.2) and ‘emerging adult responsibility’ (mean = 2.5, SD = 1.2). This pattern of top stressors remained consistent across the three age groups. The detailed means and standard deviations for all life stressors are available in the supplementary materials (see Table S2). ‘Family relationships’ and ‘personal health’ had the lowest ratings. Older adolescents reported a greater number of stressors (median = 4, IQR = 1–7) than their younger counterparts (12–13 years old: median = 1, IQR = 0–4; 14–16 years old: median = 2, IQR = 0–5).

Additionally, those with somatic symptoms reported a greater number of stressors (median = 6, IQR = 4–8) than those without somatic symptoms (median = 1, IQR = 0–3). Similarly, those with anxiety or depressive symptoms reported a greater number of stressors (median = 7, IQR = 5-8.5) compared to those without anxiety or depressive symptoms (median = 1, IQR = 0–4). Overall, the majority of the participants (53.1%) reported at least two life stressors, whereas more than a quarter (26.7%) reported experiencing six or more life stressors.

Logistic regression

Table 3 presents findings from the logistic regression analyses. According to the univariable analyses, no sociodemographic variables were significantly associated with somatic symptoms. However, the presence of anxiety and depressive symptoms was significantly associated with somatic symptoms. These associations remained significant in the multivariable analyses, even after adjusting for all other variables in the model. The odds of reporting somatic symptoms were 2.91 times greater for those with anxiety (95% CI: 1.55–5.45, p < 0.001) and 6.54 times greater for those with depression (95% CI: 3.75–11.6, p < 0.001) than for those without these mental health concerns. The association between depressive and somatic symptoms remained significant even after removing overlapping items (fatigue and sleep disturbance) from the somatic symptom scale. In the revised model, the odds ratio decreased to 5.32 (95% CI: 2.98–9.60) (see supplementary materials, Table S3). Additional analyses modelling anxiety and depression as outcomes revealed that while somatic symptoms were significantly associated with both, sociodemographic factors were not (see Tables S4 and S5 in the supplementary materials).

Table 3.

Regression results for predictors of somatic symptoms (N = 546)

	Univariable			Multivariable
Characteristic	OR1	95% CI1	p-value	OR1	95% CI1	p-value
Age			0.055			0.570
12–13	-	-		-	-
14–16	0.97	0.57–1.67		0.86	0.47–1.59
17–19	1.54	0.93–2.62		1.11	0.61–2.04
Gender			0.405			0.346
Female	-	-		-	-
Male	0.86	0.59–1.23		0.81	0.52–1.26
Ethnicity			0.213			0.074
Chinese	-	-		-	-
Malay	1.49	0.93–2.36		1.84	1.05–3.17
Indian	0.69	0.29–1.49		0.64	0.24–1.57
Others	1.61	0.41–5.74		2.12	0.46–8.46
Housing Type			0.583			0.589
1–3 room HDB flats	-	-		-	-	-
4–5 room HDB/Executive flats	1.11	0.67–1.88		1.17	0.64–2.22
Condominiums	1.41	0.79–2.53		1.44	0.73–2.91
Landed Properties	1.52	0.55–3.98		1.90	0.59–5.72
Anxiety			< 0.001			< 0.001
No	-	-		-	-
Yes	9.27	5.73–15.37		2.91	1.55–5.45
Depression			< 0.001			< 0.001
No	-	-		-	-
Yes	11.10	7.07–17.64		6.54	3.75–11.58

¹ OR = Odds Ratio, CI = Confidence Interval

Discussion

Our study revealed that an estimated 32.3% of community-dwelling adolescents in Singapore reported clinically significant somatic symptoms. This prevalence rate aligns with broader concerns about youth mental health in Singapore [25–27], emphasising the need for focused attention to somatic presentations of distress. Our findings provide a localised perspective on the global trends observed from research focusing mainly on European countries [16]. The most frequently reported somatic symptoms were fatigue (65.7%), sleep problems (53.9%), and stomach pain (48.9%). These symptoms have also been prominently observed in adolescents experiencing distress in other contexts, such as following trauma [30]. This overlap suggests that fatigue, sleep disturbances, and stomach pain might represent common pathways through which adolescents, in particular, express distress, regardless of their specific origin.

Trends with demographic characteristics

In our study, older adolescents (aged 17–19 years) presented a greater prevalence of somatic symptoms (37.1%) than younger adolescents (28–29%). Similarly, anxiety and depression are more common in older adolescents, which is consistent with trends in Singaporean primary school children aged 8–12 years [41]. Older adolescents may face unique stressors, such as academic pressures, pubertal changes, relationship or sexuality issues, and future uncertainties, which could contribute to a higher burden of somatic symptoms. Additionally, more females reported somatic symptoms (34.9%) than males (29.9%), consistent with the literature suggesting a greater sensitivity to stress, possibly due to hormonal changes or socialisation patterns that encourage emotional expression [10, 42, 43].

Somatic symptoms were more prevalent among Malay adolescents (42.2%) than among their Chinese counterparts (29.8%), possibly influenced by cultural factors, such as norms around the expression of distress [44] and stressors linked to minority identity [45]. These findings highlight the role of cultural nuances in symptom expression, suggesting further exploration of ethnic differences in somatic symptom reporting [41, 46]. Moreover, adolescents in private housing, indicative of higher SES, reported a greater prevalence (36.0%) of somatic symptoms than those in public housing (30.5–32.1%). Higher SES backgrounds may lead to unique stressors, such as pressure to meet high expectations, which could manifest as somatic symptoms [47].

However, caution is needed when interpreting these findings, as housing type alone may not fully represent SES, and the wide confidence intervals add uncertainty to the prevalence estimate. These patterns suggest that other factors, such as psychological stressors or life events, may play a more direct role in somatic symptom development. While demographic characteristics provide context, they may function as distal factors that do not directly influence the manifestation of somatic symptoms in adolescents. Instead, proximal factors, such as psychological and environmental influences, play a more immediate and critical role [48]. For example, school-related stressors and poor family relationships can serve as immediate triggers for somatic symptoms, with demographic factors framing these experiences rather than directly causing them [49].

Mental health-related factors

Our findings revealed a strong association between somatic symptoms and the presence of anxiety and depressive symptoms in adolescents. The prevalence of somatic symptoms was significantly greater among those reporting anxiety (70%) and depression (69.1%) than among their counterparts (22.5% and 18.8%, respectively). Regression analyses revealed that adolescents with anxiety were 2.91 times more likely, and those with depression were 6.54 times more likely to report somatic symptoms. While overlapping fatigue and sleep-related items in the depression and somatic symptom scales likely inflated these associations, as these were among the most commonly reported somatic symptoms, supplementary analysis confirmed that the association persisted even after these items were removed. Factors such as circadian rhythm shifts, poor sleep hygiene, and early school start time can exacerbate sleep difficulties in adolescents [50], contributing to daytime fatigue and, consequently, to the heightened reporting of somatic symptoms. Additionally, generalised anxiety disorder, which is often comorbid with depression, has been linked to insufficient sleep [51], further contributing to this potential confound.

While these results underscore the strong association between anxiety and depression with somatic symptoms, our supplementary regression analyses modelling anxiety and depression as outcomes showed consistent patterns, with somatic symptoms also associated with both anxiety and depression. Although the present study is cross-sectional and does not permit causal inference, these findings align with longitudinal studies demonstrating bidirectional relationships where somatic symptoms predict future anxiety and depression, highlighting their role as early indicators of mental health challenges [52, 53].

Several mechanisms may explain this relationship between somatic symptoms and mental health conditions. Shared biological vulnerabilities, such as heightened stress reactivity, may contribute to both somatic distress and mental health difficulties [54]. Dysregulation of neurotransmitters such as serotonin and norepinephrine, which influence both pain and mood, may underlie the well-evidenced link between pain and depression [55]. Psychological factors, such as pain catastrophising and emotional dysregulation, also play a role [56]. Furthermore, the heightened focus on bodily sensations often associated with anxiety can lead to misinterpretation of typical physical experiences as threatening, thus perpetuating somatic complaints [57].

Recognising these complex interrelationships is crucial for clinicians, as somatic symptoms may not only indicate a physical disorder but also reflect culturally influenced expressions of distress that require careful interpretation to ensure appropriate and holistic treatment [46, 58]. This underscores the need for ongoing mental health education and training for primary care physicians and paediatricians [59], particularly in settings such as polyclinics and outpatient paediatric clinics where adolescents with somatic symptoms often present [18]. Given that young people may struggle to articulate emotional distress, early screening and appropriate right-siting of care in these settings offer valuable opportunities to assess mental health, address underlying stressors, and ensure timely intervention and holistic care [60, 61].

Life stressors

Our findings showed that adolescents with somatic symptoms reported significantly more life stressors (median = 6) compared to those without (median = 1), and this trend was also observed in those with anxiety and depression, who reported a median of 7 stressors. The top-rated stressor was academic pressure (mean = 2.9, SD = 1.2), followed by future uncertainty and emerging adult responsibilities, consistent with local data [62]. These results also align with international studies that reported a strong association between academic pressure and psychosomatic symptoms, suggesting that school-related stress is a significant contributor across different cultures [16, 24].

In Singapore, where adolescents experience high academic demands, the pressure to excel in schoolwork may exacerbate somatic symptoms, potentially through unhealthy behaviours such as poor sleep and physical inactivity [63]. While academic stress was the most prominent (refer to supplementary materials, Table S2), the accumulation of various stressors, particularly as adolescents transition into adulthood, likely plays a central role in the development of somatic complaints. Older adolescents, facing heightened academic and social demands as they navigate this transition, may be particularly vulnerable. Sexuality issues and identity fluidity in this current generation can also contribute to stress [64] and may manifest as somatic symptoms because of the stigma of coming out. Their heightened sensitivity to stressors, influenced by developmental changes in stress processing systems, including hormonal fluctuations [65], may explain the higher prevalence of somatic symptoms in this age group. Tailoring interventions to reduce both academic pressure and the cumulative effect of various stressors could be key to addressing the interconnectedness of stress and somatic symptoms.

Limitations and future research

This study offers valuable insights into the prevalence of somatic symptoms and associated factors, although some limitations should be considered. While the adapted PHQ-SADS has shown preliminary validity, further research is needed to examine its psychometric properties to confirm its reliability and validity. Although the depression and somatic symptoms scales included overlapping items related to fatigue and sleep problems, which could potentially inflate the observed associations, supplementary analyses excluding these items showed that the associations remained robust. Nonetheless, future studies should consider using distinct, non-overlapping measures to more precisely assess these associations.

Our focus on internalising symptoms, anxiety and depression, was guided by their strong theoretical and empirical links to somatic symptoms and their relevance to early intervention efforts [66, 67]. However, limiting the scope to these domains may have excluded other important psychological and contextual factors. Future research should incorporate a broader range of variables, such as perceived stress, coping styles and family or school-related influences, to better understand the multifaceted nature of somatic symptoms in adolescents.

Additionally, comparing the prevalence of somatic symptoms across countries is challenging because of varying assessment tools, hindering accurate cross-cultural comparisons. As this is the first study in Singapore to assess somatic symptoms in adolescents, future research should aim to use the same measure to monitor changes locally over time. Standardised screening tools within Singapore and internationally would enable more meaningful comparisons and a deeper understanding of trends across different cultures and time points.

While this study used a locally relevant life stressor indicator for comparisons with the National Youth Survey, future investigations should adopt more robust and comprehensive tools to assess life stressors. This would allow for stronger comparisons across different populations and a clearer understanding of how various stressors contribute to the manifestation of somatic symptoms. Additionally, incorporating measures related to lifestyle behaviours and physical health, such as sleep, physical activity, and diet, could provide valuable insights into the broader contextual factors that influence somatic symptom development [63, 68]. A more comprehensive approach to assessing both stress and lifestyle factors could enhance intervention strategies aimed at improving adolescent well-being.

Conclusion

This study is the first to measure the prevalence of somatic symptoms among adolescents in Singapore, revealing that 32.3% reported clinically significant symptoms. While demographic factors do not directly explain these symptoms, our findings underscore the crucial role of mental health and life stressors in understanding these physical manifestations of distress. The strong associations between somatic symptoms, anxiety, and depression suggest a complex interplay in which these conditions may exacerbate one another, emphasising the need for a holistic approach to adolescent healthcare that recognises the interconnectedness of mental and physical well-being. Future research should focus on developing culturally sensitive, multifaceted strategies that address academic stress, build resilience against accumulated life stressors, and promote mental health literacy among adolescents, their families, and educators. Understanding and addressing the factors contributing to somatic symptoms in Singaporean adolescents has the potential to reduce the burden of these symptoms and promote holistic well-being during this critical developmental period.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

PHQ-SADS

Patient Health Questionnaire Somatic, Anxiety, and Depressive Symptoms Scales

CI

Confidence Interval

HDB

Housing and Development Board

SES

Socioeconomic status

GAD-7

Generalised Anxiety Disorder 7-item scale

DSM-5-TR

Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR)

OR

Odds Ratio

MCAR

Missing Completely At Random

VIF

Variance Inflation Factor

STROBE

Strengthening the Reporting of Observational Studies in Epidemiology

IQR

Interquartile Range

Author contributions

Conceptualisation – DM, VJM, HYC, SCS, YCLH, CBLL; Methodology – DM, YCLH, CBLL; Project administration – DM, MSLC, NP, CZHH; Data acquisition, analysis, and interpretation – DM, MSLC, NP, CZHH; Supervision – JML, CBLL; Writing – original draft – DM, JML, CBLL; Writing – review and editing – all authors. All authors reviewed and approved the final manuscript.

Funding

This research is supported by the National Medical Research Council (NMRC) through the SingHealth PULSES II Centre Grant (CG21APR1013).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Studies involving human participants were reviewed and approved by the SingHealth Centralised Institutional Review Board (CIRB; reference number: 2023/2508). Written informed consent was not required to participate in this study in accordance with the national legislation and institutional requirements.

Competing interests

The authors declare no competing interests.

Consent for publication

Not applicable.