Functional gastrointestinal disorders (FGIDs) often present with prominent somatic symptoms and high prevalence of psychological comorbidity. This study aimed to examine psychosomatic status in FGID outpatients and test the hypothesis that depression and anxiety are closely associated with somatic symptom burden. A secondary aim was to assess whether depressive symptoms modify the association between anxiety symptoms and somatic symptoms.
This was a retrospective cross-sectional study conducted at the Gastroenterology Clinics of the Featured Medical Centre of Army Medical University (Daping Hospital). Data were collected from patients who attended the clinic between March 2021 and December 2024. The institutional review board granted a waiver of informed consent for this retrospective chart-review study.
Inclusion required outpatient diagnosis of FGIDs and completion of the Self-Rating Anxiety Scale (SAS), Self-Rating Depression Scale (SDS), Symptom Checklist-90 somatisation subscale (SCL-90-SOM) and heart rate variability (HRV) testing. Exclusions included missing required data, presence of serious organic lesions or critical illnesses preventing completion of scales, and age <18 years. Demographic data recorded were age and gender.
SAS and SDS are 20-item self-report scales quantifying past-week anxiety and depressive symptom severity; standard cut-offs used were SAS ≥50 for anxiety symptoms and SDS ≥53 for depressive symptoms. The SCL-90-SOM subscale is a 12-item cluster assessing somatic symptom burden, including headaches, gastrointestinal distress and musculoskeletal pain.
RR interval signals were recorded in a relaxed, seated 5-minute protocol using the PEM-D Psychosomatic Holistic Assessment and Intervention System. Frequency-domain analysis produced low- and high-frequency components, and the reported HRV-derived ANS index refers to the system’s frequency-domain output used as a marker related to autonomic regulation.
Analyses used complete cases only; no imputation was performed. Continuous variables are presented as mean±SD or median and IQR as appropriate; categorical variables as counts and percentages. Univariable linear regression screened associations with SCL-90-SOM; variables with p<0.05 were entered into multivariable linear regression. Continuous predictors were standardised when constructing the interaction term. An interaction model tested the SAS×SDS term. All tests were two-sided with p<0.05 considered significant; no correction for multiple testing was applied.
The analytic sample comprised 655 patients (mean age 45.07 years), 381 female (58.17%) and 274 male (41.83%). Anxiety symptoms (SAS ≥50) were present in 198 patients (30.23%), and depressive symptoms (SDS ≥53) in 378 patients (57.71%). Mean scores were: SAS 45.79±9.98, SDS 53.63±11.76, and SCL-90-SOM 1.84±0.60. The HRV-derived ANS index median was 0.50 (25th percentile 0.33, 75th percentile 0.97).
In univariable linear regression, female gender, higher SAS, higher SDS, and lower HRV-derived ANS index were associated with higher SCL-90-SOM scores; age showed no significant association.
In multivariable linear regression adjusting for available covariates, female gender and higher SAS remained positively associated with SCL-90-SOM. The coefficient for SDS became negative after inclusion of SAS in the model, and the HRV-derived ANS index was no longer statistically significant. Spearman correlation showed a strong positive correlation between SDS and SAS (rho=0.653, 95% CI 0.614 to 0.723, p<0.001). Both SAS and SDS were positively associated with SCL-90-SOM in single-variable models; however, when entered together, the SDS coefficient reversed sign.
An interaction term between standardised SAS and SDS was significantly associated with somatic symptoms (B=0.042, 95% CI 0.018 to 0.067, p<0.001). In the interaction model, standardised SAS had a strong positive association with SCL-90-SOM (B=0.451, p<0.001). Simple-slope analysis showed that when depressive symptoms were high (+1 SD), the positive association between anxiety and somatic symptoms was stronger (steeper slope). When depressive symptoms were low (−1 SD), the association was weaker (gentler slope). The authors note the negative SDS coefficient in adjusted models should not be interpreted as depressive symptoms reducing somatic symptoms but likely reflects shared variance, suppression effects, or symptom-scale overlap.
This outpatient FGID sample showed relatively high rates of anxiety and depressive symptoms and a measurable somatic symptom burden. Female gender and anxiety symptoms were independently associated with greater somatic symptom scores after adjustment. The significant positive interaction between anxiety and depression indicates that higher depressive symptom levels magnify the association between anxiety and somatic symptoms.
The HRV-derived ANS index related to somatic symptoms in univariable analysis but lost significance after covariate adjustment; the authors suggest HRV associations may be affected by transient physiological states, medications, sleep and other unmeasured factors.
Strengths: the study used a substantial sample size (n=655) from a gastroenterology clinic and standardised, validated self-report instruments (SAS, SDS, SCL-90-SOM), supporting measurement reliability.
Limitations: the cross-sectional design precludes causal inference. Single-centre recruitment may limit generalisability. Residual confounding is possible because important variables—gastrointestinal symptom severity and duration, psychiatric diagnoses or history, medication use (antidepressants, anxiolytics, antispasmodics, proton pump inhibitors, prokinetics), sleep disturbance, major medical comorbidities, and lifestyle factors—were not systematically available. Selection and reporting biases are possible because analyses used only patients who completed all scales. Symptom-scale overlap between anxiety and depression measures may complicate interpretation of adjusted coefficients.
In this single-centre outpatient FGID sample, anxiety symptoms were strongly associated with higher somatic symptom burden, and depressive symptoms modified this relationship: higher depressive symptom levels amplified the association between anxiety and somatic symptoms. Interpretation is limited by the cross-sectional design, potential symptom-scale overlap and residual confounding. The findings support further prospective studies and suggest that assessing co-occurring anxiety and depressive symptoms may be clinically relevant in the care of patients with FGIDs.