Psychometric Properties of the Self-Report Instrument for Somatic Symptoms in General Hospitals

0
0
0

Psychometric Properties of the Self-Report Instrument for Somatic Symptoms in General Hospitals

Medically unexplained physical symptoms (MUPS) and related disorders are highly prevalent in healthcare-seeking populations, yet they often remain unrecognized or underdiagnosed. These conditions are associated with significant disability, healthcare costs, and prolonged untreated illness, with studies reporting a mean duration of untreated somatoform disorders exceeding 25 years. A critical barrier to early identification is the lack of effective screening tools tailored for use in general hospital settings. To address this gap, the Self-Screening Questionnaire for Somatic Symptoms (SQSS) was developed as a multidimensional instrument designed to assess somatic symptoms, negative perception, illness behavior, and social function in Chinese populations. This article evaluates the psychometric properties of the SQSS within general hospitals, focusing on its reliability, validity, and diagnostic utility.

Development and Structure of the SQSS

The SQSS was conceptualized to address cultural and clinical nuances in assessing MUPS among Chinese patients. It comprises four dimensions:

  1. Somatic Symptoms (SS): Physical manifestations reported by patients.
  2. Negative Perception (NP): Patients’ maladaptive interpretations of symptoms.
  3. Illness Behavior (IB): Actions taken in response to symptoms, such as frequent medical consultations.
  4. Social Function (SF): Impact of symptoms on daily life and interpersonal relationships.

Initial validation in psychiatric settings demonstrated promising reliability and validity. However, since most patients with somatoform disorders in China first present to general hospitals, further validation in this context was necessary.

Study Design and Methodology

The research was conducted in two stages:

Stage 1: Psychometric Evaluation

A multicenter cross-sectional study enrolled 1,558 outpatients from gastroenterology, neurology, and cardiology departments in three Beijing general hospitals. Inclusion criteria required participants to be aged 18–65, possess at least junior high school education, and report somatic symptoms as their primary concern. Exclusion criteria included cognitive impairment, severe mental disorders, or physical conditions preventing questionnaire completion.

Participants completed the SQSS and the Patient Health Questionnaire-15 (PHQ-15), a validated measure of somatic symptom severity. Confirmatory factor analysis (CFA) was performed using AMOS 25.0 to evaluate the SQSS’s structural validity. Reliability was assessed via Cronbach’s alpha and Spearman-Brown split-half coefficients. Criterion validity was examined through correlations with PHQ-15 scores. Measurement invariance was tested across age, gender, and diagnostic groups.

Stage 2: Diagnostic Accuracy and Group Comparisons

A second sample of 279 participants was recruited, including patients with somatoform disorders (n=72), anxiety/depression (n=63), physical diseases without mental health issues (n=72), and healthy controls (n=72). Receiver operating characteristic (ROC) curve analysis determined the optimal SQSS cutoff score for identifying somatoform disorders. Between-group differences in SQSS scores were analyzed using ANOVA.

Key Findings

Structural Validity and Model Fit

CFA confirmed the four-factor structure of the SQSS, with robust model fit indices:

  • Chi-square/degrees of freedom = 5.508
  • Goodness-of-Fit Index (GFI) = 0.936
  • Comparative Fit Index (CFI) = 0.924
  • Root Mean Square Error of Approximation (RMSEA) = 0.054
  • Standardized Root Mean Square Residual (SRMR) = 0.039

All factor loadings exceeded 0.4, supporting the scale’s structural coherence. The four dimensions correlated moderately with one another (r = 0.378–0.612) and strongly with the total score (r = 0.726–0.858).

Reliability and Validity

  • Internal Consistency: Cronbach’s alpha for the total scale was 0.899, with subscale alphas ranging from 0.724 (SF) to 0.877 (SS).
  • Split-Half Reliability: Spearman-Brown coefficient = 0.865.
  • Criterion Validity: SQSS total scores correlated strongly with PHQ-15 scores (r = 0.683, p < 0.01).

Measurement Invariance

The SQSS demonstrated configural, metric, and scalar invariance across age, gender, and diagnostic groups, confirming its suitability for diverse populations.

Diagnostic Utility

ROC analysis yielded an area under the curve (AUC) of 0.818 (p < 0.001) for distinguishing somatoform disorders from other groups. A cutoff score of 29 maximized diagnostic accuracy:

  • Sensitivity = 78.1%
  • Specificity = 71.4%
  • Youden Index = 0.494

Between-Group Differences

ANOVA revealed significant differences in SQSS scores across diagnostic groups (p < 0.001):

  1. Somatoform Disorders: Highest scores across all subscales (SS = 19.6, NP = 12.3, IB = 9.8, SF = 8.4, Total = 50.1).
  2. Anxiety/Depression: Elevated scores compared to physical disease and healthy groups, suggesting somatic symptom comorbidity.
  3. Physical Disease and Healthy Controls: Lowest scores, confirming the SQSS’s ability to discriminate between somatic and non-somatic conditions.

Clinical Implications

The SQSS addresses critical gaps in somatic symptom assessment by integrating culturally relevant dimensions of illness perception and behavior. Its strong psychometric properties support its use as a screening tool in general hospitals, where early identification of somatoform disorders can reduce diagnostic delays and improve patient outcomes. The optimal cutoff score of 29 provides a practical threshold for clinicians to initiate further evaluation or referral.

Limitations and Future Directions

While the study demonstrated robust validity in Chinese populations, cross-cultural validation is needed to assess the SQSS’s applicability in other settings. Longitudinal studies could further establish predictive validity and responsiveness to treatment.

doi:10.1097/CM9.0000000000001615

Was this helpful?

0 / 0