Participants
The study had a cross-sectional study design. Participants were recruited from among the patients who visited the Department of Psychosomatic Medicine at Toho University Medical Center Omori Hospital between February and March 2021. The inclusion criteria were as follows: 1) age 20–79 years; 2) accurate understanding of the purpose and process of the study and signing an informed consent form; 3) meeting the diagnostic criteria for somatic symptoms and related disorders [1]. Exclusion criteria included diagnosis of 1) schizophrenia spectrum disorder and other psychotic disorders; 2) dementia (such as Alzheimer's dementia, vascular dementia, Parkinson's disease dementia, and Lewy body dementia); 3) neurodevelopmental disorders (such as autism spectrum disorder, attention deficit/hyperactivity disorder, communication disabilities); 4) dissociative disorders; and 5) patients who could not be accurately assessed for any reason.
The diagnoses were made by multiple physicians using the Diagnostic and Statistical Manual of Mental Disorders, 5th ed. (DSM-5) [1]. Data regarding age, sex, education, and duration of treatment were collected as background factors from all participants.
Questionnaires
The SSS-8 [5] was used to assess physical symptoms; the Japanese version of the SSS-8[12] has been validated linguistically and psychologically and has internal consistency [13].
The Central Sensitization Inventory (CSI) [14] was used to assess central sensitization. The CSI consists of two parts. Part A assesses subjective symptoms common to CSS and Part B asks whether the subject has had CSS in the past. CSI is a questionnaire with high reliability and internal consistency, and the reliability and validity of the Japanese version of CSI have already been verified in a previous study [15]. The CSI correlates with quantitative sensory tests used for inferring CSS [16,17], and a cut-off point of 40 or higher on the CSI-A has been reported to be useful for discriminating CSS in outpatient clinics [18]. In this study, patients with a CSI-A of 40 points or higher were included in a group with severe comorbid CSS.
We assessed the participants' state of anxiety, depression, and catastrophic thinking, which are psychological states that have been reported to be related to central sensitization in previous studies [19–21].
The Hospital Anxiety and Depression Scale (HADS) [22] is a questionnaire consisting of seven items each on anxiety and depression. Both the anxiety and depression scales are scored from 0 to 21 points and are used as clinical indicators of psychiatric symptoms in general practice [23]. The HADS has also been reported to be associated with quality of life [24], and the Japanese version of the HADS has been validated for reliability and validity [25]. The Pain Catastrophizing Scale (PCS) [26] is a 13-item questionnaire with three subscales (rumination, helplessness, and magnification) that assesses catastrophic thinking and has shown high reliability and validity. The reliability and validity of the Japanese version of the PCS were also confirmed [27].
Data analysis
For differences in background factors and endpoints between the CSS and non-CSS groups, nominal variables were subjected to chi-square or Fisher’s test, continuous variables to t-test, and categorical variables and non-normally distributed continuous variables to Mann–Whitney U test.
To evaluate the utility of the SSS-8 in discriminating between the CSS and non-CSS groups, two propensity scores were calculated by logistic regression analysis. One was the propensity score with CSS as the dependent variable, SSS-8 as the independent variable, and PCS, HADS, age, sex, and CSI-B as confounders of CSS, and the other was the propensity score with CSS as the dependent variable and PCS, HADS, age, sex, and CSI-B as independent variables. Receiver operating characteristic (ROC) curve analyses were performed on the propensity scores [28] to statistically compare the area under the curve (AUC) with and without SSS-8 as an independent variable. The optimal cut-off point of SSS-8 was determined by the Youden Index to distinguish the group with severe CSS, and the accuracy of the test was evaluated by its sensitivity and specificity.
All analyses in this study were performed using EZR Version 1.32 [29]. Two-tailed P-values less than 0.05 were considered statistically significant.