Validity and Reliability of the Japanese Version of the Cognitive Scale for Functional Bowel Disorders

The Cognitive Scale for Functional Bowel Disorders (CS-FBD) is a useful measure to assess maladaptive cognition, and focuses on how functional bowel disorders relate to negative thoughts, perfectionism, and social This study aimed to conrm the reliability and validity of the Japanese version of the CS-FBD (CS-FBD-J).

group, non-patient IBS group (students with IBS except two who received treatment), and control group, there were signi cant group effects in the CS-FBD-J (IBS patient > non-patient IBS > control). The internal consistencies of the CS-FBD-J were high (α = 0.95). The item-total correlation analysis for the CS-FBD-J showed that the correlations between each item and the total score were signi cant.

Conclusion
This study con rmed the reliability and validity of the 23-item version of the CS-FBS with the deletion of two items with low factor loadings.

Background
Irritable bowel syndrome (IBS) is a functional gastrointestinal disorder characterized by persistent chronic abdominal pain and disturbance of bowel movements. It is a typical psychosomatic disorder of the digestive system that is exacerbated by psychological stress. Many previous studies have reported the effect of cognitive behavioral therapy CBT) or cognitive therapy on abdominal symptoms and emotional distress in patients with IBS [1,2]. Our previous study [3] suggested that the cognitive-emotion process is related to the aggravation of anxiety in individuals with IBS; severe anxiety sensitivity in individuals with IBS related to their symptom-related cognition and the altered cognition associated with increased anxiety can lead to the development of a disabling condition including avoidant behavior.
The Cognitive Scale for Functional Bowel Disorders (CS-FBD [4]) is a useful measure to assess maladaptive cognitions, focusing on functional bowel disorders related to negative thoughts, perfectionism, and social desirability. The CS-FBD is a valid and reliable scale that can be used as an outcome measure in evaluating the e cacy of different forms of psychotherapeutic intervention for functional bowel disorders, including IBS, and can also serve as a helpful assessment tool for health professionals working with patients diagnosed with IBS [4]. CS-FBD has been used in many previous studies, including clinical trials of CBT [5][6][7][8][9][10]. The Visceral Sensitivity Index (VSI [11]) has been developed to assess gastrointestinal speci c anxiety, cognitive, affective, and behavioral responses to fear of GI sensations, symptoms, and the context in which they occur, and it has been translated into Japanese [12]. The CS-FBD focuses on maladaptive cognition that is not speci c to anxiety, and its content is differentiated from the VSI. CS-FBD items were generated from the diaries of automatic thought records in patients with functional gastrointestinal disorders, and included a variety of themes: bowel performance anxiety, out of control, perfectionism, anger/frustration, self-e cacy, social approval, embarrassment/shame, heightened sensitivity to social rules and norms, and self-nurturance. The CS-FBD was utilized along with the VSI in previous studies (e.g., [9]). However, a Japanese version of the CS-FBD has not yet been developed.
Therefore, the purpose of this study is to con rm the reliability and validity of the Japanese version of the CS-FBD (CS-FBD-J).

Participants
Participants were recruited between September 2013 and March 2021. Participants of the present study were outpatients who were diagnosed with IBS in the gastroenterology and psychosomatic medicine departments of hospitals in Tokyo and Kanagawa and college students, including IBS patients or nonpatient IBS who met diagnostic criteria (Rome III criteria) and those who did not meet the diagnostic criteria. The present study was approved by the ethical committee of Yokohama City University (B130905050) and the National Center of Neurology and Psychiatry (A2018-046). Written informed consent was obtained from all participants.

Measures
Cognitive Scale for Functional Bowel Disorders translated in Japanese [4] The Cognitive Scale for Functional Bowel Disorders (CS-FBD) includes 25 items that measure maladaptive cognition related to abdominal symptoms, with responses ranging from 1 (strongly disagree) to 7 (strongly agree). The total score for this scale ranges from 25 to 175. One of the authors (N.S.), who is a psychologist knowledgeable about functional gastrointestinal disorders and psychosomatic medicine, translated the original version into Japanese. Another author (F.G.) who is bilingual in English and Japanese, back-translated the Japanese items of the CS-FBD into English. We compared the results of the back translation to the original version and con rmed agreement among the researchers about the nuances.
Japanese version of the Rome III modular questionnaire (only items related to IBS) [13,14] The Rome III Diagnostic Criteria for Functional Gastrointestinal Disorders [15] have been widely used to de ne IBS. Rome IV has not been published at the start of this investigation. Using the Rome III modular questionnaire, the presence of IBS was determined if the participants had abdominal pain or discomfort for at least 3 weeks (at least once per week) in the previous 3 months and two of the three following symptoms: (1) pain or discomfort that improves or stops after a bowel movement, (2) a change in the number of bowel movements when the pain or discomfort starts, and (3) either softer or harder stools than usual when the pain or discomfort starts.

IBS Severity Index [16,17]
The IBS Severity Index (IBS-SI) was used to assess the severity of gastrointestinal symptoms. This scale consists of 5 items: abdominal pain (2 items), abdominal distension, bowel movements, and quality of life, and the total score ranges from 0 to 500. The IBS-SI classi es the severity of IBS as mild (75-174), moderate (175-299), or severe (300-500) based on clinical observations of patients with IBS.
Japanese version of Visceral Sensitivity Index [11,12] The Visceral Sensitivity Index (VSI) is a self-reported measure of gastrointestinal symptom-speci c anxiety in patients with IBS. The VSI includes 15 items with responses ranging from 1 (strongly agree) to 6 (strongly disagree), and scores ranging from 0 to 5.

Japanese version of the 24-item Dysfunctional Attitudes Scale (Tajima et al., 2007 [18])
The 24-item Dysfunctional Attitudes Scale (DAS-24) is a self-report scale for depression schemata with responses ranging from 1 (strongly disagree) to 7 (strongly agree). The DAS-24 contains three factors: "achievement," "self-control," and "dependency." The achievement subscale consists of eight items, for example, "If I fail partly, it is as bad as being a complete failure." The dependency subscale consists of eight items, for example, "I am nothing if a person I love doesn't love me." The self-control subscale consists of eight items, for example, "A person should do well at everything he undertakes." Hospital Anxiety and Depression Scale [19,20] The Hospital Anxiety and Depression Scale (HADS) is a self-report questionnaire for anxiety and depression. The HADS includes of 14 questions on a 4-point scale, consisting of an anxiety subscale with 7 items and a depression subscale with 7 items. This psychometric instrument was chosen because all its items refer solely to an emotional state and do not consider somatic symptoms.

Patients
Patients with IBS were given written and oral explanations about the purpose of this study, the protection of their personal information, and that participation was voluntary. Patients with IBS who agreed to participate in this study provided informed consent. Participants then completed questionnaires containing the physical and psychological scales mentioned above.

College students
The researchers recruited college students after class hours or via the university's online bulletin board. Participants recruited by the latter method were limited to those who met the diagnostic criteria for IBS. They were given written and oral explanations about the purpose of this study, the protection of their personal information, that participation was voluntary, and that they would not be punished for not cooperating. After the above explanations were given, only those students who agreed to participate were asked to complete a series of questionnaires.

Statistical analysis
The factor structure of the CS-FBD-J was con rmed by exploratory factor analysis. Construct validity was con rmed using Pearson's correlation coe cient analysis between the CS-FBD-J and scores on the IBS-SI, VSI, DAS, and HADS. Criterion-related validity was con rmed using one-way analysis of variance to compare the CS-FBD-J score between IBS patient group, non-patient IBS group and control group, and a post-hoc test with Scheffe's method was employed to test the differences between groups. The internal consistency reliability of the CS-FBD-J was determined using Cronbach's α. The correlations between each item and the total score of the CS-FBD-J were used to conduct item-total correlation analysis. The signi cance level was set at less than 5%. Data analyses were performed using SPSS software (version 25.0; IBM Corp., NY, USA).

Results
Descriptive results Table 1 presents the descriptive results for the participants. In our study, 203 college students were enrolled, and 192 students (20.2±3.0 yrs.) provided valid data. There were 76 students who met the diagnostic criteria for IBS, and 2 students received treatment for IBS. Twenty-two outpatients who were diagnosed with IBS in the gastroenterology and psychosomatic medicine departments of hospitals in Tokyo and Kanagawa (38.0±13.0 yrs.) provided valid data. Eleven participants who received treatment for IBS received psychiatric medication. The IBS group (IBS patients and non-IBS patients) included 28 participants with diarrhea-predominant IBS (IBS-D), 9 participants with constipation-predominant IBS (IBS-C), 55 participants with mixed type IBS (IBS-M), and 6 participants with unclassi able IBS (IBS-U). Table 2 shows the differences in each variable between the IBS patient group, non-patient IBS group, and control group. There were signi cant group effects in the IBS-SI and VSI, and all comparisons between all groups in these variables except that between the IBS patient group and the non-IBS group in IBS-SI showed signi cant differences under multiple comparison analysis.  Factor analysis Exploratory factor analysis (using maximum likelihood estimation and promax rotation) without assigning the number of factors revealed the presence of ve components with eigenvalues exceeding 1. However, inspection of the scree plot revealed a break after the rst component. Therefore, we decided to retain a unidimensional factor structure for further investigation (Table 3). Factor analysis showed that the factor loadings for two items (items 6 and 18) were less than 0.4. After excluding these items from the analysis, the factor loadings for all items were above 0.4, and a single factor of the 23-item version accounted for 45.85% of the total variance.

Construct validity
The results of the correlation analysis between the CS-FBD-J and other variables in the IBS and control groups con rmed the construct validity ( Table 4)

Criterion-related validity
We compared the scores of the 25-item and 23-item versions of the CS-FBD-J between the IBS patient group, non-patient IBS group, and control group to con rm the criterion-related validity ( Table 2). There were signi cant group effects in both versions of the CS-FBD-J, and multiple comparison analysis showed signi cant differences between all groups (i.e., IBS patient > non-patient IBS > control).

Discussion
While the CS-FBD-J had a one-factor structure in common with the original version, two items (items 6 and 18) the Japanese version had low factor loadings and were deleted from this version. These items of the original version were also reported to not have high factor loadings. Because the automatic thoughts referenced in these two items could not be interpreted by some people as being related to abdominal symptoms, the factor loadings of the two items were lower in this study.
We con rm that the construct validity of the CS-FBD-J is su cient. Regardless of the presence or absence of IBS in the participants, the CS-FBD-J and IBS symptom-related scales (IBS-SI and VSI) were positively correlated. Thus, the CS-FBD-J was appropriate as a scale re ecting gastrointestinal symptomspeci c cognition. In contrast, general dysfunctional cognition not speci c to gastrointestinal symptoms (DAS-24) was not prominently correlated with CS-FBD-J. Although this result differs from the original version, it can be said that it clearly shows the difference between general dysfunctional cognition and IBS-speci c cognition. In addition, anxiety and depression (HADS) were correlated with CS-FBD-J only in the IBS group. Since cognition of abdominal symptoms has been reported to be associated with anxiety and depression in IBS patients [3,12], the CS-FBD-J may re ect cognition speci c to gastrointestinal symptoms that affect their emotions in individuals with IBS. In both the previous study [12] and the present data, the VSI was correlated with anxiety but not with depression, whereas the CS-FBD-J was correlated not only with anxiety but also with depression. Therefore, the CS-FBD can be distinguished from the concept measured by the VSI in that it re ects cognitions that affect both anxiety and depression. The CS-FBD and the VSI can be used for different purposes or can be combined to elaborate on cognitions related to gastrointestinal symptoms.
Su cient criterion-related validity of the CS-FBD-J was also con rmed. The CS-FBD-J scores were higher in IBS patients, non-IBS patients, and the control group, in that order. In this study, the intensity of IBS symptoms (IBS-SI) by group was also in this order. This suggests that there is a large difference in CS-FBD-J scores not only in the presence or absence of IBS, but also in the high-and low-severity groups.
The high reliability of the CS-FBD-J was con rmed. The alpha coe cient was greater than 0.9, indicating a high internal consistency. The item-total correlation was signi cant for all items, and exceeded 0.4 for the 23-item version.
The present study has several limitations. First, there were few patients with IBS among the participants in this study. In a previous study of the original version, all participants were patients with functional bowel disorder. As shown by the data of this study, the severity of IBS and non-IBS patients may differ. If we collect data from a su cient number of patients with IBS or FBD and conduct factor analysis, the results may be different from those of this study. Second, the diagnostic criteria for IBS used in this study were not the latest Rome-IV criteria, which had not been published at the time the study was started. In the latest diagnostic criteria, the presence of abdominal pain and discomfort is essential for the diagnosis of IBS. Based on the above issues, further validation of CS-FBD-J is required by extracting individuals with IBS using the latest diagnostic criteria and collecting su cient data from FBD patients.

Conclusion
The The present study was approved by the ethical committee of Yokohama City University (B130905050) and the National Center of Neurology and Psychiatry (A2018-046). Written informed consent was obtained from all participants.