Introduction

Irritable bowel syndrome (IBS) represents disorders of gut-brain interaction and is a common disorder with a worldwide prevalence of 4.1% according to the latest Rome IV criteria1. IBS presents with recurrent abdominal pain that meets two of the following three criteria: changes in defecation, changes in frequency of defecation, and changes in stool shape (appearance)1. IBS symptoms involve multiple factors, including socio-psychological factors, genetic predisposition, gut microbiome, and diet, which complicates treatment2,3. Given that IBS lowers patients’ quality of life4 and imposes a heavy social and economic burden5, it is important to address this issue.

It has been reported that comorbid psychiatric symptoms such as depression and anxiety disorders are common in patients with IBS6. Even in the absence of diagnostic-level psychiatric symptoms, there is a tendency for patients with IBS to perceive normal bodily functions negatively because of their health anxiety and symptom-specific anxiety, and they may have a hypersensitivity to symptoms as well as attention bias7,8. A study reporting that subjective pain scores were higher in patients with severe IBS despite there being no difference in rectal pain sensory thresholds between patients with severe and mild IBS9 suggests the importance of cognition in the severity of IBS, and neuroimaging studies support this concept10. Psychological characteristics reported in IBS patients include low self-esteem11, high levels of self-blame12, “all-or-nothing” behavior13, shame and defective schemas about oneself12, and catastrophizing14. Catastrophizing mediated depression and IBS symptoms15, while dysfunctional beliefs and avoidance-oriented coping strategies mediated IBS symptoms and reduced health-related quality of life16. These attentional biases, irrational belief systems, and associated maladaptive coping strategies are thought to increase perceived stress and contribute to symptom exacerbation; cognitive-behavioral therapy, an accepted effective treatment for IBS17, targets these factors. Addressing psychological symptoms is key to the treatment of IBS, as a meta-analysis found that psychotherapy overall has a number needed to treat of four patients (95% confidence interval [CI]: 3–5) and is more effective compared with most medications17.

The etiology of the psychological symptoms associated with IBS remains uncertain. It is well known that IBS patients have more childhood and adult traumatic experiences compared with healthy controls (HCs)18,19. Traumatic experiences in IBS patients are associated with IBS symptoms. Adverse childhood experiences (ACEs), including abuse and neglect, increase the prevalence of IBS18, and the number of ACEs exposed correlates with IBS severity20. In a meta-analysis, the odds ratio (OR) for IBS due to traumatic experiences in both childhood and adulthood was 2.22 (95% CI 1.72–2.86)21, and that for IBS due to PTSD in adulthood was 2.80 (95% CI 2.06–3.54)22.

Complex post-traumatic stress disorder (CPTSD) is a condition that occurs following repeated and severe traumatic experiences, unlike PTSD, which often results from a single traumatic event such as an accident or combat experience. CPTSD was listed for the first time as an official diagnosis in the 11th revision of the International Classification of Diseases (ICD-11) published by the WHO in 2018. The diagnosis of PTSD includes three core symptom clusters that occur after the traumatic experience: re-experiencing in the here and now (RE), avoidance of traumatic reminders (AV), and sense of current threat (TH). In addition to the core symptoms of PTSD, CPTSD includes three additional symptom clusters: affective dysregulation (AD), negative self-concept (NSC), and disturbances in relationships (DR), collectively referred to as "disturbance of self-organization" (DSO)23. CPTSD is diagnosed when symptoms of both PTSD and DSO are satisfied after a period of time following an event or series of events of an extremely threatening or horrific nature, including torture, slavery, genocidal campaigns, prolonged domestic violence, repeated child sexual or physical abuse. CPTSD is more likely to result from childhood trauma compared with PTSD24, with more childhood and lifetime trauma experiences and more severe social dysfunction25.

DSO is the symptom that distinguishes CPTSD from PTSD and it reflects the pervasive impact of complex trauma on an individual’s psychological functioning and overall well-being beyond PTSD. DSO is based on disorders of extreme stress not otherwise specified (DESNOS), which was proposed in the 1990s26. DESNOS are persistent effects on a person’s cognition, emotions, and sense of self caused by long-term or repeated traumatic experiences, including female sexual abuse, child abuse, captivity, or refugee experiences, and include symptoms such as problems with the regulation of affect and impulses, memory and attention, self-perception, interpersonal relations, somatization, and systems of meaning26. DESNOS has been shown to have a profound impact that can change people’s lives27.

PTSD is associated with physical illnesses (e.g., cardiovascular, metabolic, and musculoskeletal disorders)28. In addition, a recent study revealed associations between CPTSD and somatization29. Not only symptoms of CPTSD but also those of DSO alone were found to mediate between ACEs and somatization symptoms30. Considering the close relationship between ACEs and IBS, as Drossman notes31, the psychological and physical dysfunction based on the long-lasting trauma experience may impact the development of IBS. Although the relationship between IBS and PTSD has been well established in previous studies22, the relationship between IBS and CPTSD and/or DSO has not yet been investigated, despite the many similarities between IBS and CPTSD. Both have high ACE exposure, high use of maladaptive coping strategies (in those experiencing multiple ACEs, not CPTSD)32, hyperarousal33, and the NSC observed in DSO is very similar to the psychological characteristics of IBS. Similar to somatization, it is possible that CPTSD or DSO may mediate the pathway between ACEs and IBS.

Based on the above, we propose the following hypotheses in this study. #1. The prevalence of PTSD symptoms, CPTSD symptoms, and DSO-alone symptoms is higher among those with IBS symptoms compared with HCs. #2. PTSD, CPTSD, and DSO-alone symptoms are risk factors for IBS. #3. Groups with PTSD, CPTSD, and DSO have a higher severity of IBS compared with those without PTSD, CPTSD, and DSO. #4. PTSD and DSO symptoms mediate the relationship between ACEs and IBS severity. #5. A wide range of ACEs, represented by bullying, are associated with IBS symptoms.

To test the above hypotheses, we conducted an online survey of individuals with IBS and HCs from the general population in Japan (See Supplemental Fig. 1 for a flowchart of the survey). From the general population of 47,000, 721 individuals who met the Rome IV criteria for IBS and did not meet the exclusion criteria were selected as the IBS group. Then, 721 people in the HC group were randomly selected with demographic information matched to the IBS group. Both groups were then surveyed using three main survey items: the IBS severity index (IBS-SI), Japanese version; a 10-question ACE questionnaire from the US Centers for Disease Control and Prevention (CDC) plus 5 author-created questions; and the International Trauma Questionnaire (ITQ), which can screen for CPTSD according to the ICD-1134. We did not assess the details of the traumatic event in the ITQ, only the traumatic symptoms. Responses were obtained from 669 participants in the IBS group and 659 in the HC group.

Because this study also focused on whether ACEs other than the CDC’s 10 ACEs are involved in IBS, we created our own ACE questionnaire. The CDC’s 10-question ACE questionnaire examines family adversity, including psychological/physical/sexual abuse, physical/emotional neglect, parental separation or divorce, witnessing intimate partner violence (IPV), living with a mentally ill or addicted household member, and incarceration of a household member. According to previous studies35,36, the suicide rate in Japan is higher than that in Western countries, despite lower scores on the CDC’s ACE questionnaire (https://data.oecd.org/healthstat/suicide-rates.htm). Childhood bullying victimization has been shown to be associated with attempted or completed suicide among youth37. Therefore, it was considered that ACEs other than the 10 adversities, such as bullying, may be closely related to IBS and PTSD or CPTSD. We focused specifically on the effects of bullying but we also added questions about the death or disability of one or more parents and/or siblings38,39, interpersonal stress with people other than parents or similar-aged peers, accident or natural disaster, and low socioeconomic status (SES) during childhood40,41 to the CDC questionnaire, for a total of 15 questions. In this study, this 15-question questionnaire is hereafter referred to as the Comprehensive ACE (C-ACE) questionnaire and its score as the C-ACE score in order to distinguish it from the CDC’s 10-question ACE questionnaire and score.

The Rome Foundation Global Epidemiology Study (RFGES)42,43,44, which was conducted across 33 countries, concluded that participants underreported symptoms in face-to-face surveys compared with online surveys owing to the embarrassment of reporting gastrointestinal symptoms in person. Accordingly, we considered an online survey to be an appropriate research method for this study, which deals with sensitive issues such as traumatic experiences as well as IBS symptoms.

Results

Demographic information

The respondents’ demographic information is summarized in Table 1, with complete details being presented in Supplementary Table 1. There were no significant differences between the HC and IBS groups in terms of age, gender ratio, body mass index, family income, education, region of residence, and type of business. The IBS group was significantly more single and had fewer children compared with the HC group. The percentage of drinkers, smokers, and those who reported having acute gastroenteritis immediately before the onset of IBS symptoms and IBS-SI score was significantly higher in the IBS group compared with the HC group. There were no major differences between the two groups in terms of diet.

Table 1 Summary of participant characteristics.
Full size table

Comparison of comprehensive ACE scores between the HC and IBS groups

The IBS group had a significantly higher percentage of ACE exposures compared with the HC group. The percentage of participants with a C-ACE score of 0 (no ACEs) was 60% in the HC group and 29% in the IBS group (p < 0.001). The proportions of C-ACE and ACE scores for each group are shown in Supplementary Fig. 2. There was a significant difference in mean C-ACE scores between groups: 0.87 (Standard deviation (SD) 1.58) in the HC group and 2.41 (2.73) in the IBS group (p < 0.001).

Comparison of the prevalence of each ACE

The prevalence of each ACE was compared as follows: HC males vs. IBS males, HC males vs. HC females, HC females vs. IBS females, and IBS males vs. IBS females (Fig. 1). The precise figures are given in Supplementary Table 2. The prevalence was significantly higher in the IBS group compared with the HC group for all ACEs. For men and women combined, the most prevalent ACE in the IBS group was bullying (30.1% of men and 38.7% of women), followed by psychological abuse (19.9% of men and 30.0% of women) and emotional neglect (16.7% of men and 26.0% of women). The HC group had the highest prevalence of death or disability of one or more parents and/or siblings (12.9% for men and 15.3% for women), followed by parental separation or divorce (8.6% for men and 12.4% for women) and bullying (10.0% for men and 10.6% for women).

Figure 1
Figure 1
Full size image

Comparison of the prevalence of each ACE. The following comparisons were made: HC males vs. IBS males, HC males vs. HC females, HC females vs. IBS females, and IBS males vs. IBS females. ACEs, adverse childhood experiences; HC, healthy controls; IBS, irritable bowel syndrome; IPV, intimate partner violence; SES, socioeconomic status. Pearson’s Chi-square test. a, p < .001 versus HC/male; b, p < .01 versus HC/male; c, p < .05 versus HC/male; d, p < .001 versus HC/female; e, p < .05 versus HC/female; f, p < .01 versus IBS/male; g, p < .05 versus IBS/male; h, p < .01 versus IBS/female; i, p < .05 versus IBS/female. Light green columns, HC males; green columns, HC females; orange columns, IBS males; red columns, IBS females.

Comparison of prevalence of PTSD, CPTSD, and DSO between the HC and IBS groups

ITQ was utilized to determine the prevalence of PTSD symptoms, CPTSD symptoms, and DSO-only symptoms (PTSD group, CPTSD group, and DSO group), and these were compared between the IBS and HC groups (Fig. 2). Of the respondents in the HC and IBS groups, respectively 13 (2.0%) and 31 (4.6%) had PTSD symptoms, 39 (5.9%) and 116 (17.3%) had CPTSD symptoms, and 14 (2.1%) and 86 (12.9%) had DSO alone symptoms. The prevalence of PTSD, CPTSD, and DSO was significantly higher in the IBS group than in the HC group.

Figure 2
Figure 2
Full size image

Comparison of prevalence of PTSD, CPTSD, and DSO between the HC and IBS groups. Numbers represent percentages. CPTSD, complex post-traumatic stress disorder; DSO, disturbance of self-organization; HC, healthy controls; IBS, irritable bowel syndrome; PTSD, post-traumatic stress disorder. Pearson’s Chi-square test, ***p < .001, **p < .01. Green, participants without PTSD/CPTSD/DSO symptoms; gray, with PTSD symptoms; orange, with CPTSD (PTSD + DSO) symptoms; red, with DSO symptoms.

Comparison of trauma scores between the HC and IBS groups

The trauma scores for the HC and IBS groups, as measured by the ITQ, are shown in Fig. 3. The PTSD score (a) is the sum of the RE, AV, and TH scores, while the DSO score (b) is the sum of the AD, NSC, and DR scores. The respective means and SDs for each score in the HC group vs. IBS group were as follows: RE score, 0.71 (1.54) vs. 2.06 (2.17); AV score, 0.79 (1.74 ) vs. 2.41 (2.36); TH score, 0.75 (1.67 ) vs. 2.45 (2.41); PTSD score, 2.25 (4.67) vs. 6.92 (6.26); AD score, 0.69 (1.59) vs. 2.38 (2.36); NSC score, 0.74 (1.71) vs. 2.95 (2.74); DR score, 0.83 (1.73) vs. 2.86 (2.57); DSO score, 2.25 (4.75) vs. 8.19 (7.03). All scores were significantly higher in the IBS group than in the HC group (all p < 0.001).

Figure 3
Figure 3
Full size image

Comparison of trauma scores by international trauma questionnaire between the HC and IBS groups. (a) The PTSD score is the sum of the RE, AV, and TH scores. (b) The DSO score is the sum of the AD, NSC, and DR scores. Numbers represent percentages. AD, affective dysregulation; AV, avoidance of traumatic reminders; DR, disturbances in relationships; DSO, disturbance of self-organization; HC, healthy controls; IBS, irritable bowel syndrome; NSC, negative self-concept; PTSD, post-traumatic stress disorder; RE, re-experiencing in the here and now; TH, sense of current threat. Mann–Whitney U test, ***p < .001. Green columns, HC group; red columns, IBS group.

IBS-SI score by C-ACE score group and PTSD/CPTSD/DSO group

All participants (n = 1,328) were classified into four groups: the 0 ACEs group (n = 584) for C-ACE scores of 0, the low ACEs group (n = 527) for C-ACE scores of 1–3, the medium ACEs group (n = 147) for C-ACE scores 4–6, and the high ACEs group (n = 70) for C-ACE scores of ≥ 7, and the IBS-severity index Japanese version (IBS-SI-J) score for each group was examined. IBS-SI score was 86.32 (SD 98.18) in the 0 ACEs group, 130.83 (106.72) in the low ACEs group, 171.24 (116.30) in the medium ACEs group, and 233.74 (128.73) in the high ACEs group, showing a significant increase in dose-dependence with increasing score (Fig. 4a). We then compared the IBS-SI scores of all participants by PTSD category group. The no PTSD/CPTSD/DSO symptom group, PTSD group, CPTSD group, and DSO group had IBS-SI scores of 104.35 (SD 103.20), 165.27 (104.64), 169.14 (130.92), and 200.32 (115. 64), respectively. The PTSD, CPTSD, and DSO groups had significantly higher IBS-SI scores compared with the no PTSD/CPTSD/DSO symptom group, with a trend toward higher IBS-SI scores in the order of PTSD, CPTSD, and DSO groups (Fig. 4b).

Figure 4
Figure 4
Full size image

IBS-SI scores of all participants by C-ACEs score group and PTSD/CPTSD/DSO group (both n = 1,328). (a) IBS-SI scores by C-ACEs score group. 0 ACEs, C-ACEs score 0; Low ACEs, C-ACEs score 1–3; Mid ACEs, C-ACEs score 4–6; High ACEs, C-ACEs score ≥ 7. (b) IBS-SI scores by PTSD/CPTSD/DSO group. C-ACEs score, comprehensive adverse childhood experiences score; CPTSD, complex post-traumatic stress disorder; DSO, disturbance of self-organization; IBS, irritable bowel syndrome; IBS-SI, IBS severity index; PTSD, post-traumatic stress disorder. Kruskal–Wallis’s test after Bonferroni adjustment. a, p < .001 vs. 0 ACEs; b, p < .001 vs. Low ACEs; c, p < .01 vs. Low ACEs; d, p < .05 vs. Mid ACEs; e, p < .001 vs. No PTSD/CPTSD/DSO group; f, p < .01 vs. No PTSD/CPTSD/DSO group.

Logistic regression analysis

A logistic regression analysis model was created with IBS as the objective variable for all participants (Table 2). The variable with the highest OR for IBS was DSO-alone symptoms (p < 0.001, OR = 3.718). The variables with ORs of > 2 were, in descending order, living with an addicted family member (p = 0.002, OR = 2.842), interpersonal stress (p = 0.034, OR = 2.323), living with a family member with mental illness (p = 0.006, OR = 2.226), and probiotics intake (p < 0.001, OR = 2.144). Variables with ORs between 1.7 and 1.9 were, in descending order, CPTSD symptoms, C-ACE score of ≥ 1, bullying victimization, and psychological abuse. The OR for having PTSD symptoms was 2.047 but did not reach statistical significance, albeit by only a small margin. Variables that showed negative effects on IBS were physical neglect (p = 0.039, OR = 0.464) and witnessing IPV (p = 0.007, OR = 0.401). The variance inflation factor (VIF), which indicates the multicollinearity of each factor, was low at a maximum of 2.067 (See Supplementary Table 3).

Table 2 Logistic regression analysis for irritable bowel syndrome.
Full size table

Mediation analysis

Based on the above results, we predicted that the relationship between C-ACE scores and IBS-SI scores would be mediated by the PTSD and DSO scores. Therefore, we created Model A (Fig. 5a), in which the total effect of the C-ACE score on the IBS-SI score was 0.33, the path coefficient to the PTSD score was 0.48, and the path coefficient to the DSO score was 0.53, all of which were significant at p < 0.001. The path from PTSD score to IBS-SI score was insignificant at p = 0.208, negating any indirect effect of PTSD symptoms. The path coefficient from the DSO score to IBS-SI score was 0.28, with a significant p-value, and the path coefficient for indirect effects due to DSO symptoms was significant at 0.15. C-ACEs accounted for 20% of the variance to IBS-SI scores, with DSO symptoms mediating 45%. In addition, among all participants, the rate of those who had experienced bullying was compared between the group without DSO (no PTSD/CPTSD/DSO and the PTSD group) and that with DSO (CPTSD and DSO alone) and was found to be much higher in the group with DSO (58.0% in the group with DSO vs. 14.4% in the group without DSO). Because these results suggested a strong relationship between bullying and DSO, we predicted that DSO may mediate the relationship between bullying and IBS-SI score. To examine this assumption, we created Model B (Fig. 5b), in which the total effect of bullying on IBS-SI score was 0.22, accounting for 18% of the variance to IBS-SI. Nevertheless, the direct effect of DSO on IBS-SI score was non-significant at 0.03 (p = 0.38) when the indirect effect of DSO was considered, which means that DSO completely mediated the effect of bullying to IBS-SI score.

Figure 5
Figure 5
Full size image

Two models (model A (a) and B (b)) of mediation analysis (both n = 1,328). ACEs, adverse childhood experiences; β, standardized coefficients; CI, 95% confidence interval; DSO, disturbance of self-organization; IBS-SI, irritable bowel syndrome severity index; PTSD, post-traumatic stress disorder; R2, square of multiple correlation coefficient. CIs were calculated based on 5,000 bootstrapped samples. ***p < .001.

Discussion

To our knowledge, this study is the first to suggest that IBS is associated with CPTSD according to ICD-11. Regarding hypothesis #1, the prevalence of PTSD, CPTSD, and DSO alone was significantly higher among participants with IBS compared with HCs. The OR for IBS was significantly higher for participants with CPTSD symptoms and those with DSO symptoms alone, but not for those with PTSD symptoms, partially proving hypothesis #2 (However, it is possible that the reason the OR for PTSD symptoms was not significant was that the number of people in the PTSD group was small). IBS-SI tended to be higher in the DSO group, followed by the CPTSD group, PTSD group, and no PTSD/CPTSD/DSO symptoms group, which supported hypothesis #3. Although PTSD symptoms did not mediate IBS symptoms from ACEs as postulated in hypothesis #4, DSO symptoms accounted for approximately half of the mediating effect, which provided support for hypothesis #4. DSO fully mediated the relationship between bullying and IBS-SI. In addition, a wide range of ACEs, including bullying, interpersonal stress, and experiencing an accident or disaster, increased the OR for IBS, confirming hypothesis #5. The higher prevalence of CPTSD (PTSD + DSO) symptoms in individuals with IBS symptoms compared with HCs and the fact that CPTSD symptoms were a risk factor for IBS were new but not surprising findings. However, the fact that DSO-alone symptoms, which are not usually given a diagnosis, had a much higher OR for IBS compared with PTSD and CPTSD symptoms, and that the DSO group had the highest severity of IBS in the group comparison of PTSD categories were novel findings. In the mediation analysis, DSO symptoms accounted for about half of the variance from C-ACEs to IBS-SI, and DSO symptoms completely mediated the variance from bullying to IBS-SI, suggesting the importance of DSO symptoms in the development of IBS.

The high rate of ACE exposure and the high prevalence and scores of trauma symptoms among individuals with IBS symptoms, and the dose–response relationship between the C-ACEs score and IBS severity in the present study were consistent with previous studies18,19,20. More than one-third of the participants with IBS in this study showed symptoms of trauma when DSO-alone symptoms were included. This result is consistent with previous studies, assuming that the presence and severity of trauma are related to the severity of IBS, in which 70% of IBS patients have mild symptoms, while the rest have moderate or higher symptoms9. These findings suggest that a trauma-focused treatment may be beneficial when treating patients with severe IBS who have strong social-psychological symptoms and a traumatic background. Furthermore, psychological assessment to determine psychological symptoms and trauma history and trauma-informed care based on this assessment may be helpful in reducing patients' IBS symptoms and is worth incorporating into future IBS practice.

The present results suggest that at least some of the psychological symptoms of IBS patients may be DSO symptoms resulting from traumatic experiences. It is not clear why DSO-alone symptoms increase the risk of IBS more compared with CPTSD. There were no significant differences in DSO scores between the CPTSD and DSO groups. In the mediation analysis, DSO symptoms mediated between childhood adversity and IBS severity, whereas PTSD symptoms did not. Therefore, the same mediation analysis was conducted with the DSO subscales AD, NSC, and DR as mediators, and the indirect effect was significant only for NSC (See Supplementary Fig. 3). These results suggest that NSC is the most important factor in DSO for IBS and that NSC may play a critically important role in the pathogenesis of IBS. Interestingly, IBS patients with DSO and HCs with DSO tended to have had different types of ACEs (see Supplementary Fig. 4). IBS patients with DSO had higher rates of psychological abuse, physical abuse, emotional neglect, bullying, accidents and disasters, while HCs with DSO had higher rates of parental separation and divorce, incarceration of a household member, and low SES during childhood. Experiences of psychological abuse, physical abuse, emotional neglect, and bullying, as well as parental separation and divorce and low SES during childhood may lower children’s self-esteem and sense of self-worth but may be of different quality. Although parental separation/divorce and low SES during childhood are clearly not the fault of the children themselves, psychological abuse, physical abuse, and bullying may be associated with the formation of a stronger NSC and consequently IBS, given that the child is the target of direct aggression and has their self-esteem undermined (and psychological abuse and bullying actually increased the odds of IBS). Of course, NSC might be caused not only by traumatic experiences but also by IBS itself, or it might also derive from broader life experiences, including ACEs. Recognizing and treating DSO in IBS patients may improve their quality of life, resulting in broader public health benefits, including reduced healthcare costs and lower social and economic burdens. In addition, DSO is likely to be involved in other physical illnesses as well as IBS. For a deeper understanding and treatment of chronic physical illnesses, it seems necessary to position DSO as its own category of pathophysiology and study it accordingly.

Of the CDC’s 10 ACEs in this study, psychological abuse and living with a household member with addiction and mental illness had high ORs. It has been shown that if a mother is addicted to drugs, her attachment to her child is likely to be of the insecure type45. This may contribute to the onset of IBS by shaping the child’s NSC. The same might occur if the caregiver is mentally ill, but further research is needed to confirm this. It is not clear why lower ORs were obtained for physical neglect and witnessing IPV. Physical neglect and witnessing IPV were more prevalent in the IBS group, and both are undoubtedly traumatic experiences for childhood development46,47,48. Possible interpretations of these ORs include the issue of DSO quality discussed above and the fact that this study did not examine protective factors against ACEs. It is known that if someone experiences adversity, the effects of ACEs are mitigated when there is a protective experience (e.g., having at least one secure caretaker, having good friends)49. Previous research indicating that DSOs alone mediated ACEs and somatic symptoms while resilient coping strategies alleviated them30 suggests the importance of protective factors. Future research should focus on protective factors.

ACEs other than the CDC’s 10 ACEs also increased the risk of IBS. Bullying and interpersonal stress increased the risk of IBS, which also tended to increase with accidents and disasters. In a meta-analysis, 57% of bullying victims showed PTSD symptoms above the cutoff value50. Bullying may contribute to the onset of IBS through PTSD and DSO. Future research on the association of ACEs with chronic illnesses, including IBS, should focus on the development of a questionnaire that can measure a broad range of ACEs, including bullying.

Since IBS varies physiologically and symptomatically by subtype, future analysis of whether the relationship between psychological factors and IBS symptoms varies by subtype is also needed.

Although the OR for IBS was higher in participants taking probiotics, the evidence that probiotics are beneficial for IBS is largely established51. This may indicate that those with IBS symptoms were taking more probiotics than those without IBS symptoms.

This study has several strengths and limitations. The first strength is that this is the first study to investigate the ITQ in symptomatic IBS patients. Second, because many symptomatic individuals do not seek medical care52, surveying the general population likely provided a better picture of the general condition of IBS-symptomatic individuals. However, a follow-up study in a group of patients with professionally diagnosed IBS is needed for reproducibility. Third, we believe that using an online survey format lowered the psychological hurdle of answering sensitive questions and that the accuracy of the responses was achieved by avoiding missing responses and incorrect selections. The first limitation of this study is that it was cross-sectional in design, and thus causal relationships cannot be determined. Second, recall bias is always an issue concerning ACEs, considering that the responses are retrospective53. Third, because we chose to use an online survey for the reasons discussed above, we were unable to scrutinize the content of the traumatic experiences and make a clinical diagnosis of PTSD or CPTSD. Therefore, we cannot rule out the possibility that PTSD and DSO symptoms stem from other mental disorders. To rule out this possibility, a clinical PTSD/CPTSD/DSO judgment in a group of patients diagnosed with IBS is needed. However, ACEs that do not meet the current PTSD event criteria, including psychological abuse and bullying, may be important as traumatic events for IBS. In addition, Brewin et al. argue that it is preferable to diagnose PTSD on the basis of symptoms rather than events because the events that can cause PTSD vary according to individual differences in vulnerability to stress54. Fourth, the C-ACEs questionnaire was not validated before this study. However, the Japanese version of the 10-item ACEs questionnaire was well validated55, and Cronbach-alpha of C-ACEs showed reliable value. In future studies, a validated questionnaire that assesses a wide range of ACEs needs to be developed. The validation of ITQ Japanese version is in progress for completion, but Cronbach-alpha showed an excellent value (0.912). Fifth, because the present study did not investigate protective factors for ACEs, additional protective factors should be measured. Sixth, the important biological factors associated with IBS, such as gut microbiota, genes, and function of the hypothalamic–pituitary–adrenal axis, have not been investigated. Seventh, the survey targeted a Japanese population (or more precisely, residents of Japan). The same results are expected in samples from other races or cultures, but a follow-up study is needed.

Conclusion

The results of this study underscore the importance of CPTSD, especially DSO, in the development and maintenance of IBS. Future studies should follow up with another sample and clarify how CPTSD, especially DSO, is related to the pathology of IBS. Promoting trauma-informed care in the general public and medical settings could benefit society by helping to prevent IBS and improve the quality of life of IBS patients. In addition, appropriate diagnosis and intervention of DSO may be beneficial in the prevention and treatment of IBS. Therefore, DSO should be studied and treated clinically as a condition involved in various diseases, including IBS.

Method

Survey procedure

This survey was conducted from March to April 2023 using an online survey platform called Freeasy, operated by I-Bridge Corporation (Osaka, Japan). Freeasy is a system that distributes questionnaires to users registered on their reward sites and collects their responses in exchange for rewards, with a user base of 13 million. The survey was anonymous and did not include any additional questions related to personal information beyond the nine pieces of information on age, gender, marital status, presence of children, area of residence, household income, type of business, occupation, and type of residence that were available in advance. The survey was conducted in three steps: a two-stage screening survey of 47,000 individuals aged 18 to 75 to select IBS that met Rome IV criteria and HC groups, followed by the main survey (See Supplementary Fig. 1 for a flowchart of the survey process). The population ratios by age and gender of primary screening respondents were designed to be the same as those in Japan. The first screening investigated the following Rome IV criteria. It requires recurrent abdominal pain on average at least one day/week in the last three months, associated with two or more of the following three criteria: (i) related to defecation, (ii) associated with a change in frequency of stool, and (iii) associated with a change in form (appearance) of stool. In addition, these symptoms had to appear at least six months before. Respondents with uninterrupted daily pain were excluded. The first screening survey yielded 919 symptomatic IBS individuals who satisfied the criteria (1.96% incidence). The secondary screening survey then investigated the overall history, including IBS and laparotomy, medications, and current medical history. Respondents who satisfied the Rome IV criteria but had a surgical history of malignancy or resection of the gastrointestinal tract; current malignancy, including gastric and duodenal ulcers, inflammatory bowel disease, or colorectal cancer; or a history of colorectal cancer were excluded from the IBS group. The secondary screening survey narrowed the IBS group down to 721 individuals. To determine the HC group candidates, 4,667 individuals who did not meet the Rome IV criteria and had normal stool shape, matched for age, gender, household income, and region of residence to these 721 IBS symptomatic individuals, were randomly selected from the primary screening respondents and subjected to a second screening survey. The contents of the secondary screening survey were the same for both the IBS and HC groups. From among the respondents of the secondary screening survey, 721 individuals who matched the characteristics of the IBS group and were considered healthy with no current or past history of disease were randomly selected to form the HC group. Both groups were asked to complete the main survey; ultimately, valid responses were received from 669 individuals with symptomatic IBS and 659 healthy subjects (response rates of 92.8% and 91.4%, respectively). In this online survey system, hashed email addresses are matched with attribute information to exclude duplicate registrants, ensuring surveys are distributed without allowing duplicate responses. A proprietary algorithm eliminates potential inappropriate respondents. IP filtering is not performed. Respondents of both the screening and main survey were compensated by the research company.

Ethical considerations

For both the screening survey and the main survey, informed consent was obtained from all participants online before requesting responses. In particular, the advance instructions for this survey clearly stated that the questions included sensitive content such as sexual abuse, that answering the questions might lead to physical or mood changes, that respondents were free to stop answering at any time during the survey, and that they would receive reward only if they answered all questions. Respondents were asked to contact the research staff if any physical or mental health problems were exposed as a result of participating in the survey, but none did. This study was approved by the Ethics Committee of Tohoku University Graduate School of Medicine (No. 2022–1-1059, December 21, 2021). This study was conducted in accordance with the principles of the Declaration of Helsinki and its later amendments.

Primary endpoints

The primary endpoints were the IBS-SI-J, the C-ACE questionnaire, and the ITQ.

IBS-SI-J

The IBS-SI-J is a validated Japanese translation of the IBS-SI56, and its contents are the same as those of IBS-SSS43. It asks about the severity and frequency of abdominal pain, the degree of abdominal distension, the degree of satisfaction with abdominal symptoms, and the impact of abdominal symptoms on daily life. The score is expressed as a continuous number from 0 to 500, with a higher number indicating more severe IBS symptoms.

Comprehensive ACE questionnaire

The C-ACE questionnaire is a 15-question questionnaire consisting of 10 questions from the ACE Questionnaire57 used by the CDC in the U.S., plus five additional author-created questions on bullying, death or disability of one or more parents and/or siblings, interpersonal stress other than parents or children similar in age, accident or disaster, and low SES. The CDC’s 10 ACE questions, which ask about ACEs in the family (psychological/physical/sexual abuse, emotional/physical neglect, parents' divorce or separation, witnessing IPV, living with household member with addiction and mental illness, and incarceration of household member), are widely used worldwide. The validated Japanese version was used in this study55.

We created the five additional questions partly because there is no uniform questionnaire for examining CAEs outside the home and partly to explore the possibility that there is adversity unique to Japan.

The following four types of bullying were identified: psychological bullying (e.g., being bad-mouthed, made fun of, ignored, ostracized), physical bullying (e.g., being punched, kicked), sexual bullying (e.g., being touched, having one’s genitals exposed), and cyberbullying (e.g., being the subject of negative posts on social media sites, not being allowed to join groups). The survey asked, “Did you have a hard time being bullied by other children in your neighborhood, at school, etc.?” and then asked them to answer “never,” “sometimes,” “often,” “very often,” or “constantly” for each form of bullying. With reference to the ACE study questions58, if the respondent answered “often,” “very often,” or “always” for one or more types of bullying, he or she was considered to have experienced bullying. The bullying score was converted to never = 0, sometimes = 1, often = 2, very often = 3, and constantly = 4, and the bullying score, which ranged from 0 to 16, was the sum of the scores for each type of bullying. The internal reliability alpha for the bullying score in this sample was 0.811, which is considered good.

Regarding the death or disability of one or more parents and/or siblings, respondents were asked if “one parent had died,” “both parents had died,” “one or both parents had had a serious illness or disability,” “a sibling had died,” or “a sibling had had a serious illness or disability”59. If none of these applied to them, the respondents were asked to answer “No,” and if any of them did, they were asked to select the items that applied to them (multiple responses allowed). If one or more of the above were true, the respondent was considered “exposed”; if none of the above was true, the respondent was considered “not exposed.”

For interpersonal stress, respondents were asked to answer “yes” or “no” to whether they had been exposed to stressful experiences with relatives other than their parents (or equivalent caregivers) or with people other than friends similar in age, and if “yes,” to briefly describe the details. This enabled us to study the effects of corporal and other forms of punishment in schools and other settings. Even in cases where the respondents answered “yes,” the responses that were considered to be equivalent to abuse by parents or bullying by similar aged peers were excluded and included in the “no” category in order to avoid duplication with the questions about abuse by parents or bullying.

Natural disasters and accidents can also have a significant impact on a child’s subsequent health. In response to the question, “Have you been in any kind of major accident or disaster that forced you to change your life?”, the respondents were asked to select “None of the above” if none applied to them or to select from the following options if they had any such experiences (multiple responses allowed): car accident, train accident, plane accident, ship accident, earthquake, typhoon, tsunami, eruption, fire or explosion, crime or incident, exposure to harmful substances, and other. If the respondent chose “other,” they were asked to briefly describe the details of the experience. If the respondent had one or more of these experiences, they were considered to have experienced an accident or disaster.

To ascertain SES in childhood, the respondents were asked to answer the question “How would you rate the economic status of the family in which you grew up compared with other families?" using a five-point scale: “high,” “middle high,” “normal,” “middle low,” and “low.” The “low” response was considered to be “low SES”41.

All ACEs were defined as exposures that occurred when the respondents were under the age of 18. The ACEs score is the sum of the number of ACEs determined to be exposed. Scores on the C-ACE questionnaire ranged from 0 to 15. The internal reliability alpha of the C-ACE questionnaire in this sample was 0.808, which is considered good.

ITQ

The ITQ is a validated self-administered questionnaire that can screen for both PTSD and CPTSD according to the ICD-11 diagnostic criteria34. In the present study, we used the Japanese version developed by Kim et al.60. The ITQ first asks participants to describe the event in their life that distressed them the most, and to what extent they were troubled by three categories of PTSD symptoms related to that event (RE, AV, and TH) in the past month, and to what extent their daily life was disrupted (dysfunctional symptoms). Each of the three PTSD symptoms has two subscales, for a total of six questions. The respondents answered on a five-point scale: “not at all” (= score 0), “a little” (= score 1), “moderately” (= score 2), “a lot” (= score 3), and “very much” (= score 4). PTSD is diagnosed when at least one response in all three categories has a score of ≥ 2 and the respondent satisfies the criteria for dysfunctional symptoms. In addition, the respondents are asked about the three categories of DSO symptoms (AD, NSC, and DR) as symptoms of CPTSD and how they usually feel about the dysfunctional symptoms caused by these symptoms. Each of the three DSO symptoms has two subscales as well, and the respondent is considered to have DSO symptoms when at least one response in all three categories has a score of ≥ 2 and they satisfy the criteria for dysfunctional symptoms. A patient is diagnosed with CPTSD if they satisfy both the PTSD and DSO criteria, and with PTSD if they satisfy the PTSD criteria but not the DSO criteria. The PTSD and DSO scores, which are the sum of the scores of the six subscales of PTSD and DSO, respectively, can also be used as quantitative measures. In this study, if a respondent satisfied the criteria for DSO but not for PTSD, they were included in the analysis in the “DSO group” in order to determine the impact of DSO alone. The internal reliability alpha in this sample was 0.912, which is considered good.

Secondary assessment items

Secondary endpoints included height, weight, highest educational attainment, diet, exercise habits, presence and extent of alcohol consumption and smoking, probiotic intake, and presence or absence of acute gastroenteritis immediately preceding the onset of IBS symptoms.

For probiotics consumption, respondents were asked, “On average, how many days a week do you take probiotic foods, supplements, or medicines?”, and asked to select from 0 to 7 days. If they took probiotics at least one day per week, they were considered to have a probiotic intake habit.

Regarding the presence or absence of acute gastroenteritis immediately prior to the onset of IBS symptoms, we asked, “If you have symptoms troubling your digestive tract (esophagus, stomach, or intestines), did those symptoms first begin to appear immediately after you had an intestinal infection (acute gastroenteritis)?” The respondents were asked to choose one of the following options: “Yes,” “No,” “I don’t remember,” or “Not applicable” (i.e., no troubling symptoms in the gastrointestinal tract). If the respondent chose “Yes,” they were considered to have had acute gastroenteritis.

For details on the methodology regarding last education, diet, exercise habits, and presence and extent of alcohol consumption and smoking, see the Supplementary Materials.

Statistical analysis

For comparisons between the two groups, a chi-square test was used for qualitative variables and a Kolmogorov–Smirnov's test of normality was used for quantitative variables, followed by a t-test for variables showing a normal distribution and a Mann–Whitney’s U test for variables that did not. For multigroup comparisons, the Kruskal–Wallis test was used because the data were non-normally distributed by the Kolmogorov–Smirnov normality test, and the Bonferroni-adjusted probability of significance was used. For grouping of C-ACEs scores for IBS-SI intergroup comparisons, see the Supplementary Materials. Models for logistic regression analysis were created after confirming multicollinearity among variables. Basic factors that may be involved in the development of IBS and general health status were selected as possible confounders. For the mediation analysis, two models were created. Model A had the C-ACE score as the explanatory variable, IBS-SI as the outcome variable, and PTSD score and DSO score as the mediating variables, whereas Model B had the bullying score as the explanatory variable, IBS-SI as the outcome variable, and DSO score as the mediating variable. Controlling for age, gender, number of cigarettes smoked, and net alcohol intake, 95% confidence intervals were calculated based on 5,000 bootstrapped samples61. There were no missing values in the data. All p-values above were set at < 0.05 bilaterally. Statistical analysis was performed using IBM SPSS Statistics ver. 29.0.0.0, and mediation analysis was performed using PROCESS Procedure for SPSS version 4.2.