Psychological distress and coping efficacy in children with disorders of gut–brain interaction

Multiple psychological factors influence disorders of gut–brain interaction (DGBIs). We aimed to evaluate psychological distress in Colombian schoolchildren with and without DGBIs.


| INTRODUC TI ON
Disorders of gut-brain interaction (DGBIs) include functional abdominal pain disorders (FAPDs) such as functional dyspepsia (FD), irritable bowel syndrome (IBS), functional abdominal pain-not otherwise specified (FAP-NOS), and abdominal migraine as well as other non-pain manifestations such as functional constipation (FC), non-retentive fecal incontinence, aerophagia, cyclic vomiting syndrome, functional vomiting, functional nausea, and rumination. 1,2e prevalence rates of FAPDs differ across cultures, with evidence that lower socioeconomic status is associated with higher pain prevalence. 3Ethnicity, race, climate, society, and culture may also affect the epidemiology of FAPDs. 2 In South American children, for example, our group found a prevalence rate of 16.8%, which is higher than the prevalence rates of FAPDs found in European children (10.5%). 4ltiple psychological factors including psychological distress and coping efficacy influence the development, progression, and resolution of symptoms of DGBIs.10][12][13][14][15][16][17] Variability among cultural and socioeconomic factors [18][19][20][21] that are associated with DGBIs is understudied, and the specific influence of psychological factors such as anxiety and somatization in children with DGBIs in Latin America has never been studied.11,22 Understanding of the impact of psychological functioning on DGBIs in children of different geographical regions can advance our understanding of the pathophysiology of DGBIs and help design more tailored treatments for FAPDs.Thus, we aimed to analyze psychological functioning in children and adolescents with DGBIs compared with healthy peers in Colombian schools.We sought to establish the association between psychological factors and symptoms in children with DGBIs.We explored similar associations in the subset of children with FAPDs. In addition,we attempted to identify other factors from item-level analysis of questionnaires that would differ between children with DGBIs and healthy peers.We hypothesized that children in Colombia with DGBIs would have higher anxiety and somatic symptoms, and lower coping efficacy than healthy peers.

| MATERIAL S AND ME THODS
This study analyzed a dataset of responses from children ages

| Spanish translation of questionnaires
All questionnaires were originally developed for use in the English language.To conduct this study, first, the questionnaires were

| Statistical analyses
Data from participants who had DGBIs and those who did not have DGBIs (healthy peers) per the ROME III criteria were compared.
We further subclassified those with DGBIs as FAPDs versus nonpain-predominant DGBIs.Data analyses were performed using SAS version 9.4 (SAS Institute, Cary, North Carolina).We assessed reliability of the measures in our study using Cronbach's alpha.
In general, a score of more than 0.7 was considered acceptable.
Continuous variables were reported as mean ± SD, and categorical variables were reported as count and percentage.Comparisons between groups were analyzed using Student's t-test for normally distributed and Mann-Whitney U or Kruskal-Wallis rank test for non-normally distributed continuous data.Subgroup analysis was used to inspect other factors affecting outcome measures, given the modest sample size.To evaluate group differences, univariate and multivariable analysis was performed between each of the exposure variables of interest and the effect variable.Correlation among various measures was analyzed using Spearman correlation coefficient.p Values of less than 0.05 were considered statistically significant.Item-level analysis for all DGBIs measures was analyzed using chi-squared test, and for cell counts less than 5, Fisher's exact test with themes/subgroups of interest being identified for comparison between the groups.specific) is designed to measure subjective, consciously perceived feelings of tension, worry, and apprehension that differ in intensity and fluctuate over time, that is, how one feels at a particular moment in time.Trait anxiety evaluates differences between children in propensity to experience anxiety conditions; that is how one generally feels.The scores for each scale can range from 20 to 60 with higher scores suggesting worse anxiety.In addition to these subscales, we assessed differences in items on the state anxiety measure related to a positive affect versus a negative affect.

Questionnaire on Pediatric Gastrointestinal Symptoms-Rome III Version
Children's Somatization Inventory (CSI) 25,26 : The Children's Somatization Inventory (CSI) is a 35-item measure of the severity of bothersome nonspecific somatic symptoms experienced by children and adolescents (ages 8-17 years) and was originally designed for children with recurrent abdominal pain.Participants rate the items on a 5-point scale over a period of 2 weeks (Cronbach's alpha = 0.9).Scores can range from 0 to 140 with higher scores indicating higher level of somatization.Although it needs standardization and validation, the endorsement of 13 or more symptoms (corresponding with a score at or greater than 13) is generally considered to indicate clinically significant somatic distress.The scale is further split into subscales of GI and non-GI-related somatization items.In addition to evaluating these subscales, we also conducted an item-level analysis and attempted to identify themes in the questionnaires, such as differences in specific pain-related items versus other items between the groups.
Coping efficacy: We used a 7-item questionnaire that assessed general coping in children derived from the "How I Cope" measure of the Children's Coping Strategies Checklist (CCSC): Divorce Adjustment Project scale (test-retest reliability ranging from 0.64 to 0.80). 27For the proposed study, participants were asked to rate both their feelings and perceptions around addressing problems, either in the context of the past 6 weeks or in the future on a 4-point scale.
Total coping score was a summation of items 1-7 and ranged from 7 to 28.Coping within the past 6 weeks was measured via the summation of items 1-4, with a score range between 0 and 12. Future coping was measured via summation of items 5-7, with a score range between 0 and 9.During item-level analysis, we grouped items that measured past coping with future coping efficacy.

| Demographic and baseline characteristics
Of 1842 participants approached, 152 declined participation, and 165 did not meet inclusion criteria.Of the remaining 1525 participants, 29 did not complete questionnaires and were not included in the analyses.Of 1496 children, 281 (19%) met criteria for a DGBIs (mean age 12.9 ± 2.2 years, 49.8% females) (Table 1).Children without DGBIs (n = 1215) had a mean age of 12.7 ± 2.1 years and 50.6% females.The DGBIs group had a trend for more children in the 13-18 years compared with healthy youth (p = 0.09).There were no other baseline differences between children with DGBIs and healthy peers (p > 0.05).

DGBIs and healthy peers
Anxiety: Children with DGBIs had higher trait (p < 0.001) and lower state (p = 0.04, Table 2) anxiety than healthy peers.There was no significant difference between DGBIs and healthy peers while assessing positive affect (p = 0.16) and negative affect (p = 0.22) subscales of state anxiety between children with DGBIs and healthy peers.
Somatization: Children with DGBIs had higher somatization scores than their healthy counterparts (p < 0.001, Table 2 and Figure 1).In subgroup analysis of scales, both GI (p < 0.001) and non-GI symptoms (p < 0.001) reported on the CSI were higher in those with DGBIs.Those with DGBIs also had higher scores on painrelated (p < 0.001) and non-pain-related CSI items (p < 0.001).
Coping efficacy: Children with DGBIs had lower coping efficacy compared with healthy peers (p = 0.02, Table 2).Similarly, past coping was lower in those with DBGIs (p = 0.02), while there was a trend for future coping to be lower compared with healthy peers (p = 0.06).
After adjusting for age and sex, children with DGBIs had higher trait anxiety (p < 0.001) and somatization scores (p < 0.001), but they had lower coping efficacy (p = 0.02) and a trend for lower state anxiety (p = 0.06, Table 3) than healthy peers.

| Correlation between scores in children with DGBIs
There was a strong positive correlation between total somatization and GI as well as non-GI symptom subscales (r = 0.84 and r = 0.98, respectively, Table 4).There was a moderate positive correlation between somatization and trait anxiety scores (r = 0.50) and between past and future coping efficacy scores (r = 0.43).

| Differences in scores based on demographic characteristics
Females had higher levels of trait anxiety (p = 0.04) and somatization (p = 0.005) comparing all DGBIs.Scores differed based on different cities (La Union, Cartagena, Florencia, Quilichao) with state anxiety (p = 0.02), trait anxiety (p = 0.03), and coping efficacy (p < 0.001).However, they did not differ based on age, only child status, separated parents, or nutritional status between DGBIs cases and healthy children (Table 5).

| Anxiety
The internal consistency for state and trait anxiety calculated in our study was found to be 0.75 and 0.83, respectively (Cronbach's alpha).This is comparable with prior validation studies that reported state anxiety values of 0.82 for males and 0.87 for females, and trait anxiety values of 0.78 for males and 0.81 for females. 24Within the state anxiety measure, internal consistency of items related to positive affect versus negative affect was found to be 0.85 and 0.84, respectively (Cronbach's alpha).

| Somatization
The internal consistency calculated for total CSI was 0.87 (Cronbach's alpha 0.66 and 0.83 for GI and non-GI CSI scales, respectively) and was comparable with prior validation studies (Cronbach's alpha = 0.9 25 and 0.88, 26 respectively).Cronbach's alpha for painrelated items and for non-pain-related items compared to total CSI was 0.91 and 0.98, respectively.

FAPDs and healthy peers
Children with FAPDs had higher trait anxiety (p = 0.02) and somatization (p < 0.001, Table 5) than healthy peers.There was a trend for lower state anxiety in FAPDs compared with healthy peers (p = 0.07).However, coping efficacy did not differ between FAPDs and healthy peers (p > 0.05).After adjusting for age group and sex, children with FAPDs had higher trait anxiety (p = 0.02), and somatization (p < 0.001, Table 3) than healthy peers.There was a trend for lower state anxiety in FAPDs compared with healthy peers (p = 0.08), but coping efficacy did not differ (p ≥ 0.05).Females with FAPD had higher levels of somatization (p = 0.04).Scores also differed based on different cities (La Union, Cartagena, Florencia, Quilichao) with state anxiety (p = 0.03) and trended to differ with trait anxiety (p = 0.08), somatization (p = 0.05), and coping efficacy (p = 0.08).They did not differ based on age, only child status, separated parents, or nutritional status between children with FAPDs and healthy peers.

| DISCUSS ION
Disorders of the gut-brain interaction result from the interplay of genetic, biological, psychological, social, and environmental factors. 1,2Similar to rates in the US (25% and 13.3%, respectively 28

TA B L E 4 Correlation between psychological measures in children with disorders of gut-brain interaction.
Total CSI

Trait anxiety (STAIC2)
Total this rate of FAPDs was lower than that previously published in South American children, comparable rates have been found for FAPDs in Europe (10.5% and 7.7%, respectively). 4,17The aim of this study was to explore the association of psychological outcomes in Latin American children with DGBIs and compare with healthy peers.We found higher rates of trait anxiety, somatization, and lower coping efficacy in children with DGBIs compared with healthy peers.While females had worse anxiety and somatization which differed based on geographic location, other demographic and socioeconomic factors did not differ.Importantly, this is one of the largest studies to report psychological functioning outcomes in children with DGBIs and healthy peers.
Trait anxiety was higher in children with DGBIs, while state anxiety at any given point of time had a trend to be lower than healthy peers.This is consistent with higher rates of anxiety (40%-77%) reported in prior studies [7][8][9]11,[29][30][31][32][33][34] of DGBIs including cyclic vomiting (64%), 31 fecal incontinence (77%), 34 rumination, 29,30 and nausea. 33 The dfference between the two subscales can potentially be explained by the fact that the trait anxiety is generally more stable and generally demonstrated better internal consistency (Cronbach's alpha >0.78) and test-retest reliability after 6 weeks.24 One important finding from this study is that individuals with DGBIs differed from healthy peers in terms of their emotional sensitivity (e.g., worry about making mistakes, worry about what others thought of them, feeling like crying) as well as being more attentive to bodily sensations (e.g., greater symptoms of stomach pain in addition to more frequent headaches, lower back pain, etc.). Suh youth may have a shared risk factor, such as high behavioral inhibition, a pattern of behaviors involving fear/avoidance and over-arousal of the sympathetic nervous system, which is a predictor of the development of both DBGI 35,36 and anxiety disorders 37 in youth. 7 omatization includes experiencing symptoms affecting multiple organ systems that cannot be explained medically. 9,32 In ition to endorsing many GI-specific symptoms (e.g., abdominal pain, nausea), children with DGBIs reported other mental health symptoms (e.g., worry, feeling like crying) and physical symptoms such as widespread pain complaints including headaches, chest pain, lower TA B L E 5 Psychological measures based on demographic and baseline characteristics in children with disorders of gut-brain interaction (n = 281).back pain, joint pain, and pain in the arms or legs.Thus, an itemlevel analysis allowed us to gain a nuanced understanding of common physical and mental symptoms impacting youth with DGBIs.
We found higher total scores, GI, and non-GI somatic symptoms in children with DGBIs than controls.Both GI and non-GI subscales highly correlated in our sample.This is similar to previously reported literature within both adolescents and adults in the US 31 as well as children in Germany. 17 addition, we also subgrouped pain-related and non-painrelated items which were both higher in children with DGBIs than healthy controls.Somatization has resulted in worse psychosocial and functional outcomes in DGBIs. 9Somatic symptoms frequently co-occur with chronic abdominal pain in children, 37 and sometimes non-GI somatic symptoms precede the diagnosis of functional abdominal pain (FAP). 12,35In fact, somatization has shown to mediate the relationship between anxiety and symptom severity in adult and pediatric IBS patients. 9,10,32This is consistent with our findings where somatization scores correlated with trait anxiety.
We also found lower general coping efficacy and lower resilience in handing difficult situations in children with DGBIs compared to healthy controls.Both past coping and future coping efficacy subscales were lower in children with DGBIs.Thus, conflict resolution, assertiveness training, or resilience building could be beneficial to improve outcomes for this population.Unlike prior studies, we did not find a strong correlation between general coping efficacy and somatization or anxiety. 38This may be secondary to culturally specific factors such as mental health stigma which may be more prevalent in Latin American countries. 39,40ere is conflicting literature on the impact of demographic and socioeconomic factors on DGBIs in Latin America.Some studies have shown that older children, children with a higher socioeconomic status, 21 females, 20 and children whose parents were separated 11 have a greater prevalence of DGBIs.In contrast, another study did not find any difference in the prevalence of DGBIs with race, family composition, parent marital status, or history of DGBIs in the household. 20We found that the DGBIs group had more children in the 13-18 years compared with healthy youth.
However, we did not find any difference in prevalence of DGBIs, or score responses with age, composition of household, marital status of parents, or nutritional status.In addition to prevalence, we compared score responses based on different demographic and socioeconomic factors in children with DGBIs.Females had worse trait anxiety and somatization than males with DGBIs.
Coping efficacy was lower in La Union and Florencia compared with Cartagena and Quilichao regions.Somatization was higher in Florencia compared with La Union, Cartagena, and Quilichao regions.While the reason for these differences is unclear, these cities are in different regions of the country where there are climatic, social, cultural, and economic differences.They have differences in the number of inhabitants.Some are state capitals, and while others are county cities, and these could potentially explain the differences.
FAPDs in particular have been related to negative psychological outcomes like anxiety [6][7][8]10,12 and depression, 10,13,15,16 somatization, and poor coping. Siilarly, we found that children with FAPDs had worse trait anxiety and somatization, and a trend for lower trait anxiety compared with healthy peers.
Anxiety is reported in 42%-85% youth with FAPDs and has been related to increased functional disability, somatization, social and academic impairment, heightened pain sensitivity, and long-term persistence of pain. 7,41Somatic symptoms have been associated with greater pain severity and pain-related functional disability, negative pain cognitions, and emotional distress. 10,12,36,42,43In our sample, coping efficacy did not differ between FAPDs and healthy peers unlike all DGBIs.This is contrary to prior studies reporting passive coping responses to be associated with more pain via somatization and either anxiety or depression. 7,10The caveat could be that these studies used pain-specific coping measures which may be more useful while assessing coping mechanisms in FAPDs.
In terms of social and demographic determinants, coping efficacy in patients with FAPDs was lower in La Union and Florencia compared with Cartagena and Quilichao regions, while somatization was higher in Florencia compared with La Union, Cartagena, and Quilichao regions.A German study cited higher prevalence of FAPDs within children not living within one household compared with those who had parents in one household. 17However, we did not find a such a difference within our cohort.
The strength of our study was the large sample size of Colombian youth with a prospective study design which facilitated analyzing multiple aspects of different psychological dimensions.We performed an item-level and subscale analyses to infer meaningful comparisons.We used child-reported measures in our sample in contrast to caregiver reports.While emotional and behavioral health has been reported to be more impacted in children 6-10 years of age according to caregiver reports, 17 other studies have shown that children may be more sensitive informants of their own anxiety in comparison. 6This stresses the importance of screening with childreported measures to avoid missing internalizing symptoms.We included social determinants of health such as parental status, descendant status, and geographical location in our assessments.This may be particularly important and affect health care disparities and access to care in children with DGBIs.DGBIs have been thought to be diagnosed less frequently in lower socioeconomic populations as these patients and families may not often be seen by medical providers. 44Our study provided data in the general population demonstrating a more real-life prevalence.
We were limited by the cross-sectional nature of the study allowing for only a snapshot for comparison between children with DGBIs and healthy peers rather than assessing the stability of these measures over time.Limiting the investigation to participants from public schools alone could limit the generalizability of the study.
With our current study design, we were unable to compare differences between FAPDs and non-pain-related DGBIs.However, this could be the basis for future studies to understand how their differences could affect treatment strategies as well as outcome.
The influence of behavioral therapy on psychological measures can also be explored in future studies.In conclusion, our study demonstrated the role of trait anxiety, somatization, and lower coping efficacy in children with DGBIs compared with healthy peers in Latin America.Most treatment strategies in children with DGBIs are focused on gut-mediated therapies, despite widespread extraintestinal manifestations. 45Our study emphasizes the role of psychological functioning in these children, and the need for effective screening and treatments targeted to their improvement which can in turn, improve GI symptoms and optimize outcomes.
There may be a shortage of mental health providers or barriers to access in Latin America.However, newer technologies such as internet-delivered therapy, telephone-guided treatments, and mobile applications may be of assistance in such situations to optimize care.

8 -
18 years from the largest public school in five geographical dispersed regions of Colombia: Cartagena (Atlantic region), Florencia (Amazonic region), La Unión (Andean region) and Quilichao (Pacific region).Data were collected by members of the research team of Functional International Digestive Epidemiological Research Survey (FINDERS), the largest international consortium dedicated to accruing data on children with DGBIs.This study was approved by the Institutional Review Board and Human Subjects Committee of Universidad del Valle of Cali, Colombia, and the academic authorities of each school.Demographic information (marital status of parents, size of household) and information on gastrointestinal (GI) family history and the participant's past medical history were obtained from the parents.The study included children from third to eleventh grade without a history of organic medical conditions.Children received instructions regarding the study questionnaires completion by the research team at each school.At the end of the instruction session, children were encouraged to ask for clarification on questions or wording that they may have not understood.A member of the research team was present during study completion to assure confidentiality and provide assistance in case the children had difficulties completing the questionnaires.Measures of DGBIs classification (QPGS-III questionnaire), anxiety, somatization, and coping efficacy were collected (see below).
translated into Spanish and adapted to the local language by three bilingual physicians of the Functional International Digestive Epidemiological Research Survey FINDERS.The Spanish versions subsequently underwent reverse translation and were assessed for fidelity by comparison with the original English versions of each questionnaire with the translated version.Focus groups of schoolchildren and adolescents confirmed their understanding of the terms of the questionnaire.To evaluate the possibility of transcriptional errors, 10% of the records were reviewed and compared with the original forms.

1 . 2 .
Children with disorders of gut-brain interaction (DGBIs) have higher anxiety, emotional sensitivity and somatization, and lower coping efficacy compared with healthy youth.Children with functional abdominal pain disorders, in particular, have worse anxiety, and somatization compared with healthy youth.3.This highlights the importance of appraising psychological distress characteristics as well as incorporating conflict resolution, assertiveness training, and resilience building during the treatment of DGBIs.

(
QPGS-III) 23 : The QPGS-III is a validated questionnaire based on Rome 3 criteria to diagnose DGBIs in children 4-18 years of age.Questions review gastrointestinal symptom location, frequency, and severity as well as related disability and somatic symptoms on a fivepoint scale.Patients filled the QPGS-III questionnaire for Rome 3 criteria since the Rome 4 diagnostic questionnaire was not validated at the time of data collection.State-Trait Anxiety Inventory for Children (STAIC) 24 : The State-Trait Anxiety Inventory for Children (STAIC) is a measure of general anxiety symptoms in children.It is validated for children ages 9-12 years but can be used in populations under 18 years of age.The child-response questionnaire consists of two 20-item subscales with responses scored on a three-point system.State anxiety (context- migraine, 14 functional abdominal pain syndrome, 21 FAP-NOS.The others had non-pain-related DGBIs: 138 FC, 1 non-retentive fecal incontinence, 4 aerophagia, 6 cyclic vomiting, and 7 had rumination. and healthy peers.However, healthy peers reported increased ability to improve difficult situations in the past 6 weeks (p = 0.02) and in the future (p = 0.009) compared to children with DGBIs.

F I G U R E 1
Differences in somatization between children with DGBIs and FAPDs with healthy peers.Panel A. Total somatization scores between DGBIs and healthy peers.Panel B. GI CSI subscales between DGBIs and healthy peers.Panel C. Non-GI CSI subscales between DGBIs and healthy peers.Panel D. Total somatization scores between FAPDs and healthy peers.CSI, children's somatization inventory; DGBIs, disorders of gut-brain interaction; FAPDs, functional abdominal pain disorders; GI: gastrointestinal.† All data are reported as mean (SD).p Values <0.05 indicate statistical significance.

children (n = 1215) DGBIs (n = 281) Total (N = 1496) p Value Mean (SD) Mean (SD) Mean (SD)
to healthy peers.Interestingly, they also had worse mental health symptoms and emotional sensitivity than healthy peers.For instance, they reported feeling more upset, worried about making mistakes and what others thought of them, had trouble deciding what to do, and felt like crying more often than healthy peers.Moreover, they reported greater psychosomatic symptoms such as "noticing their hearts beating fast" and "having a funny feeling in their stomachs."Theyhadhigher pain and fatigue related non-GI somatic symptoms such as headaches, chest pain, musculoskeletal pain, muscle weakness, gait disturbances and lower energy than their healthy counterparts.Lastly, they reported other extraintestinal symptoms such as dyspnea, hot or cold spells, lump in throat, loss of voice, deafness, blindness, and dysuria.Coping efficacy (general coping to past and future situations) did not differ much between children with DGBIs TA B L E 2 Comparison of psychological measures and subscales between children with disorders of gut-brain interaction and healthy peers.MeasureHealthy Note: Bold values indicate p values ≤0.05.Abbreviations: DGBIs, disorders of gut-brain interaction; CSI, children's somatization inventory; GI, Gastrointestinal; STAIC 1: state-trait anxiety inventory for children-state anxiety; STAIC 2: state-trait anxiety inventory for children-trait anxiety.a Subscales identified during the study.