Systematic Review and Meta-analysis on the Incidence, Prevalence and Determinants of Discomfort in Inflammatory Bowel Disease

Abstract Background The symptom burden in inflammatory bowel disease (IBD) has a significant negative impact on the health-related quality of life (HRQOL). Patients with IBD report physical, psychological and social discomfort even during remission. Aim To synthesize the best available evidence to determine the worldwide incidence, prevalence and determinants of discomfort in adults with inflammatory bowel disease (IBD). Methods Following PRISMA recommendations, we searched the Medline, CINAHL, PsycInfo, Embase, Cochrane, Campbell and JBI Evidence Synthesis databases for studies on either incidence or prevalence of discomfort in English until January 2021. Data were extracted using the Joanna Briggs Institute’s standardized extraction tools. Data that directly reported or could be used to calculate the incidence and prevalence of discomfort were extracted. Ten studies were eligible for inclusion in this review. Overall, the methodological quality of the included studies was considered moderate. Data measuring the incidence of discomfort in 6 out of 10 identified studies using the same measurement tool (EQ-5D) were pooled in a meta-analysis. Additional results have been presented in a narrative form, including tables. Results There is no standardized definition or tool utilized to describe or measure discomfort in IBD. Synthesized findings demonstrate that discomfort is prevalent among adults living with IBD. Determinants of discomfort included health literacy, disease activity, hospitalization/surgery, age and gender, delayed diagnosis, local practice standards and quality of IBD care. Conclusions More research is needed to identify the impact of discomfort on health-related outcomes for people with IBD and consequently appraise discomfort interventions for their efficacy.


Introduction
Inflammatory bowel disease (IBD) has overwhelming consequences on the health-related quality of life and health care utilization (1). It has been estimated that IBD affects five million people worldwide (2), with a prevalence of 0.5% in North America and a rising incidence posing concerns for an emerging epidemic (3) with unsustainable long-term care (4,5). The complex pathogenesis of IBD leads to a life-long, unpredictable, relapsing and remitting illness course (6,7). The symptom clusters in IBD (pain, discomfort, anxiety and depression) are comparable to those in cancers and have been associated with a decreased health-related quality of life (HRQOL) (1). Thus, there is a critical need to understand the broader impact of the multidimensional nature of IBD discomfort.
There is no published systematic review on the incidence, prevalence, and determinants of discomfort in IBD. An increased understanding of the incidence and prevalence of discomfort would be invaluable for IBD care providers and policymakers to develop effective strategies to manage IBD, provide targeted support for individuals, and address healthcare systems' implications. Therefore, it is necessary to synthesize the findings of studies conducted on this area to appraise the strengths and limitations of such studies and identify evidence on the prevalence and incidence of discomfort among adults with IBD.

Objective
The aim of this review was to synthesize the best available evidence to determine the worldwide incidence, prevalence, and determinants of discomfort in adults with IBD. Therefore, the following research questions are addressed: (1) What is the global incidence of discomfort in adults with IBD? (2) What is the global prevalence of discomfort in adults with IBD? and (3) What are the determinants of discomfort in IBD?

Participants
This review considered studies conducted worldwide involving adults (18 years and older) with IBD reporting discomfort. Studies examining the experience of discomfort in children with IBD have been excluded.

Concept
The concept of discomfort has been recognized as a component of illness and suffering, and its conceptualization is essential for measurement (8). The definition of discomfort used to select studies was that of 'a negative physical and/or emotional state, causing unpleasant feelings or sensations' (9). Although a familiar concept in clinical practice, research studies have misused discomfort as a surrogate for pain (9). However, while pain can lead to discomfort, not all discomfort is a consequence of pain (9).
The deleterious impact of discomfort on patient outcomes has been previously documented. For example, anal pain or discomfort has been reported as the most critical factor in Crohn's disease (10). Abdominal discomfort has been described in the context of disease severity, frequent surgical interventions, and hospitalizations (11), abdominal cramping and diarrhea (12). Other sources of discomfort include perianal disease (13), abdominal lump (14), changes in bowel function (15), abdominal distention and flatulence (16), decreased appetite, nausea, vomiting, abdominal tenderness, difficulty passing gas, and sleep disturbance (14,17), food triggers and restrictions (18,19), and dietary concerns (20). Additional sources of discomfort include emotional, relational, familial and employment-related challenges (21), restless leg syndrome (17), worrying about discomfort even in the absence of actual symptoms (22), talking about the negative aspects of the disease (23), lifestyle limitations, social interactions (15) and lack of accommodation by others (24).
In women with IBD, vulvar and vaginal discomfort consisting of pruritus, burning and irritation (25) and sexual dysfunction (26) have been associated with decreased functional status and lower scores on Patient-Reported Outcomes Measurements Information System (PROMIS) measures of emotional and mental health (25). Furthermore, physical and psychological discomfort persist in IBD even in remission (27,28). Therefore, it is imperative to concede, evaluate, and prevent the sources of discomfort to provide patient-centred IBD care (29).
This review primarily considered studies that assessed and reported on the incidence and prevalence of discomfort in IBD. The symptom of discomfort was captured as reported within the studies. Those with concept confusion, where discomfort was not defined or was used to describe pain in the context of endoscopic or imaging evaluations, have been excluded.

Search Strategy
The essential information sources in this systematic review included Medline (Ovid), CINAHL (Ebsco), PsycInfo (Ovid), Embase (Ovid), Cochrane (Wiley), Campbell and JBI Evidence Synthesis (see Supplementary Appendices I and II). The search strategy was designed and conducted in collaboration with an experienced scientific librarian and aimed to identify the published and unpublished studies written in English. The included search terms were 'discomfort' , 'inflammatory bowel disease', 'Crohn's' and 'colitis' .. The search for unpublished studies and grey literature (33) included trial registers and ProQuest Dissertations and Theses. The electronic searches were enhanced with hand-searching of reference sections from studies retrieved via databases. Databases were searched from their inception to January 2021. Next, the reference lists of all reports and articles selected for critical appraisal have been screened for additional studies. Finally, an author search has been conducted on the authors' names known to have researched the review objective.

Study Selection
All the identified citations have been collated and uploaded into Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia) (34) and duplicates removed. Following a pilot test, the full text of selected citations has been assessed in detail against the inclusion criteria by two independent reviewers (O.I.N., R.W.), with arbitration about final inclusion from a third reviewer (C.G.) where required.

Data Extraction and Analysis
Potentially relevant studies have been retrieved in full, and their citation details were imported into the Unified Management, Assessment and Review of Information ( JBI SUMARI; JBI, Adelaide, Australia) (35). A priori protocol guided this review, with the intent to pool the included studies, where possible, in a statistical meta-analysis using JBI SUMARI. The selected studies were then assessed for methodological quality by two independent reviewers using the standardized critical appraisal instrument for prevalence studies ( Joanna Briggs Institute Critical Appraisal Checklist for Studies Reporting Prevalence Data) (30) and the critical appraisal checklist for cohort studies (36). Due to the variability across studies, attributed to methodological, clinical, geographical and statistical differences, the critical appraisal results are presented in Tables 1 and 2 with an accompanying narrative.

RESULTS
After the databases were searched and duplicates were removed from retrieved records, 1531 titles and abstracts were screened. Articles that did not fit the inclusion criteria were excluded resulting in 40 citations identified as appropriate for detailed, full-text assessment. Thirty studies were excluded after full-text evaluation. Data were only included once for studies reporting on the same participants (n = 4). The remaining 10 articles were selected for critical appraisal.

Characteristics of Included Studies
The final 10 studies included in this systematic review consist of two prospective observational cohort studies (29,37), two retrospective chart review and patient self-completion questionnaires (38,39), one retrospective observational study (40), a cross-sectional study design (online survey) (41,42), one large, multinational, cross-sectional survey (43), a cohort study (44) and a multicenter, observational study (45). In addition, the authors have been contacted where data were missing to seek further clarifications (n = 6), with one response (41) providing more results. The characteristics of the ten included studies are shown in Table 4. The data extracted included specific details about the participants, condition, other characteristics and outcome information, including the proportion of people reported with either current or period or lifetime prevalence of discomfort in IBD or outcome data. Only the baseline data were extracted in cohort study designs that measured the prevalence of discomfort with multiple data points.  (40), Spain (43) and the USA (38,39,41,43). Sample sizes ranged from 52 (37) to 2093 (44) participants with IBD, and the proportion of female participants ranged from 19 (37) to 1250 (41). The age of participants ranged from 18 to 84 years. Two studies enrolled participants from outpatient settings (29,45), IBD patient registry (37), online platforms (41,42), multinational participants (43) and multicenter participants (45). The observational period ranged from 2 (43) to 180 months (40). One study did not specify the observational period (42). Three studies collected data retrospectively (38)(39)(40), three collected data prospectively (29,37,45), three were cross-sectional studies (41)(42)(43), and one was a cohort study (44). Participants in remission ranged between 19% (38) and 55.44% (43).

Methodological Quality
All the 10 included studies underwent critical appraisal, data extraction and synthesis to capture valuable insights regardless of their methodological quality, to gain a richer understanding of discomfort. As a result, no study was excluded based on the quality assessment. However, the nine included prevalence studies scored between 8 to 9 out of 9 on the Critical Appraisal Checklist for Studies Reporting Prevalence Data (30) ( Table 3). One cohort study (44) scored 5 out of 11 on the critical appraisal checklist for cohort studies (36). Five out of ten studies answered 'Yes' for each critical appraisal question, meaning that the risk of bias across included studies was moderate. The reviewers ensured that the participants in the included studies were representative of the target population and were adequately recruited. Most of the included studies provided adequate details about their participants and study settings. However, one study did not specify the observation period (42).

Discomfort Measurement Instruments
Six (29,(37)(38)(39)42,43) out of the ten included studies used the standardized, Component 1 of the EuroQoL 5 Dimensions (EQ-5D) to measure discomfort. The EQ-5D is a non-diseasespecific instrument that evaluates HRQoL while capturing the patients' experiences with pain/discomfort, rated from 1 = no problems to 3 = severe/extreme problems (29). While the psychometric properties of EQ-5D have been established, EQ-5D does not differentiate further between pain and discomfort and lacks the required sensitivity to differentiate further at the subgroups analysis (42).
One study reported the number of patients with perianal discomfort as a chief complaint without specifying a measuring tool (40). Another study reported the number of women with IBD experiencing vulvar and vaginal discomfort based on the patients' response to a vulvovaginal symptom survey which included three separate questions about feeling itching, burning or irritation in the past month (41). However, no specific phenotype of vulvovaginal discomfort was found to be dominant (41). A single study used the validated Structured Assessment of Gastrointestinal Symptoms (SAGIS) to assess the severity of pain/discomfort at defecation on a 5-point scale (44). Finally, the last included study described using the 'Standard Set of Patient-centered Outcomes for Inflammatory Bowel Disease -an International, Cross-disciplinary Consensus', a prestandardized set of patient-centred outcome measures for IBD developed by an international working group (45).

Prevalence of Discomfort
The prevalence of discomfort was synthesized for each included study. The prevalence was expressed as the proportion/percentage of study participants with discomfort as determined by the measuring tool. Data on the number of patients with IBD reporting discomfort ranged from 7.5% of women with IBD reporting vaginal discomfort while in remission (41) to 88.7% of patients with moderate to severe Crohn's disease (39) (Table 4).

Meta-analysis
Since all the included studies used different measurement tools for discomfort, pooling all the studies was not suitable. However, the six studies which measured discomfort using the EuroQoL 5 Dimensions (EQ-5D) component 1 were pooled separately using a random-effects model in JBI SUMARI. The prevalence of discomfort among participants living with IBD was generally high. In the six studies using the EuroQoL 5 Dimensions (EQ-5D) component 1, the prevalence of discomfort ranged from 43.3% to 99.1%. The pooled prevalence was 71.4%, with a 95% CI of 48.6% to 89.5% (Figure 2).
Heterogeneity was determined using the I 2 statistic and was very high (>99.5%). The number of included studies in the pooled analysis limited the probe of this level of heterogeneity. The observed heterogeneity could be attributed to the definitions for discomfort and its components, the various contexts in which discomfort was measured, and IBD practice standards across several countries, which may have impacted participants' observed levels of discomfort. The number of studies and data identified in this review prevented us from reporting discomfort rates for any specific subgroups.

Incidence of Discomfort
Nine out of 10 included studies offered data that permitted the calculation of discomfort rates per person-year, using the

Health Literacy (HL)
In one included study (37), patients with CD reporting pain/ discomfort had a significantly lower median HL (12.5 [10.3-14.8], P = 0.02). However, this prospective and observational study did not establish a causal relationship between the two.

Disease Activity
The studies included in this systematic review used a range of disease activity indices. A significantly higher median Crohn's Disease Activity Index (CDAI) has been described in CD patients who faced more difficulties with pain/discomfort (123. 6 (41).

Hospitalization/Surgery
More patients with CD who required surgery or hospitalization (54%) reported pain/discomfort when compared to those who did not (49%) (37).

Age and Gender
Younger female participants' were more agreeable to respond to inquiries about vulvovaginal symptoms (41). Similarly, more women with CD (805 of 1144 (70%)) completed the vulvar and vaginal symptom survey when compared with women with UC/IC (445 of 690 [65%]), P = 0.01. In addition, the participants with perianal discomfort were significantly younger than those reporting other symptoms (40).

Delayed Diagnosis
Patients with perianal discomfort had active symptoms for an extended period before their diagnosis was established (40). Perianal discomfort was found to be a significant and only factor associated with a long diagnostic delay (OR 10.23, 95%CI: 1.93-54.37) in Crohn's disease (P = 0.006), but not in UC (40).

Local Practice Standards
Patients with CD presenting with perianal discomfort in Korea experienced longer physician-dependent delays the period from initial presentation to diagnosis) (45). The authors attributed these findings to local practice standards, which differ from the West, and include referring patients presenting with perianal discomfort to a surgeon or general practitioner instead of a gastroenterologist as IBD is uncommon in the East (40).

Quality of IBD Care
The patient satisfaction with the quality of their IBD care was lower in patients experiencing pain or discomfort in a bivariate analysis evaluating patient-centred outcomes (P = 0.043) (45).

Discussion
This is the first review that systematically synthesized the best available evidence to determine the worldwide incidence,   Journal of the Canadian Association of Gastroenterology, 2022, Vol. 5, No. 2 prevalence and determinants of discomfort in adults with IBD. The characteristics of the adult population with IBD experiencing discomfort support the underlying complexities of IBD across disease type, disease activity, age, gender, range of measurement tools, practice standards and relationships with the health care teams. The findings of this review suggest that discomfort is prevalent in adults living with IBD. Therefore, health care professionals providing care to adults with IBD should recognize discomfort as an independent symptom that can impact their quality of life and perception of quality care. Consequently, standardized measurement and evaluation strategies of discomfort should be considered in clinical practice to deliver patient-centred services.
The variability of the studies reporting discomfort in IBD limits our understanding of this symptom. Patient-centred research will allow the exploration of patients' care priorities and recognize their sources of discomfort or distress (29). Thus, future research efforts should include studies to evaluate the efficacy of discomfort interventions for people with IBD.

LIMITATIONS
The inclusion of only English-language studies may have led to the exclusion of seminal studies. Given the quality of the included studies, the lack of differentiation between pain and discomfort in the EURO-QOL 5 domain, and the high level of heterogeneity observed during the meta-analysis, the findings of this review must be interpreted with caution. In addition, there was insufficient data to conduct subgroup analyses.

CONCLUSION
Discomfort is prevalent in adults with IBD. Determinants of discomfort amenable to intervention may include health literacy, disease activity, hospitalization, surgery, age and gender, delayed diagnosis, local practice standards and quality of IBD care. Unfortunately, there is no specific, validated measure of discomfort in IBD. Therefore, a consensus on how discomfort is defined, acknowledged and investigated is recommended to represent patients' experiences and care needs accurately.

AUTHOR CONTRIBUTIONS
O.I.N.: Substantial contributions to the conception and design of the study; acquisition and analysis of study data; drafting the work; development and final approval of the manuscript; agreement to be accountable for all aspects of the work. R.W.: Substantial contributions to the conception and study design, acquisition, and analysis of data, drafting the work, development and final approval of the manuscript, agreement to be accountable for all aspects of the work. C.G.: Substantial contributions to the conception and study design, data analysis, development and final approval of the manuscript, agreement to be accountable for all aspects of the work. A.R.W.: Substantial contributions to the conception and study design, scientific library management.

Funding
A Nursing-Led IBD Research Grant supports this systematic review through Crohn's and Colitis Canada and the Canadian IBD Nurses (CANIBD).