There is a high prevalence of mental health disorders (MHD) in those with established Inflammatory bowel disease (IBD). Two large meta-analyses have focussed on anxiety and depression, finding anxiety disorder present in 21% (but anxiety symptoms in up to 35%) and depressive disorder in 15% (with depressive symptoms in up to 25%) (1, 2). Comparatively, when last surveyed 17% of the United Kingdom (UK) general population met criteria for common mental health disorders (3). Both studies found a higher prevalence of depression and anxiety symptoms in those with Crohn’s disease (CD) than Ulcerative colitis (UC) and those with active disease over remission. Links have also been established between IBD and broad range of other MHD including deliberate self-harm, eating disorders, bipolar disorder and post-traumatic stress disorder (4–6). The relationship between IBD and MHD can be viewed as bi-directional. Both anxiety and depression are associated with higher rates of treatment escalation, hospitalisation and emergency service utilisation amongst IBD patients, whilst active IBD is linked to the future development of anxiety or depression (7, 8). A large primary care study has highlighted this amongst UK IBD populations (9). It has been demonstrated that targeted psychological interventions, where a need is identified, can have a positive impact on IBD associated outcomes (10).

Though attempts at determining the prevalence of MHD at the first presentation of IBD have been limited, the UK IBD standards group make clear the need for psychological assessment in the initial management of patients newly diagnosed with IBD (11, 12). A 2023 review focussing on the practicality of delivering psychological support in established IBD patients highlighted the time constraints placed upon outpatient clinics (13). The authors recommended the use of psychometric questionnaires such as the Hospital Anxiety and Depression Scale (HADS), which can take up to 5 minutes to complete (14). Time is often limited in new patient clinics and diagnostic uncertainty can increase the burden of anxiety. The option to utilise a tool that combines assessments of disease activity, disability and mental health burden is appealing. The IBD Disk ( Supplementary Appendix 1 ) was first developed in 2017 (15). It represents a shortened self-administered version of the IBD Disability Index (IBD-DI) (16). The IBD-DI was developed as a multi-organisation cooperation based on four preparatory studies, a consensus conference and a subsequent operationalisation process (16). Conversely, the IBD disk followed an abridged development process and arose from an iterative Delphi consensus based upon opinion from selected gastroenterologists attending an industry funded training programme. Nonetheless, it was developed specifically with the ‘busy outpatient clinic’ in mind. It is deliberately broad in scope and employs a visual analogue scale. It has been validated as a measure for daily life burden in a large multicentre study (17). Disability as measured by the IBD disk has been shown to correlate well with the IBD-DI, as well as with C-reactive protein (CRP) and faecal calprotectin (FCP) (18). More recently, albeit on a smaller scale, multiple disk domains have been shown to correlate with disease activity in ileal CD, as measured by bowel wall thickness on abdominal ultrasound (19). Though it is reasonable to extrapolate these findings to disease onset, the utility of the IBD disk as a measure specifically of psychological disease burden, has not been validated. The HADS score is a long-established validated measure to screen for major depression and anxiety amongst individuals with physical health problems (20, 21). As such, it represents a suitable comparator for other scoring systems.

The primary objective for this work was to determine if disease associated disability at first presentation, as determined by the IBD disk, could be used to predict outcomes during the first year of treatment. This included the need for treatment escalation, the presence of persistent disease activity and the need for inpatient admission. A secondary aim, carried out as a sub-study, was to determine if the IBD disk could be utilised at first presentation to screen for clinically significant mental health symptoms.

Between February 2021 and June 2024, patients were triaged to a dedicated rapid access clinic for suspected IBD. Triage was based on having symptoms broadly compatible with IBD and an elevated (no mandated threshold) faecal calprotectin (FCP) level. The collection of the established and validated clinical indices and patient reported outcome measures (PROM) employed in this study took place as part of a wider prospective observational cohort study to which patients provided informed consent (IRAS 287279, approved by Bloomsbury Research Ethics Committee [REC reference 21/PR/0515]). During the first outpatient appointment, prior to the establishment of a diagnosis, patients were asked to complete an IBD disk score. This was completed in the clinic room, following discussion regarding the suspected diagnosis and investigative plan. A subgroup of patients were also asked to complete a simultaneous HADS score. Traditional clinical and biochemical indices, symptom duration and longitudinal treatment outcomes were collected prospectively. Repeat scores were taken at follow up attendances. Treatment was not protocolised for this study, but all patients were managed by the same multi-disciplinary group of IBD clinicians within a single hospital and treatment commenced on the day of colonoscopy where endoscopic findings were supportive. Pre-existing mental health diagnoses were obtained from clinical history taking and coded primary care diagnoses. IBD disk scores from patients in whom IBD was subsequently excluded have been omitted from the analysis.

At each follow-up visit, current treatments were recorded and an assessment of disease activity undertaken. Diagnoses of IBD were established in line with European Crohn’s and Colitis Organisation guidelines (22). The criteria utilised to define ‘Inactive’ disease are shown in Supplementary Appendix 2 . A hierarchy of importance was employed in decision making. Where clinical indices alone were available these were used. If both CRP and FCP were available, these were preferred over clinical symptoms. If an endoscopic or radiological assessment had been performed, it superseded both clinical and biochemical indices.

Presented statistical analyses were primarily undertaken in JASP version 0.183 (23). Non-parametric tests were undertaken for all IBD disk related analyses as both the total IBD disk score (Shapiro-Wilk W 0.981 p=0.012) and each individual domain score did not follow a normal distribution. The Mann-Whitney U (U) test has been utilised for grouped differences in continuous variables. A chi-squared (X²) test is employed to determine proportional difference in categorical variables. For receiver operating characteristic (ROC) curves and logistic regression modelling, Jamovi version 2.6.26 was utilised alongside a dedicated R package and integrated modules (24–27).

IBD disk scores were collected from 188 patients (97 CD, 91 UC) at their first outpatient appointment, prior to diagnosis. 95 (45 CD, 50 UC) completed a simultaneous HADS score. 128 patients seen via this pathway completed an IBD disk and subsequently had IBD excluded. Follow up IBD disk scores were collected post treatment (median interval 117 days) from 82 patients, of whom 37 completed a paired post-treatment HADS score. The cohorts contributing to each analysis are shown in Table 1 .

The CD cohort was characterised by a significantly higher proportion of pre-existing MHD than those with UC (CD 22% [21/97], UC 11% [10/91], X² 3.88 p=0.049). Including those with multiple diagnoses; depression, anxiety disorder and mixed anxiety and depression were the most frequent MHD, accounting for 73% (27/37) of all pre-existing MHD. The most frequent anti-depressant prescribed was Sertraline (9 [41% of all antidepressants]) followed by Citalopram (4 [18%]).

The data presented signifies a novel evaluation of the IBD disk in a cohort of IBD patients assessed at the point of first presentation, prior to diagnosis. Whilst it is acknowledged that obtaining disk scores when diagnostic uncertainty remained may have heightened factors such as anxiety, our approach allowed for the quantification of baseline disease associated disability prior to any treatment being commenced and uninfluenced by any preconceptions about a given IBD type (28).

In IBD, disease associated disability has been shown to negatively impact upon health-related quality of life and socioeconomic productivity (29, 30). Disability determined by other PROM including the IBD-DI has been linked to adverse future treatment outcomes in both established and newly diagnosed IBD cohorts (31, 32). Moreover, normalised quality of life and the absence of disability now forms a long-term treatment target listed in STRIDE 2 (33). For the first time we have demonstrated that disability determined by the IBD disk can help identify both UC and CD patients at an increased risk adverse treatment outcomes in the first 12 months after diagnosis. The strength of association, particularly within UC where IBD disk scores also correlated with biochemical and endoscopic disease activity, highlights disability as an important marker of disease severity and treatment outcome. Whilst overall predictive capacity was not superior to traditional disease activity indices, our data supports the complementary role of the IBD disk as a PROM at IBD diagnosis across disease subtypes. Alongside these established markers, higher IBD disk scores at presentation should prompt consideration of a more aggressive therapeutic approach with earlier treatment escalation. In CD, whilst the cohort proceeding to surgical resection within 12 months was small, the disk was more able to identify these patients than the Harvey-Bradshaw index.

We have been able to demonstrate the high prevalence of significant psychological symptoms across IBD patients at presentation. This is particularly the case in patients subsequently diagnosed with CD, aligning with studies in those with established disease (1, 2). IBD disk scores, particularly in the ‘Emotions’ domain, strongly associated with HADS anxiety and depression scores. An ‘Emotions’ score of ≥7 identified all patients with HADS determined moderate or severe depressive symptoms at IBD presentation, but in many these symptoms appear driven by disease behaviour with longitudinal scores highlighting a greater psychological burden enduring in those with persisting disease activity at follow up appointments. Whilst the IBD standards emphasise the importance of holistic psychological assessment at IBD diagnosis, separating psychiatric illness from a psychological response to severe physical symptoms is not possible at first presentation, with prompt establishment of disease control through shared treatment plans that prioritise health related quality of life the key first step to be taken (12, 34). In those with persisting disability and psychological disease burden over and above other markers of disease activity, targeted psychological intervention might then be considered, as has been shown to be effective in those with established disease (10, 35, 36). Though the strength of specific IBD disk cut-offs for predicting moderate-severe depression and anxiety symptoms fell in our underpowered post-treatment cohort, IBD disk ‘Emotions’ scores and HADS scores remained strongly linked. In our work, patients identified with persisting moderate depression or anxiety symptoms did not receive targeted intervention, but this data will be used to support service expansion to facilitate such measures being available in future.

Though longer symptom durations prior to diagnosis associated with increased psychological symptoms, a pre-existing MHD was not associated with delayed diagnosis in our cohort. The framework of our study does not allow for the contribution of recall bias to these differences. However, patients with pre-existing MHDs were more likely to receive courses of oral corticosteroids, despite being no more likely to progress to advanced therapies or have ongoing disease activity after a year of treatment. Whilst this finding has been previously noted in active IBD and may relate to increased service utilisation amongst these patients, it remains noteworthy given the long-established potential for steroids to cause neuropsychiatric complications (37–40).

The difference in size of the cohorts for each analysis, in particular the longitudinal cohort being less than half the size of the overall cohort (and consequently underpowered for validation) is a key limitation. Furthermore, all scores are also derived from a single IBD centre and whilst patients were managed by a single MDT, post diagnosis treatment was not protocolised. Additionally, whilst data was gathered regarding pre-existing MHD, no data on other confounding socioeconomic factors were collected. Modelling relating to psychological burden was adjusted for baseline characteristics, but this was not the case in analyses relating to 12-month outcome, increasing the risk of uncontrolled confounding.

Utilisation of the IBD disk is not time consuming and whilst we opted to complete it in the clinic room given our pre-diagnosis setting, it can be completed electronically or in the waiting room prior to consultation (41). Implementing it as a PROM at IBD presentation can help in the early identification of those who are likely to require more aggressive IBD treatment to obtain disease control. Moreover, it can reliably detect clinically relevant depressive symptoms. Whilst these may represent a reaction to physical symptoms at IBD presentation, use of the IBD disk during longitudinal follow-up may help identify those with a persisting psychological symptom burden beyond that attributable to disease activity alone.