Search in the Abstract Database

Abstracts Search 2017

* = Presenting author

P655 A study to determine factors affecting clinical decision making in outpatients with inflammatory bowel disease

Derwa Y.1, Gracie D.J.*1,2, Williams C.J.1, Sood R.1,2, Mumtaz S.1, Bholah M.H.1, Hamlin P.J.1, Ford A.C.1,2

1St. James's University Hospital, Leeds Gastroenterology Institute, Leeds, United Kingdom 2University of Leeds, LIBACS, Leeds, United Kingdom


Evaluation of patient-reported symptoms forms the mainstay of disease activity assessment in patients with inflammatory bowel disease (IBD). Despite this, the correlation between symptom reporting and the presence of mucosal inflammation is poor, particularly in Crohn's disease (CD). [1] We conducted a cross-sectional study to determine factors affecting clinical decision making in IBD.


Validated questionnaires were used to assess clinical disease activity, self-reported flare, the presence of irritable bowel syndrome (IBS) symptoms and anxiety, depression and somatisation in a cohort of 274 patients with IBD. Patients who had undergone investigations in the preceding 90 days were excluded. Clinicians were blinded to faecal calprotectin (FC), with a cut off of 250 μg/g used to define active mucosal inflammation. Logistic regression analysis was performed to determine the association between these factors and clinical decision making (investigation requesting or escalation of medical treatment). Results were reported as odds ratios (OR) with 95% confidence intervals (CI).


18 (50%) of 36 CD patients, and 10 (27.7%) of 36 ulcerative colitis (UC) patients referred for investigations had evidence of mucosal inflammation defined by FC. The proportion of patients with mucosal inflammation who were neither investigated nor received escalation of medical treatment was 33.7% in CD and 39.4% in UC. The results of logistic regression analyses for factors associated with investigation requesting and escalation of medical treatment are displayed in Tables 1 and 2.

Table 1. Relationship between clinician investigation requesting and disease characteristics in CD and UC after logistic regression

CD and investigation requestingUC and investigation requesting
OR (95% CI)OR (95% CI)
Rome III IBS criteria fulfilled0.74 (0.25–2.15)1.07 (0.30–3.81)
Self-reported flare5.93 (1.93–18.21)7.29 (2.00–26.54)
Total HBI ≥51.43 (0.43–4.80)N/A
Total SCCAI ≥5N/A1.23 (0.37–4.15)
FC ≥250 μg/g2.09 (0.77–5.71)0.19 (0.05–0.64)
Anxiety (per 1-point change on HADS anxiety score)0.91 (0.78–1.07)0.81 (0.66–0.98)
Depression (per 1-point change on HADS depression score)0.98 (0.83–1.17)1.19 (0.96–1.47)
Somatisation (per 1-point change on PHQ-15 score)1.02 (0.86–1.20)1.10 (0.95–1.26)

Table 2. Relationship between clinician decision to escalate treatment and disease characteristics in CD and UC after logistic regression

CD and escalationUC and escalation
OR (95% CI)OR (95% CI)
Rome III IBS criteria fulfilled0.87 (0.21–3.69)1.24 (0.31–5.03)
Self-reported flare5.62 (1.24–25.5)2.39 (0.64–8.87)
Total HBI ≥54.73 (0.89–25.06)N/A
Total SCCAI ≥5N/A10.36 (2.47–43.5)
FC ≥250 μg/g2.02 (0.52–7.94)4.26 (1.28–14.2)
Anxiety (per 1-point change on HADS anxiety score)0.95 (0.77–1.17)1.04 (0.88–1.25)
Depression (per 1-point change on HADS depression score)1.07 (0.84–1.36)0.91 (0.73–1.14)
Somatization (per 1-point change on PHQ-15 score)0.94 (0.73–1.20)1.01 (0.85–1.21)


Bar self-reported flare, factors influencing clinical decision making are uncertain in CD. Clinicians treat, rather than investigate active UC, presumably because of the superior correlation between symptoms and inflammation in these patients. [1] At the time of investigation requesting, 50% of CD and 72.3% of UC patients had no evidence of mucosal inflammation, suggesting that these investigations could have been avoided. One third of patients with FC ≥250μg/g received no intervention and were, potentially, managed inappropriately. The introduction of routine point-of-care FC testing may reduce unnecessary investigations and improve the appropriateness of resource allocation, particularly in CD.


[1] Gracie DJ et al. (2016), Poor Correlation Between Clinical Disease Activity and Mucosal Inflammation, and the Role of Psychological Comorbidity, in Inflammatory Bowel Disease, Am J Gastroenterol, 541–51., 111(4)