The impact of timely diagnosis and recurrent health seeking behaviour on the subsequent clinical course of Inflammatory Bowel Disease: A UK population based study

Date of Approval: 
2021-01-08 00:00:00
Lay Summary: 
Inflammatory Bowel Diseases (IBD) are chronic bowel conditions inclusive of Ulcerative colitis and Crohn's disease. They present with debilitating symptoms including diarrhoea, rectal bleeding and abdominal pain. Current treatments include steroids and immune-suppressant medications, which reduce bowel inflammation. However, about a third of patients require surgery for their disease at some point. Timely diagnosis in IBD is essential. Delays in diagnosis create uncertainty and worry about the cause of symptoms to the individual experiencing them and delays starting treatment, which may lead to complications. Some people with IBD are thought to be misdiagnosed with irritable bowel syndrome or experience unexplained gastrointestinal symptoms for a long period before diagnosis is reached. A study showed 14% of individuals diagnosed with IBD in the UK report such symptoms to their primary care physician 3 years before receiving a diagnosis. The consequences of time taken to diagnose IBD and recurrent presentations to healthcare have on the course of disease outcome remains uncertain. Some studies have found a link between the time from symptom onset to date of diagnosis and increased risks of disease complications, others have not. Most studies are from single centres outside of the UK. Using Clinical Practice Research Datalink, a nationally representative primary care research database, we aim to examine impact of time to diagnosis and recurrent health seeking behaviour on the clinical course of IBD.
Technical Summary: 
Timely diagnosis in inflammatory bowel disease (IBD) is essential to enable early treatment to alleviate symptoms and prevent disease progression. The impact of time interval from first health seeking behaviour to the point of IBD diagnosis on disease outcome in IBD is unclear. Some studies demonstrate a correlation between this time period and increased risk of intestinal surgery in Crohn’s disease (CD), others did not. The influences of this time period on the course ulcerative colitis (UC) are limited. The impact of this time period on the subsequent clinical course of IBD has not been examined on a population level in the United Kingdom. Furthermore what is less clear is the impact of recurrent health seeking behaviour with undiagnosed gastrointestinal symptoms on the subsequent clinical course of IBD. The study proposed is to examine the impact of time to diagnosis, defined as the time from first health seeking behaviour for unexplained gastrointestinal symptoms to IBD diagnosis date, on the subsequent clinical course of IBD. Our outcome measures are time to; first steroid use, steroid dependency, immunomodulator initiation, biologics initiation, hospital admission and surgical intervention. Analysis will be separate for CD and UC. We will also evaluate the impact of recurrent consultations with GI symptoms on the clinical course of IBD. We will identify index presentation to primary care with gastrointestinal symptoms in the three years prior to IBD diagnosis date. We will analyse time‐dependent survival outcomes using the Kaplan–Meier method and generate survival curves to estimate the cumulative probability of the primary outcomes by each exposure group (a) time to diagnosis and (b) recurrent consultations for gastrointestinal symptoms. The log‐rank test will be used to compare outcomes. We will use a Cox proportional hazard model to calculate hazards ratio for our primary outcomes.
Health Outcomes to be Measured: 
SECTION C Primary outcome measures The impact of time to diagnosis and recurrent health seeking behaviour on the clinical course of IBD (described in section K and M) will be explored by examining the following outcome measures. 1. Medical intervention (a) Steroid dependence: We will identify IBD patients who received prolonged courses of oral corticosteroids therapy. We have defined steroid dependence as either; i) the inability to stop steroids within 3 months after the initiation of treatment or ii) in cases where steroid treatment is re-instituted within 3 months of stopping a previous course. These definitions are in keeping with the European ECCO definition of steroid-dependency. [10] Early steroid use (within 90 days of diagnosis) - this has been shown to be an independent marker for severity with an increased need for earlier surgery in both UC and CD. [11,12] This proxy marker for disease severity has also been used in previous CPRD work on IBD outcomes. [13,14] I. We will evaluate and compare rates of steroid dependency between the defined exposure groups (b) Immunomodulator therapy initiation I. We will evaluate and compare rates of immunomodulator therapy initiation between the defined exposure groups We will use thiopurine (inclusive of immunomodulator therapy) exposure as a proxy marker for disease severity. In UC, exposure to thiopurines within 3 years of diagnosis is associated with increase in colectomy risk. [15] Similar patterns have been seen in CD. [16] Patients will be subdivided into either exposed or non-exposed for this variable, dependent on whether they have a prescription for these medication in the period of follow up. (c ) Biological therapy initiation I. We will evaluate and compare rates of biologics therapy initiation between the defined exposure groups We will also use biologic exposure as a proxy marker for disease severity. Patients will be subdivided into either exposed or non-exposed for this variable, dependent on whether they have a prescription for these medication in the period of follow up. 2. IBD related Hospital admission I. Using HES linked data we will calculate hospital admission rates following diagnosis. We will further subcategorise admissions into those which were elective or emergency. 3. Surgical intervention I. We will calculate rates of surgical intervention for IBD. CD surgery will be subcategorised as either major (intestinal) abdominal surgery or perianal surgery. Colectomy surgical procedures will be explored for UC
Application Number: 
20_000248
Collaborators: 

Sonia Saxena - Chief Investigator - Imperial College London
Nishani Jayasooriya - Corresponding Applicant - St George's, University of London
Alex Bottle - Collaborator - Imperial College London

Linkages: 
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Practice Level Index of Multiple Deprivation