Chronic Inflammatory Disease, Lifestyle and Treatment Response

Chronic inflammatory diseases (CID), including inflammatory bowel diseases (IBD) (of which Crohn's disease \[CD\] and ulcerative colitis \[UC\] are the two most prevailed entities), rheumatic conditions (rheumatoid arthritis \[RA\], axial spondyloarthropathy \[axSpA\], psoriatic arthritis \[PsA\]), skin diseases (psoriasis \[PsO\], hidradenitis suppurativa \[HS\]), and eye disease (non-infectious uveitis \[NiU\]), are diseases of the immune system that are managed with biological agents targeting the pro-inflammatory cytokine tumour necrosis factor-α (TNF), i.e. TNF inhibitors. Design In this prospective cohort study disease activity prior to and after (14-16 weeks) initiation of biologic treatment will be assessed. The endpoint is the treatment outcome defined as A: Responder according to the specific criteria described below (incl. drug-continuation) or B: Non-responder (incl. drug-discontinuation due to unacceptable side effects). Whether a patient will discontinue therapy is assumed to be based on a certain degree of shared decision making between the patient and physician supported by principles from national guidelines for each CID as recommended in the respective national guidelines and laboratory data. Setting All patients assigned for initiation of biologic treatment at 1) Department of Gastroenterology and Hepatology, Aalborg University Hospital; 2) Department of Hepatology and Gastroenterology, Aarhus University Hospital; 3) Diagnostic Centre, Silkeborg Regional Hospital; 4) Department of Internal Medicine, Herning Regional Hospital; 5) Department of Gastroenterology, Herlev Hospital; 6) Organ Centre, Hospital of Southern Jutland; 7) Department of Gastroenterology Hospital of South West Jutland; 8) Department of Medical Gastroenterology, Department of Rheumatology, Department of Dermatology and Allergy Centre, and Department of Ophthalmology, Odense University Hospital from 1st of April 2017 and until 31th of Marts 2019 or until a minimum of 100 patients with IBD, 100 patients with RA, and 120 patients with axSpA, PsA, PsO, HS and NiU are achieved. Clinical data consist of personal data, data on health and disease, lifestyle, laboratory measure, and disease activity scores including using patient-reported outcome measures (PROMs), clinical assessments, and laboratory data. Each participant will fill out validated questionnaires on disease activity, quality of life, and lifestyle using an electronic link. Studies have found electronic questionnaires to be equivalent to paper-based in relation to PROMs. Data management The electronic questionnaire is in Danish language and the participants will have access to the questionnaire by an electronic link sent to their personal, electronic mailbox. All data will be stored in a secure research storage facility. Information registered by clinicians and technicians will be transferred from paper format to electronic format using either double entry of data or automated forms processing. Statistical methods The investigators will use this rigorously designed, prospective cohort study to explore the ability to predict clinical response across the conditions included (Y=primary endpoint), and explore whether patients who are on a diet high in fibre AND low in red and processed meat (X=assessed at baseline) is an informative prognostic factor. Per default, the statistical model will include condition (any of the CID conditions included), and clinical centre (site #1 to #8) as fixed effects. Sample size considerations: It is a well-known difficulty for exploratory prognostic factor research studies like this, to formalize how many participants (i.e. with events) to include. In order to consider an adequate number of outcome events, the investigators apply "the rule of thumb" that dictates that 10 outcome events are needed for each independent variable (possible predictors); the investigators plan to enrol 320 patients in total, and anticipate that 50% of these will experience a clinical response during the 14-16 week period after therapy is initiated. With this in mind: Anticipating that at least 160 will achieve clinical responses (among the 320 patients), this study will have a reasonable power to explore the impact of as many as 16 independent (predictor) variables (including condition and clinical centre). Focusing on the contrast between groups, for a comparison of two independent binomial proportions (those with high fibre AND low meat intake vs other) using Pearson's Chi-square statistic with a Chi-square approximation with a two-sided significance level of 0.05 (P\<0.05), a total sample size of 318 assuming an allocation ratio of 1 to 2 has an approximate power of 0.924 (i.e. \>90% statistical power) when the proportions responding are 60% and 40%, respectively. All the statistical programming will be done in SAS (Statistical Analysis Software), STATA or R, transparently reporting the source code used to analyse the data. All computational details will be available in the pre-specified Statistical Analysis Plan (will be finalised before data collection is complete). Project organisation The project is organised with a Clinical Research Group and an Analytical Research Group. The clinical group includes specialists from the medical, gastroenterological, rheumatological, dermatological and ophthalmological departments that are sampling the cohort. The analytical group will perform the analyses on the biological material.