Behçet's Disease Overall Damage Index

DESIGN OF THE STUDY

This observational multicenter study will be developed in 2 subsequent phases, in their turn, consisting of different steps (figure 1):

1. BODI Development 1.1) Development of a preliminary version of BODI (p-BODI) based on literature review and inspired by pre-existing tools according to OMERACT guidelines; 1.2) Delphi process aimed to review and improve the p-BODI.
2. Validation of the new tool according to OMERACT filter 2.0 2.1) Application of the BODI on a multicentre cohort of BD patients. 2.2) Assessment of face, content, construct and criterion validity, and sensitivity to change.

2.3) Assessment of reliability based on scoring exercise on clinical vignettes.

All subjects involved in the study are experts in the management of BD. They will work in two different groups having different roles:

• Work and facilitator group (WFG): consisting of 3 Physician from the Coordinating Centre. The tasks of the WFG are: (a) to review literature and develop a preliminary version of BODI (p-BODI), including: definition of damage, rules for scoring and identification of potential items of damage; (b) to coordinate the Delphi process for editing and optimizing p-BODI; (c) to analyse data from the validation process.
• Expert Panel (EP): consisting of multidisciplinary experts in Behçet's disease and Patient's delegate from different countries. EP members will be selected and invited to participate in the study by the WFG. The tasks of the EP are to express their opinion on p-BODI and propose editing for its improvement within a Delphi process;
• Clinicians group (CG): consisting of one Clinician from each Centre, with expertise in Behçet's disease but not involved in the Delphi process. The tasks of this group are: (a) to apply the BODI on 30 consecutive patients with BD (10 for Neurologist and Ophthalmologist) from each Centre involved in the study; (b) to score a set of clinical vignettes using the BODI, to assess reliability of the instrument; (c) to express an overall judgement (in an online survey) regarding to general credibility, comprehensiveness and feasibility of the BODI.

The member of EP and GC, involved in the study so far, with respective affiliations, are reported below:

EP members - CG members - Affiliations.

• Prof. Alessandro Mathieu - Dr. Matteo Piga - Chair and Unit of Rheumatology, University and AOU of Cagliari, Monserrato - Italy;
• Prof. Govoni Marcello - Dr. Lo Monaco Andrea - Department of Medical Sciences, Section of Rheumatology, AOU S. Anna, the University of Ferrara - Italy;
• Prof. Iannone Florenzo - Dr. Lopalco Giuseppe - Department of Emergency and Organ Transplantation, Rheumatology Unit, University of Bari - Italy;
• Prof. Neri Piergiorgio - Dr. Pirani Vittorio - Ocular Immunology Service, the Eye Clinic, Polytechnic University of Marche- Ancona - Italy;
• Prof. Cantarini Luca - Dr. Orlando Ida - Rheumatology Unit, University of Siena. Policlinico Le Scotte;
• Prof. Martins Silva Ana - Dr. Santos Ernestina - Neurology Department, Hospital Santo António, Centro Hospitalar do Porto - Portugal;
• Prof. Vasconcelos Carlos - Prof Correia Joao; Dr. Raquel Faria - Clinical Immunology Unit, Centro Hospitalar do Porto and UMIB, ICBAS, Universidade do Porto, Portugal;
• Prof. Cervera Ricard - Dr. Espinosa Gerard, Dr. Ignasi Rodriguez - Department of Autoimmune Diseases, Hospital Clinic, IDIBAPS, University of Barcelona - Spain;
• Prof. Bertsias George - Dr. Nikos Kougkas - Department of Rheumatology, Clinical Immunology and Allergy, University of Crete School of Medicine - Grece;
• Patient Association delegates.

OPERATIVE STUDY DEVELOPMENT AND DATA ANALYSIS

1. BODI Development 1.1) Development of preliminary version of BODI (p-BODI) based on literature review and inspired to pre-existing tools A preliminary list of damaged items with respective definition will be generated after reviewing the literature and preexisting damage indices (e.g VDI for vasculitides and SLICC damage index (SDI) for systemic lupus erythematosus). 1.2) Reviewing and editing of the p-BODI by Delphi process A multi-step Delphi process, based on reaching experts consensus, will be performed in order to review and refine the p-BODI and the respective glossary. Through an online survey, each member EP will be asked to rate: (a) how strongly they agree with the general rules for damage scoring; (b) how strongly they agree to include the listed damage items within the BODI (e.g. to which extent do you agree that "Cataract" should be included in the BODI?); (c) how strongly they agree with the provided item definition. Ratings will be scored on a 5-point Likert scale (5 = strongly agree; 4 = agree; 3 = unsure; 2 = disagree; 1 = strongly disagree). The EP members will be asked to give considerations and arguments to support their opinion, especially if they score less than 4. They will also be given the opportunity to suggest alternative wordings, to suggest additional items, or to make any other comment. In the reviewed p-BODI for the subsequent Delphi round, each issue (scoring rules, damage items, item definitions) if: (a) ≥80% of EP members scores ≥4: will be included; (b) 50-79% of EP members scores ≥4: will be included in the second questionnaire after reviewing according to comments and suggestions provided in the previous round (c) <50% of EP members scores: will be excluded. The WFG has the right to make alternative decisions after reviewing the EP responses (e.g propose to change an item definition according to a suggestion from the expert panel, even if the item was scored >4 by ≥80% of EP). An anonymous feedback report will be provided with the second questionnaire, in order to offer the opportunity to reconsider and, if appropriate, to change a previous opinion in light of the anonymous responses and considerations of the other EP members. The panel members will again be asked to give their opinion on each issue in subsequent rounds until complete consensus will be achieved (all issues scored ≥4 by ≥80% of EP) and the definitive version of BODI will be developed.
2. Validation of BODI 2.1) Application of BODI on a multi-center cohort of BD patients Each CG member will be asked to apply the reviewed version of BODI on a cohort of 30 BD patients consecutively assessed in their own center. For Neurologist and Ophthalmologist, the data from 10 consecutive patients will be enough. This is because Neurologist and Ophthalmologist usually take care of the most severe neuropsychiatric and ocular cases, which might bias the real prevalence of these manifestations in the study cohort.

A multicenter cohort consisting of 200-250 BD patients will be thus set up. Patients will undergo a routine clinical assessment, as scheduled in their follow-up program. No further clinical, laboratory or instrumental investigations will be performed in addition to those provided according to the good clinical practice. Any possible drug administration will be evaluated independently from the study and according to the good clinical practice.

Inclusion criteria will be: a) BD diagnosis according to ICBD criteria, b) disease duration ≥12 months), c) age at enrolment ≥ 18 years. For each recruited patient data will be anonymously collected at the enrolment visit (T0). For patients with more than 5 years of follow-up at the enrollment time, investigators will be further asked to perform a 5 years retrospective BODI (Tr5). The following data will be collected (see the attached Case Report Form):

• T1: demographics and medical history (gender; birthday, onset, diagnosis and enrolment date; country of origin; ethnicity; comorbidity; smoker status); cumulative clinical manifestation from onset to enrolment (oral aphthosis, genital aphthosis, skin lesions, ocular involvement, neurologic lesions, vascular lesions, pathergy test, arthritis, gastrointestinal manifestations); active clinical manifestations the enrolment visit; ongoing and past therapy; BD current activity form (BDCAF); BODI; VDI; Physician global assessment (PGA)on visual analogic scale; Patients global assessment (PtGA) on visual analogic scale; Short Form 36 (SF-36) Health Survey,
• Tr5 (retrospectively): BODI

2.2) Assessment of face, content, construct, criterion validity and sensitivity to change Validation will be performed according to the principles of OMERACT Filter 2.0, thus according to the criteria of (A) truth, (B) discrimination and (C) feasibility will be assessed.

A.1. Face validity. Since BODI will be developed through literature review and consensus between an expert in BD management, face validity criteria should be automatically met. However, face validity will be further tested by asking each CG member to complete a questionnaire investigating the suitability of the tool.

A.2. Content validity. To assure content validity, the presence of damage identified by CG but not classifiable in BODI will be assessed. Content validity will be further tested by asking each CG member to complete a questionnaire relating to the comprehensiveness of the instrument.

A.3. Construct validity. It will be assessed analyzing the agreement between results of damage assessment on our validation cohort by BODI and other damage assessment tool. Since there are no other specific and validate instruments for damage amassment in Behçet disease, SLICC and VDI will be used as a surrogate. Correlation between the different indices results will be evaluated by Pearson's or Spearman's correlation coefficients.

A.4. Criterion Validity. Criterion validity represents the correlation with results of measurement with the new tool and other long-term outcomes related to the phenomenon we are measuring. Since we expect that damage correlates with quality of life, disability and mortality, a correlation between BODI score and SF-36, HAQ and in the future also with mortality rate will be assessed.

B.1. Reliability. Clinical vignettes will be created from real cases to test the reliability of the BODI. Thus, inter-observer agreement (kappa) will be assessed by asking a physician from every center to assess the same clinical vignettes.

B.2. Sensitivity to change. The ability of the BODI to record the accumulation of damage over time will be examined by determining the change in BODI score over 5 years of follow-up on the validation cohort. The average increase of BODI score from T1 to T2 will be calculated and will be tested by the Mann-Whitney test.

1. Feasibility. To test feasibility each CG member will be asked to complete a questionnaire relating to the ease of use, consumption of time and interpretability of the instrument.