Background The SLICC, ACR and LFA embarked on a data- and expert-driven project to develop a revised systemic lupus erythematosus (SLE) organ damage index (SDI). The methodological approach includes 5 phases: updating the construct of damage (I), item generation (II), item reduction (III), item weighting and threshold determination (IV), and the assessment of validation and reliability (V). In phase I, a consensus statement was developed to define the construct of damage in SLE1. In the Item Generation phase, we aimed to develop and agree on a candidate list of items that reflect the construct of damage in SLE and are appropriate to be included in a new damage index including consideration of relevant items from adult, paediatric and young adult SLE. In this analysis, we compare the two approaches to initial item generation that were employed in a parallel process, namely a literature review and a Delphi exercise.
Methods Item generation included a literature review and 3-part Delphi exercise. A group of lupus experts conducted a literature review to identify items that reflect the construct of damage in SLE and grouped the items into organ domains. Each domain was reviewed by paediatric rheumatologists.
Snow-ball sampling was used among SLICC members, asking them to nominate 3-4 SLE experts considering a range of clinical expertise, equality, diversity and inclusiveness factors, and the global nature of SLE research. The LFA, Lupus UK, Lupus Europe and Lupus Canada were also asked to nominate 4-6 patient/carer representatives to participate in the Delphi exercise. Participants were asked to nominate items that should be included in a revised damage index based on the updated construct definition1 using a free-text option in Delphi exercise.
Results We established a group of 146 individuals (mean age 50.6 ranging from 28 to 79 years; 60.3% females; 58.9% white; clinical experience from 1 to 51 years) from 35 countries, broadly representative of the lupus research and patient community. There were 135 medical doctors, 2 allied health professionals and 9 patients. Of 135 medical doctors, 120 were rheumatologists, 7 internists, 5 nephrologists, 2 dermatologists, and 1 immunologist. The response rate after the first round Delphi exercise was 97.9%.
All items in the original SDI were nominated in both processes. Item generation yielded approximately 2,600 items. After rationalising for repetition, redundancy, and harmonisation of synonyms, 220 unique items were identified across 14 organ systems. The literature review proposed 4 (1.8%) unique items, 103 (46.8%) unique items were from the Delphi only and 113 (51.4%) items appeared in both exercises (figure 1).
Conclusion Using a combined data-driven and expert/patient-based approach, items and domains that comprise damage in SLE have been expanded. Just over half of all items were nominated by both approaches. However, the Delphi exercise which included a wide and diverse group of contributors, provided a large number of unique items for further consideration. Our data confirms the value of large group exercises early in such a process to maximise the scope of new items to consider for a revised index.
Johnson, S. R et al. Evaluating the construct of damage in SLE. Arthritis Care Res. 2021.
Lay Summary The SLICC/ACR Damage Index (SDI) (published in 1996) is widely used in clinical studies and trials to measure the long-term complications that can occur in lupus patients, such as cataracts, fractures, and kidney failure. Higher scores are associated with poorer quality of life, as reported by patients. A number of drawbacks have also been found with the SDI. We need to better understand and measure the impact of these complications from a patient and doctor’s perspective to get a much deeper understanding of how SLE affects people. We used two methods to generate new items to include in an updated SDI. First, we used the medical literature to identify possible complications of lupus. Then, we asked a large group of lupus experts and patients to nominate complications. The process generated approximately 2,600 items. After removing redundant suggestions, 220 unique items were identified. The literature review proposed 4 (1.8%) unique items, 103 (46.8%) unique items were from the large group only and 113 new (51.4%) items appeared in both exercises. Our data shows the value of large group exercises that include patient representatives, to maximise the scope of new items to consider for a revised index.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/ .
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.