New recommendations on core datasets to be used in systemic lupus erythematosus care

Systemic lupus erythematosus (SLE) is a complex and potentially life-threatening autoimmune disease. Part of the complexity stems from how it can differ from person to person—giving rise to marked heterogeneity in not only manifestations, but also in disease course and treatment response.

To support better understanding of the disease, researchers have suggested using “big data.” Traditional data is often structured and stored in databases of tables—making it easy to query and run statistics. This is fine for relatively small volumes of data with predictable formats. With big data, massive and complex datasets can be utilized with advanced tools such as machine learning to uncover patterns and insights.

However, big data copes with massive datasets in structured, semi-structured, and unstructured formats, and this information is stored in different ways, such as in data lakes without predefined schemas. The analyses from such projects could have a number of impacts, such as helping to identify patient subgroups that might be suitable for targeted clinical trials.

Although there are many registries collecting data in SLE, these do not always use the same terms or measures, and this makes it hard to combine datasets to achieve big data. To enhance clinical and multi-center research outcomes, standardized documentation of patient- and disease-related features is important.

To address these issues, EULAR put together a taskforce to define a comprehensive core dataset of the essential elements necessary to ensure complete clinical care, as well as to facilitate scientific research for the benefit of people with SLE.

In total, 25 stakeholders from 14 different countries took part. A literature search was conducted to collect relevant information, resulting in a list of 99 items to consider. In an anonymous online survey, the expert panel rated the perceived importance of each of these, followed by a Delphi survey.

The new work, published in the August 2025 issue of the Annals of the Rheumatic Diseases, includes 73 items for a clinical core dataset, and an additional eight for research purposes.

The core clinical dataset is split into three overarching groups based on timing of data collection: first visit and on-demand, yearly, and regularly. The former includes general demographic items, plus disease history and serology, the second a yearly review of comorbidities and recording of disease damage and progression, and the latter regular review of laboratory parameters, outcomes, treatment, patient-reported outcomes, and disease activity. Within each topic there are specific suggested measures.

The additional eight items in the research extension cover fulfillment of classification criteria, hematological damage, vaccinations, achievement of low disease activity, drug adherence, the use of other medications, plus health-related quality of life and work productivity.

“Harnessing big data, especially through standardized datasets, will be pivotal in accelerating research and revealing new insights that can transform how we manage and treat challenging conditions,” said Dr. Johanna Mucke—lead author on the paper and researcher at Ruhr-University Bochum, Germany. “The development of this core dataset lays a crucial foundation for achieving that standardization.”

EULAR believes that this core dataset is feasible for assessment in clinical care—especially since many of the items do not require regular assessment but only yearly or one-off evaluation. The comparability that will result from standardized datasets will facilitate clinical benchmarking, leading to advancements in our understanding and treatment of SLE. Ultimately, this project aims to improve care and quality of life for people living with SLE.

Provided by
European Alliance of Associations for Rheumatology (EULAR)

Systemic lupus erythematosus (SLE) is a complex and potentially life-threatening autoimmune disease. Part of the complexity stems from how it can differ from person to person—giving rise to marked heterogeneity in not only manifestations, but also in disease course and treatment response.

To support better understanding of the disease, researchers have suggested using “big data.” Traditional data is often structured and stored in databases of tables—making it easy to query and run statistics. This is fine for relatively small volumes of data with predictable formats. With big data, massive and complex datasets can be utilized with advanced tools such as machine learning to uncover patterns and insights.

However, big data copes with massive datasets in structured, semi-structured, and unstructured formats, and this information is stored in different ways, such as in data lakes without predefined schemas. The analyses from such projects could have a number of impacts, such as helping to identify patient subgroups that might be suitable for targeted clinical trials.

Although there are many registries collecting data in SLE, these do not always use the same terms or measures, and this makes it hard to combine datasets to achieve big data. To enhance clinical and multi-center research outcomes, standardized documentation of patient- and disease-related features is important.

To address these issues, EULAR put together a taskforce to define a comprehensive core dataset of the essential elements necessary to ensure complete clinical care, as well as to facilitate scientific research for the benefit of people with SLE.

In total, 25 stakeholders from 14 different countries took part. A literature search was conducted to collect relevant information, resulting in a list of 99 items to consider. In an anonymous online survey, the expert panel rated the perceived importance of each of these, followed by a Delphi survey.

The new work, published in the August 2025 issue of the Annals of the Rheumatic Diseases, includes 73 items for a clinical core dataset, and an additional eight for research purposes.

The core clinical dataset is split into three overarching groups based on timing of data collection: first visit and on-demand, yearly, and regularly. The former includes general demographic items, plus disease history and serology, the second a yearly review of comorbidities and recording of disease damage and progression, and the latter regular review of laboratory parameters, outcomes, treatment, patient-reported outcomes, and disease activity. Within each topic there are specific suggested measures.

The additional eight items in the research extension cover fulfillment of classification criteria, hematological damage, vaccinations, achievement of low disease activity, drug adherence, the use of other medications, plus health-related quality of life and work productivity.

“Harnessing big data, especially through standardized datasets, will be pivotal in accelerating research and revealing new insights that can transform how we manage and treat challenging conditions,” said Dr. Johanna Mucke—lead author on the paper and researcher at Ruhr-University Bochum, Germany. “The development of this core dataset lays a crucial foundation for achieving that standardization.”

EULAR believes that this core dataset is feasible for assessment in clinical care—especially since many of the items do not require regular assessment but only yearly or one-off evaluation. The comparability that will result from standardized datasets will facilitate clinical benchmarking, leading to advancements in our understanding and treatment of SLE. Ultimately, this project aims to improve care and quality of life for people living with SLE.

Provided by
European Alliance of Associations for Rheumatology (EULAR)