Uncovering the heterogeneity of care trajectories leading up to a rheumatoid arthritis diagnosis: a Swedish 10-year study.

Abstract

This study aimed to identify types of care trajectories leading to rheumatoid arthritis (RA) diagnosis, and to identify if, and when, such trajectories deviate from those in the general population. We included patients with a first-ever RA diagnosis between 2016 and 2020 in the Swedish nationwide multiregister system and matched general population controls. We performed group-based multitrajectory modelling on 9 healthcare components over the 10 years preceding the index date, grouping patients based on their pattern of healthcare use. We described these groups, or 'types of trajectories', and compared these to those of the matched population controls. Based on 6824 patients with seropositive RA, we describe 5 trajectories: (i) Healthy (60%), low healthcare use across all care components; (ii) High sick leave (18%), no musculoskeletal (MSK) or infection-related inpatient care but high sick leave; (iii) Chronic pain (10%), high use of analgesics with frequent MSK-related inpatient care; (iv) High infection (9%), frequent infection-related inpatient use; and (v) High burden (3%), with an overall high use of healthcare. Deviations from the 'expected' trajectories in the general population emerged as early as 10 (types 1, 3, 5), 3 (type 4), and 2 years (type 2) before the index date. Individuals with new-onset RA can be grouped into clusters, each with its own signature in terms of pre-RA healthcare resource utilisation and deviation from the expected years before the RA diagnosis. Our results open for identification of individuals at elevated risk, for potential preventive action, and may highlight aetiopathologically important events.