Data‐Driven Patient Clustering and Differential Clinical Outcomes in the Brigham and Women’s Rheumatoid Arthritis Sequential Study Registry

Curtis, Jeffrey R.; Weinblatt, Michael E.; Bykerk, Vivian P.; Fürst, Daniel E.; Fiore, Stefano; John, Gregory St; Kimura, Takeshi; Zheng, Shen; Bingham, Clifton O.; Wright, Grace; Bergman, Martin; Nola, Kamala; Charles‐Schoeman, Christina; Shadick, Nancy A.

doi:10.1002/acr.24471

Cited by 13 publications

(14 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This phenotype predicted poorer treatment response and a higher likelihood of adverse events in a randomized controlled trial of certolizumab pegol (27). Curtis et al also used machine learning approaches to phenotype patients with RA in the Brigham and Women's Rheumatoid Arthritis Sequential Study (28). Principal component analysis was performed on a large number of variables from patient questionnaires, physician reports, laboratory tests, and radiographs.…”

Section: Discussionmentioning

confidence: 99%

“…Subsequently, patients were clustered based on these principal components. Despite a large number of RA‐specific variables, multimorbidity burden was a key construct for differentiating clusters of RA patients (28). Our study is distinct by comprehensively characterizing the interrelatedness of chronic conditions that define unique patterns of multimorbidity in RA and demonstrating their overrepresentation among patients with RA.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Identification of Multimorbidity Patterns in Rheumatoid Arthritis Through Machine Learning

England

Yang

Roul

et al. 2022

Arthritis Care & Research

Self Cite

View full text Add to dashboard Cite

Objective. Recognizing that the interrelationships between chronic conditions that complicate rheumatoid arthritis (RA) are poorly understood, we aimed to identify patterns of multimorbidity and to define their prevalence in RA through machine learning.Methods. We constructed RA and age-and sex-matched (1:1) non-RA cohorts within a large commercial insurance database (MarketScan) and the Veterans Health Administration (VHA). Chronic conditions (n = 44) were identified from diagnosis codes from outpatient and inpatient encounters. Exploratory factor analysis was performed separately in both databases, stratified by RA diagnosis and sex, to identify multimorbidity patterns. The association of RA with different multimorbidity patterns was determined using conditional logistic regression.Results. We studied 226,850 patients in MarketScan (76% female) and 120,780 patients in the VHA (89% male). The primary multimorbidity patterns identified were characterized by the presence of cardiopulmonary, cardiometabolic, and mental health and chronic pain disorders. Multimorbidity patterns were similar between RA and non-RA patients, female and male patients, and patients in MarketScan and the VHA. RA patients had higher odds of each multimorbidity pattern (odds ratios [ORs] 1.17-2.96), with mental health and chronic pain disorders being the multimorbidity pattern most strongly associated with RA (ORs 2.07-2.96).Conclusion. Cardiopulmonary, cardiometabolic, and mental health and chronic pain disorders represent predominant multimorbidity patterns, each of which is overrepresented in RA. The identification of multimorbidity patterns occurring more frequently in RA is an important first step in progressing toward a holistic approach to RA management and warrants assessment of their clinical and predictive utility.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Identification of Multimorbidity Patterns in Rheumatoid Arthritis Through Machine Learning

England

Yang

Roul

et al. 2022

Arthritis Care & Research

Self Cite

View full text Add to dashboard Cite

show abstract

“…Clustering of comorbidities in several populations revealed similar results, with clusters for CVD, mental/behavioral disorders, and musculoskeletal disorders reported commonly (15–19). Others have utilized clustering methods in patients with RA to define patient subgroups based on clinical features and disease activity measures (20,21). Clustering of comorbidities as well as clustering of patients based on their comorbidities has been reported in other rheumatic diseases (e.g., gout) to identify clinical phenotypes (22,23).…”

Section: Discussionmentioning

confidence: 99%

Using Unsupervised Machine Learning Methods to Cluster Comorbidities in a Population‐Based Cohort of Patients With Rheumatoid Arthritis

Crowson

Gunderson

Davis

et al. 2022

Arthritis Care & Research

View full text Add to dashboard Cite

Objective To identify clusters of comorbidities in patients with rheumatoid arthritis (RA) using 4 methods and to compare to patients without RA. Methods In this retrospective, population‐based study, residents of 8 Minnesota counties with prevalent RA as of January 1, 2015 were identified. Age‐, sex‐, and county‐matched non‐RA comparators were selected from the same underlying population. Diagnostic codes were retrieved for 5 years before January 1, 2015. Using 2 codes ≥30 days apart, 44 previously defined morbidities and 11 nonoverlapping chronic disease categories based on Clinical Classifications Software were defined. Unsupervised machine learning methods of interest included hierarchical clustering, factor analysis, K‐means clustering, and network analysis. Results Two groups of 1,643 patients with and without RA (72% female; mean age 63.1 years in both groups) were studied. Clustering of comorbidities revealed strong associations among mental/behavioral comorbidities and among cardiovascular risk factors and diseases. The clusters were associated with age and sex. Differences between the 4 clustering methods were driven by comorbidities that are rare and those that were weakly associated with other comorbidities. Common comorbidities tended to group together consistently across approaches. The instability of clusters when using different random seeds or bootstrap sampling impugns the usefulness and reliability of these methods. Clusters of common comorbidities between RA and non‐RA cohorts were similar. Conclusion Despite the higher comorbidity burden in patients with RA compared to the general population, clustering comorbidities did not identify substantial differences in comorbidity patterns between the RA and non‐RA cohorts. The instability of clustering methods suggests caution when interpreting clustering using 1 method.

show abstract

“…We used LCGM, as the most common developed approach, to show patient’s disease course and heterogeneity between subjects over time. This approach has been previously used by other studies in the field of rheumatic disease to identify disease course [ 1 , 3 , 10 , 17 ] and swollen joint count trajectories in juvenile inflammatory arthritis [ 18 ]. Using clinical data from two Canadian pediatric rheumatology centers they identified five trajectory groups with significant differences in the international League of Associations for Rheumatology categorizations (ILAR) [ 18 ].…”

Section: Discussionmentioning

confidence: 99%

Disease activity trajectories for early and established rheumatoid arthritis: Real-world data from a rheumatoid arthritis cohort

Movahedi

Cesta²,

Li³

et al. 2022

PLoS ONE

View full text Add to dashboard Cite

Objectives Disease activity status described at fixed time points does not accurately reflect disease course in chronic and relapsing diseases such as rheumatoid arthritis (RA). We described longitudinal disease activity trajectories in early and established RA. Methods Patients with available 28-Joint Disease Activity Score-erythrocyte sedimentation rate (DAS28-ESR) and Clinical Disease Activity Index (CDAI) over two years were included. Using latent growth curve modelling (LCGM), subgroups of patients following distinct patterns were identified. Results 1920 patients were included with 34.4% in early RA (< 2 years’ disease duration). Three subgroups were identified using DAS28-ESR in early RA: 1) low disease activity to remission (LDA-REM: 19.1%); 2) moderate disease to remission (MD-REM: 54%); 3) high to moderate disease (HD-MD: 26.9%). The HD-MD group had a significantly higher number of comorbidities, biologic and steroid use and lower post-secondary education. Using CDAI, we identified seven subgroups with only 1.9% remission in early RA. In established RA, seven subgroups were identified using either DAS28-ESR or CDAI. Using DAS28-ESR 27.8% with HD showed improvement in disease status (14.2% HD-REM, 10.3% HD-LDA and 3.3% HD-MD) while using CDAI 17.9% showed improvement. Conclusion Disease course was different in early and established RA. Only 14.2% of established RA reached DAS28-ESR remission compared to 73.1% of early RA. Using CDAI only 1.9% of early RA and none of the established RA achieved remission, likely reflecting the impact of the patient global assessment on this score. Findings also illustrate the impact of sociodemographic characteristics and early treatment on disease course.

show abstract

Data‐Driven Patient Clustering and Differential Clinical Outcomes in the Brigham and Women’s Rheumatoid Arthritis Sequential Study Registry

Cited by 13 publications

References 28 publications

Identification of Multimorbidity Patterns in Rheumatoid Arthritis Through Machine Learning

Identification of Multimorbidity Patterns in Rheumatoid Arthritis Through Machine Learning

Using Unsupervised Machine Learning Methods to Cluster Comorbidities in a Population‐Based Cohort of Patients With Rheumatoid Arthritis

Disease activity trajectories for early and established rheumatoid arthritis: Real-world data from a rheumatoid arthritis cohort

Contact Info

Product

Resources

About