Predicting abatacept retention using machine learning

Table 1 Baseline characteristics of the combined ACTION/ASCORE study population

Variable	Overall study population (N = 5320)
Age, years	57.7 (12.7), n = 5315
Female, %	79.0, n = 4202
BMI, kg/m²	27.1 (5.7), n = 5021
RA duration, years	11.1 (9.4), n = 5283
TJC28	9.3 (6.9), n = 5065
SJC28	6.5 (5.3), n = 5084
DAS28 (CRP)	4.8 (1.1), n = 4099
DAS28 (ESR)	5.3 (1.3), n = 3996
Physician Global Assessment, VAS	58.1 (19.3), n = 4536
Patient Global Assessment, VAS	61.7 (21.5), n = 4465
Patient pain, VAS	61.7 (22.0), n = 4458
HAQ-DI	1.4 (0.7), n = 3891

All data are mean (SD) unless specified
BMI body mass index, CRP C-reactive protein, DAS28 Disease Activity Score in 28 joints, ESR erythrocyte sedimentation rate, HAQ-DI Health Assessment Questionnaire-disability index, RA rheumatoid arthritis, SD standard deviation, SJC28 swollen joint count in 28 joints, TJC28 tender joint count in 28 joints, VAS visual analog scale

ISSN: 1478-6362