Treatment response and several patient-reported outcomes are early determinants of future self-efficacy in rheumatoid arthritis

Background: Self-efficacy, or patients' confidence in their ability to control disease and its consequences, was recently prioritised in EULAR recommendations for inflammatory arthritis self-management strategies. However, it remains unclear which factors influence self-efficacy in early rheumatoid arthritis (RA). Methods: Data were analysed from the 2-year RCT Care in early RA (CareRA), which studied remission-induction treatment regimens for early RA. Participants completed the Arthritis Self-Efficacy Scale (ASES), Short-Form 36 (SF-36), Revised Illness Perception Questionnaire (IPQ-R), Utrecht Coping List (UCL), RAQoL and Health Assessment Questionnaire (HAQ). Depending on time to first remission (DAS28-CRP < 2.6) and persistence of remission, treatment response was defined as persistent response, secondary failure, delayed response, late response or non-response. The association between ASES scores and clinical/psychosocial factors was explored with Spearman correlation and multivariate linear mixed models. Baseline predictors of week 104 ASES were identified with exploratory linear regression followed by multiple regression of significant predictors adjusted for DAS28-CRP, HAQ, treatment arm, treatment response, cumulative CRP/SJC28 and demographic/serologic confounders. Results: All 379 patients had a recent diagnosis of RA and were DMARD-naive at study initiation. Most patients were women (69%) and RF/ACPA-positive (66%), and the mean (SD) age was 52 (13) years. For all tested outcome measures, better perceived health correlated with higher self-efficacy. While patient-reported factors (HAQ, SF-36, RAQoL, IPQ-R, pain, fatigue and patient's global assessment) showed moderate/strong correlations with ASES scores, correlations with physician-reported factors (physician's global assessment, SJC28),TJC28 and DAS28-CRP were weak. Only more favourable outcomes on patient-reported factors and DAS28-CRP were associated with higher ASES scores at each time point. An earlier, persistent treatment response predicted higher ASES scores at both weeks 52 and 104. Significant baseline predictors of week 104 ASES included HAQ; SF-36 mental component score, vitality, mental health and role emotional; IPQ-R illness coherence, treatment control, emotional representations and consequences; UCL Passive reacting; and the RAQoL. Conclusions: Patient-reported outcomes and treatment response were early determinants of long-term self-efficacy in an early RA trial. These results provide further relevance for the window of opportunity in an early treat-to-target strategy and could help to timely identify patients who might benefit from self-management interventions.