Article Text

THU0602 Patient Global Assessment in Rheumatoid Arthritis Conveys A Variable Blend of Disease Activity and Disease Impact: A Cross-Sectional Study with 311 Patients
  1. R. Ferreira1,2,
  2. C. Duarte1,3,
  3. C. Silva3,
  4. G. Eugénio1,
  5. M. Ndosi4,
  6. L. Gossec5,
  7. J.A.P. da Silva1,3
  1. 1Rheumatology, Centro Hospitalar e Universitário de Coimbra, EPE
  2. 2Health Sciences Research Unit: Nursing (UICiSA:E)
  3. 3Faculdade de Medicina, Universidade de Coimbra, Coimbra, Portugal
  4. 4School of Healthcare, University of Leeds, Leeds, United Kingdom
  5. 5Paris 06 University, Paris, France


Background Patient global assessment (PGA) is a key outcome measure in rheumatoid arthritis (RA) and the sole patient-reported outcome included in measures currently used to guide treat-to-target (T2T) strategies. However, it is still not clear which concepts it conveys and how appropriate they are to guide therapy.

Objectives to explore the meaning of PGA in RA patients, namely the influence of disease activity, disease impact, comorbidities, and psychological aspects.

Methods This was an observational, cross-sectional study including consecutive RA patients (ACR/EULAR 2010 or ACR 1987 criteria) followed in a tertiary rheumatology outpatient department. Data collection included PGA (100mm VAS), pain and fatigue (0–10 NRS), Health Assessment Questionnaire (HAQ), Functional Assessment of Chronic Illness Therapy (FACIT), Hospital Anxiety and Depression Scale (HADS), Happiness Subjective Scale (HSS), and Ten Item Personality Inventory (TIPI). Demographic data, comorbidities, and disease activity [tender joint (TJC28) and swollen joint counts (SJC28), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP)] were also collected. Disease activity categories were defined as: 1) remission (ACR/EULAR 2011 boolean-based definition), 2) near-remission (only PGA>1), and 3) non-remission. Univariate (Pearson correlation, student's t test and One Way ANOVA) and multivariable analysis (linear regression, with stepwise methods and assessment of multicolinearity) were performed to explain PGA.

Results 311 patients were included (82% females, 60±12 years, 11±9 years of disease duration, 8±5 years of formal education and a mean DAS(28)CRP4v = 2.9±1.2). All comorbidities assessed were associated with statistically (p<0.05) higher PGA than RA alone (Graph 1).

PGA was statistically correlated with all predictor variables, except for the personality domain “agreeableness”. In multivariable analysis, PGA was explained (R2adj.=0.58) by pain (β=0.33), fatigue (β=0.23), function (β=0.21), anxiety (β=0.13), and SJC28 (β=0.09). The predictors of PGA were different between disease activity categories (Table 1).

Table 1.

Multivariable linear regression models (stepwise-backward method) to explain PGA in RA patients

Conclusions PGA does not only convey the inflammatory activity of the disease but also pain (of whatever musculoskeletal origin), fatigue, functional capacity, and psychological variables. These predictors present different relative impacts in different disease activity categories. In near-misses, the group where PGA is decisive for T2T, PGA is determined by fatigue, TJC28 and negatively associated with CRP.

Including PGA in composite indices drifts these measures away from strict disease activity representation. This may negatively affect the adequacy of target-driven decisions regarding immunosuppressive therapy, inherently designed to control inflammation.

Disclosure of Interest None declared

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.