Influence of the different “patient global assessment” formulations on disease activity score by different indices in rheumatoid arthritis

• Published in: Clinical rheumatology Vol. 37; no. 7; pp. 1963 - 1969
• Main Authors: Ferreira, Ricardo J. O., Eugénio, Gisela, Ndosi, Mwidimi, Silva, Cristiana, Medeiros, Catarina, Duarte, Cátia, da Silva, J. A. P.
• Format: Journal Article
• Language: English
• Published: London Springer London 01.07.2018; Springer Nature B.V
• Subjects: Antirheumatic Agents - therapeutic use; Arthritis, Rheumatoid - blood; Arthritis, Rheumatoid - diagnosis; Arthritis, Rheumatoid - drug therapy; Brief Report; C-reactive protein; C-Reactive Protein - analysis; Cross-Sectional Studies; Data processing; Female; Humans; Male; Medicine; Medicine & Public Health; Middle Aged; Patients; Remission; Remission Induction; Rheumatoid arthritis; Rheumatology; Severity of Illness Index; Statistical analysis; Symptom Assessment - methods; Patient reported outcome measures; Disease activity; Patient preference; Patient global assessment; Rheumatoid arthritis; Remission induction
• ISSN: 0770-3198; 1434-9949
• DOI: 10.1007/s10067-018-4063-9

Patient global assessment (PGA) is included in almost all rheumatoid arthritis (RA) composite disease activity indices and definitions of remission. However, different PGA formulations exist and are used interchangeably in research and clinical practice. We investigated how five different PGA formulations used in four disease indices affect the remission rates. This was an ancillary analysis of data from a cross-sectional study in patients with RA. The data comprised the following: 28-joint counts, C-reactive protein, and five PGA formulations. Remission rate variation was assessed using five PGA formulations in each index (ACR/EULAR Boolean, CDAI, SDAI, and DAS28-CRP). PGA agreement was assessed by the following: Pearson’s correlation; Bland-Altman plots; paired samples t test; and establishing the proportion of patients who scored (i) all formulations within an interval of 20 mm and (ii) each formulation ≤ 10 mm. This analysis included 191 patients. PGA formulations presented good correlations (≥ 0.65), but Bland-Altman plots showed clinically significant differences, which were statistically confirmed by comparison of means. Just over a half (51.8%) of patients scored all PGA formulations within a 20-mm interval. The proportion of those scoring ≤ 10 mm varied from 11.5 to 16.2%. When different formulations of PGA were used in each index, remission differences of up to 4.7, 4.7, 6.3, and 5.2% were observed. When formulations were used in their respective indices, as validated, the remission rates were similar (13.1, 13.6, 14.1, and 18.3%). Using PGA formulations interchangeably may have implications in the assessment of disease activity and in the attainment of remission, and this can impact upon management decisions.