Background/Purpose: Several studies have shown that significant discordance exists between patient (PATGL) and physician (DOCGL) global measures of rheumatoid arthritis (RA) disease activity. These studies have focused on patient-related factors associated with discrepancy. However, relatively little is known about possible physician-related factors that might explain the DOCGL. The aim of the present study was to determine potential physician-related factors associated with the DOCGL score.

Methods: Physician data (e.g. age, sex, title and experience) were requested by questionnaire from physicians who use the Danish nationwide rheumatological DANBIO-registry in their clinical work. DANBIO data based on first encounters between patients and physicians in the registry were analysed. Patient data included: sex, age, duration of disease, Health Assessment Questionnaire (HAQ) score, C-reactive protein (CRP), swollen (SJC28) and tender joint counts (TJC28), treatment (biologic/ non bio), and 100 mm visual analogue scales (VAS) for PATGL, pain  and fatigue. Multivariate linear regression analysis was performed to determine physician- and patient-related factors that affected DOCGL. 

Results: 90 physicians returned the questionnaire (34%). The physicians were matched with 8,970 patient encounters, for which physicians were 55% females, age 52 (47- 56) years (median, Interquartile Range (IQR)), 78% were consultants in rheumatology, no. of patients seen per month per physician was 80 (50-120). Patients were 74% females, age 61 (51- 70) years, disease duration 7 years (2-15), CRP 6 (3-12) mg/L, HAQ 0.75 (0.25-1.375), SJC 1 (0-3), TJC 2 (0-6), pain 32 (14-57) mm and fatigue 42 (19-67) mm. 34% were treated with biologics. The median DOCGL was 12 (5-27) mm and PATGL 39 (17-64) mm – a difference of 27 mm.

Table 1 shows patient and physician factors associated with the DOCGL. DOCGL was higher in male physicians, in the younger half of physicians and in physicians who were not consultants. Patient gender did not affect DOCGL, whereas SJC and TJC did. Patient age was statistically but not clinically significant for DOCGL. Thus, on average, the DOCGL in a male non-consultant less than 52 years old was 9.8 mm higher than that of a female consultant above 52 years of age.

 

Table 1. Results of multivariate linear regression analysis of physician and patients factors as predictors of DOCGL, only statistically significant variables included.
Variable	Estimate	95 % CI	p
Intercept	-0.577
Physician
Sex    female	1
male	3.345	2.672 – 4.018	<0.001
Age  > 52 years	1
≤ 52 years	3.461	2.754 – 4.168	<0.001
Consultant in rheumatology, yes	1
No	3.126	1.747 – 4.505	<0.001
Patient
Age (10 years’ increase)	-0.0018	-0.0024 – -0.0014	<0.001
Patient VAS-global	0.147	0.135 – 0.159	<0.001
CRP	0.116	0.100 – 0.133	<0.001
SJC28 (range 0-28)	2.775	2.672 – 2.878	<0.001
TJC28 (range 0-28)	0.745	0.680 – 0.809	<0.001
Referent group for continuous variables is represented by each 1-unit increase in that variable when not otherwise stated. Results adjusted for patient gender, clinic site, country for medical exam, research experience, experience with joint scoring, number of patients seen monthly, and assessed level of importance of inflammation, fibromyalgia, comorbidity and patient social factors. VAS = Visual Analogue Scale, measured on a 0-100 mm scale; CRP= C-reactive protein, measured in mg/L; SJC28 = swollen joint count on 28-joint assessment; TJC28 = tender joint count on 28-joint assessment. Adjusted R2 = 0.69.

Conclusion: This study of 90 physicians matched with >8,000 RA patients treated in routine care showed clinically significant inter-physician variations in the physicians´ global score. Male physicians scored on average 3.1 mm higher than female, physicians that were < 52 years old scored 3.7 mm higher than their older counterparts, and consultants scored 3.1 mm lower than non-consultants. SJC and TJC, but not patient age and gender, also affected the physicians´ global score. The significant contribution of physician-associated factors may have implications for research involving measures of physicians´ global score and for clinical care.

---

« Back to ACR/ARHP Annual Meeting 2012

ACR Meeting Abstracts - https://acrabstracts.org/abstract/physicians-global-assessment-is-affected-by-physicians-age-and-gender-but-not-by-patient-age-and-gender-in-rheumatoid-arthritis-patients-treated-in-routine-care-data-from/