FRI0424
PAIN SCORES ARE THE PRIMARY EXPLANATORY VARIABLE FOR HIGHER GLOBAL ESTIMATES BY PATIENTS COMPARED TO DOCTORS IN PATIENTS WITH ALL RHEUMATIC DISEASES
I. Castrejon 1,*Y. Yazici 1J. Samuels 1T. Pincus 1
1Rheumatology, NYU Hospital For Joint Diseases, New York, United States
Background: Estimates of global status by doctors (DOCGL) and patients (PATGL) are discordant in about 30-40% of patients with rheumatoid arthritis (RA).1,2 This discordance has been analyzed to date only in RA patients.
Objectives: To analyze levels of discordance between DOCGL and PATGL in all patients with any diagnosis seen in usual clinical care at a rheumatology setting.
Methods: Each patient seen at an academic rheumatology clinical setting since 2005 completes a self-report MDHAQ (multidimensional health assessment questionnaire) at each visit, with scales for physical function, pain, PATGL, fatigue, anxiety, depression and quality of sleep, and demographic data. DOCGL was completed by 2 rheumatologists. One random visit of patients seen between 2005 and 2011 was analyzed, patients were classified as PATGL=DOCGL (PATGL and DOCGL within 2 of 10 units), PATGL>DOCGL (PATGL ≥2 units than DOCGL), and DOCGL≥PATGL (DOCGL ≥2 units than PATGL). Univariate odds ratios were computed to identify variables associated with discordance. Significant variables (p<0.05) were included in multivariate models, with selected variables when indicated.
Results: In a total of 980 patients studied, 509 (52%) had PATGL=DOCGL, 371 (38%) PATGL>DOCGL, and 100 (10%) DOCGL>PATGL. Patients with PATGL>DOCGL were more likely to be female, have less formal education and have higher MDHAQ scores (Table). In multivariate analysis, higher pain and fatigue scores were independent predictors of PATGL>DOCGL. If MDHAQ scores for pain and fatigue were not included in a second model, female gender, lower education and higher scores for depression and sleep problems were independent predictors of PATGL>DOCGL. In patients with DOCGL>PATGL, only lower fatigue was associated in multivariate analysis with lower odds of discordance (OR=0.88, 95% CI 0.79-0.98).
TABLE: Univariate and multivariate logistic regression of predictors of discordance between patient and physician global estimates.
|
PATGL ≥2 units greater than DOCGL |
DOCGL≥2 units greater than PATGL | |||
|
Univariate OR (95% CI) |
Model 1 Adjusted OR1 (95% CI) |
Model 2 Adjusted OR2 (95% CI) |
Univariate OR (95% CI) |
Model 3 Adjusted OR3 (95% CI) | |
|
Age |
1.00(0.99-1.01) |
-- |
-- |
1.00 (0.99-1.02) |
-- |
|
Female |
1.62*(1.2-2.2) |
0.93 (0.62-1.40) |
1.19 (0.83-1.71) |
0.99 (0.63-1.56) |
-- |
|
Education |
0.93*(0.89-0.98) |
0.97 (0.91-1.03) |
0.93*(0.88-0.99) |
1.08 (0.99-1.17) |
-- |
|
Ethnicity (Hispanic/AA) |
1.43*(1.03-2.08) |
1.03 (0.66-1.60) |
1.17 (0.80-1.72) |
0.57 (0.28-1.15) |
-- |
|
MDHAQ-FN (0-10) |
1.49*(1.37-1.62) |
0.94 (0.83-1.07) |
-- |
0.85 (0.73-1.00) |
-- |
|
MDHAQ-PN (0-10) |
1.51*(1.42-1.60) |
1.41*(1.29-1.54) |
-- |
0.98 (0.90-1.06) |
-- |
|
MDHAQ-FT (0-10) |
1.33*(1.27-1.40) |
1.16*(1.08-1.25) |
-- |
0.84*(0.77-0.92) |
0.88*(0.79-0.98) |
|
RAPID3 (0-30) |
1.29*(1.25-1.34) |
-- |
-- |
0.92*(0.88-0.97) |
0.97 (0.91-1.02) |
|
Sleep (0-3) |
1.77*(1.50-2.09) |
0.95 (0.73-1.23) |
1.67*(1.35-2.19) |
0.69*(0.51-0.94) |
0.92 (0.64-1.32) |
|
Anxiety (0-3) |
1.65*(1.36-1.99) |
1.06 (0.71-1.56) |
0.94 (0.66-1.34) |
0.70 (0.48-1.00) |
-- |
|
Depression (0-3) |
1.84*(1.51-2.25) |
1.12 (0.75-1.68) |
1.53*(1.06-2.19) |
0.83 (0.57-1.20) |
-- |
*p<0.05
Conclusions: 38% of patients estimated their status as worse than their physicians. These patients were more likely to score higher for pain and fatigue, be female and less educated than patients whose physicians estimated clinical status as similar or worse than patients.
References: 1. Barton JL, Imboden J, et al. Arthritis Care Res. 2010;62:857-64.
2. Khan NA, Spencer HJ, et al. Arthritis Care Res. 2011.
Disclosure of Interest: None Declared