Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.


The Role of Objective Measures vs. Patient Reported Outcomes (PROs) as a Reflection of Flares in Patients with RA: Results from the Brigham RA Sequential Study (BRASS).

Bykerk1,  Vivian, Solomon5,  Daniel Hal, Bingham6,  Clifton O., Frits3,  Michelle, Iannaccone4,  Christine, Weinblatt5,  Michael E., Shadick5,  Nancy A.

Brigham & Women's Hospital, Boston, MA
Brigham and Women's Hospital, Boston, MA
Brigham and Women's Hospital, MA
Brigham and Womens Hospital, Boston, MA
Johns Hopkins University, Baltimore, MD

Background:

Worsening of disease activity or RA flare can be severe enough to warrant a change of therapy. Little is known about how frequently flares occur in RA or which measures best reflect a flare of RA.

Objective:

To describe the frequency of self-reported flares in a population of patients with RA and determine which variables best relate to patients' recall of flare in the prior 6 months.

Methods:

Data were collected in a prospective, observational, single- academic center cohort of RA patients treated according to preference of their rheumatologist. Patients were questioned every 6 months if they'd had a flare of their disease, what the duration was and how they'd treated it. Variables analyzed included patient reported outcomes, composite indices of disease activity and laboratory measures. Univariate logistic regression analyses using generalized estimating equations (GEE) were performed to determine possible predictors of flare. Disease measures with a p <= 0.10 were included in a multiple logistic regression model using GEE. Additional multiple logistic regression models were performed, with patient reported measures and with objective measures.

Results:

Of 1095 RA patients, 567 with >= 3 yrs follow-up reporting on flares were included for this analysis. Baseline characteristics included mean age: 57 ±13 yrs; 85% female, 93% Caucasian, mean disease duration14 years. 75% reported flares at baseline, 59% at 6 months and 55 % reported flares every 6-months thereafter. 30% reported flares lasting > 2 weeks, 27% lasting 1–2 weeks & 43% lasting <= 3 days. Variables significantly associated with current flare in the multivariate analysis overall, were pain, physician global, tender joint count (TJC), and age (figure 1). The TJC dropped out in the model containing just objective measures (figure 1). For predicting a flare within the next 6 months, pain, emotional distress, physician global, AM stiffness, and age were statistically significant (figure 2). Patient reported treatment during a flare included increases or changes in: prednisone by 35%, DMARDs by 32%, NSAIDs by 27%, biologics by 6%, narcotics by 5%, other pain meds by 13%.

Current flare analyses indicates variables that are associated with a recent flare as reported by the patient. (Odds Ratios from GEE)

Variables significantly associated with a current flare (Multivariate Regression) All Measures:
VariableOdds RatioLower 95% CIUpper 95% CIP value
Age0.98330.97110.99570.0084
CRP1.00240.99251.01240.6361
Emotional Distress FSM150.99350.98311.0040.225
Gender0.97420.5991.58440.9161
Patient Global (MDHAQ scale)1.00540.99521.01580.2999
Pain (MDHAQ Pain Scale)1.0231.01411.0319<.0001
Sleep1.00540.99521.01580.2999
Tender Joint Count (0–28)0.96380.92950.99930.0454
Physician Global1.02681.0161.0376<.0001
AM Stiffness1.22550.80781.85910.3389
Swollen Joint count (0–28)0.99460.95981.03070.7666
Patient Reported Measures:    
VariableOdds RatioLower 95% CIUpper 95% CIP value
Age0.98350.97190.99520.0059
Emotional Distress FSM150.99410.98421.0040.2426
Gender1.01010.61551.65760.9683
Patient Global (MDHAQ scale)1.00660.99721.01610.1675
Pain1.02711.01861.0357<.0001
Sleep1.00660.99721.01610.1675
AM Stiffness1.38460.91382.09810.1249
Objective Measures:    
VariableOdds RatioLower 95% CIUpper 95% CIP value
Age0.97990.97040.9896<.0001
CRP1.00530.99931.01140.085
Gender1.03950.72031.50010.836
Tender Joint Count (0–28)0.98570.96261.00930.2336
Physician Global1.03771.03031.045<.0001
Swollen Joint Count (0–28)0.99360.97061.01730.5949

Future flare analyses: variables predicting which patients who will experience a flare in the next 6 months (Odds Ratios from GEE)

Predicting Flare in Next 6 Months (Multivariate Regression) All Measures:
VariableOdds RatioLower 95% CIUpper 95% CIP value
Age0.98890.9771.0010.0716
CRP0.99640.98771.00530.4298
Emotional Distress FSM150.98790.97870.99730.0119
Patient Global (MDHAQ Scale)0.99510.98631.00390.2733
Pain (MDHAQ Pain Scale)1.01311.00461.02180.0026
Tender Joint Count (0–28)0.9740.94341.00560.1061
Physician Global1.01371.00411.02340.0052
AM Stiffness1.39991.02411.91380.0349
Swollen Joint count (0–28)1.00650.97491.03920.6889
Patient Reported Measures:    
VariableOdds RatioLower 95% CIUpper 95% CIP value
Age0.98730.97580.9990.0339
Emotional Distress FSM150.98960.98070.99870.0246
Patient Global (MDHAQ scale)0.99680.98851.00510.4453
Pain1.0141.0061.0220.0006
AM Stiffness1.54211.1342.09710.0058
Objective Measures:    
VariableOdds RatioLower 95% CIUpper 95% CIP value
Age0.98560.97580.99540.0041
CRP1.00510.99851.01170.1297
Tender Joint Count (0–28)0.99460.97381.01580.6158
Physician Global1.01981.0131.0267<.0001
Swollen Joint count (28)0.99640.97491.01830.7426

Conclusions:

Patient self reported flares are frequent in RA; 50% last >= 1 week. Patient pain VAS, physician global and TJC best reflect flare in patients with RA. When considering PROs, only pain independently reflects flare at baseline.

To cite this abstract, please use the following information:
Bykerk, Vivian, Solomon, Daniel Hal, Bingham, Clifton O., Frits, Michelle, Iannaccone, Christine, Weinblatt, Michael E., et al; The Role of Objective Measures vs. Patient Reported Outcomes (PROs) as a Reflection of Flares in Patients with RA: Results from the Brigham RA Sequential Study (BRASS). [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1778
DOI: 10.1002/art.29543

Abstract Supplement

Meeting Menu

2010 ACR/ARHP