Background/Purpose: Though disease flares are common, very little is known about strategies RA patients use for self management (SM) of flares. We asked patients to identify SM strategies and explored potential predictors of strategies.

Methods: 512 patients in the Canadian early ArThritis CoHort (CATCH) completed the OMERACT preliminary flare questionnaire (PFQ) at clinic visits from 11-2011 through 4-2012. Patients who self-identified as being in a flare provided ratings of flare severity, pain, disability (HAQ) and identified SM strategies they were using. Strategies were selected from those previously reported during OMERACT RA Flare patient focus groups. Rheumatologists rated whether their patient was in a flare and performed joint counts. Groups were stratified based on Patient-MD agreement of flare status and compared using ANOVA. Multivariable logistic regression was used to identify potential predictors of flare SM.

Results: 512 patients with early RA who were mostly female (75%), white (82%) and well educated (57% > HS) answered the PFQ. Patients had a mean (SD) age of 53 (14) yr, 18% smoked, 65% RF+, 53% CCP+ and 24% had erosions. Mean HAQ was 1.03 (.70) and pain was 56 (27). 149 (29%) patients self-identified flare whereas MDs identified 169 cases of flare (31%); patients and MDs agreed about flare status 72% of the time (Κ=.34). Patients who were female, current smokers, RF+, Anti-CCP+, minority, living alone and ≤ HS education were significantly (p<.05) more likely to be classified as being in a flare.

The most common SM strategy was taking more analgesics (51%); in contrast, few patients reported taking more steroids (5%) and 34% tried to manage the flare without medications. Other strategies differed by patient/MD agreement on flare status (see Table). When patients and MDs agreed the patient was in a flare, 87% reported using SM strategies; whereas when patients but not MDs identified flare, 65% used SM strategies (p=.001). Patient/MD agreement about flare status was also associated with a significantly (p<.05) greater likelihood of activity reduction/avoidance. Although few patients contacted the care team for help prior to the visit, patient/MD agreement about flare status was associated with >5 fold increase in asking for help. Across strategies, predictors of SM included patient/MD agreement, female sex, and higher disability; other sociodemographic and disease characteristics were not reliably associated with SM.

Conclusion: Disease flares are common at routine care visits in early RA. Most patients recognize when they are flaring and their rheumatologists agree. Patients report using several flare SM strategies including taking more analgesics and reducing activities. Patient/MD agreement, female sex and higher disability are predictors of flare SM efforts. Notably, few patients (11%) experiencing flare in this early RA sample reported asking care providers for help prior to the routine clinic visit.