DECREASED RATES OF HEALTH PLAN DISENROLLMENT FOR PATIENTS WITH RHEUMATOID ARTHRITIS

OBJECTIVES: To estimate the time to and rates of health plan disenrollment for rheumatoid arthritis (RA) patients compared with a matched group of non-RA patients. METHODS: RA patients (≥18 years) were identified from the HealthCore Integrated Research Database (HIRD) and matched to non-RA patients using exact 1-to-l matching according to demographic characteristics, primary policy holder status, ACA plan, and length of pre-index enrollment. Patients were identified as having RA if they had ≥ 1 claim with a diagnosis of RA and ≥ 1 claim with DMARD use. Non-RA patients were those who did not meet the criteria for RA. Index date for RA cases was the earliest date at which they met the criteria for RA during 1/1/ 2007-12/31/2014. Non-RA patients were assigned same index date as their matched RA case. All patients were required to have 12 months of pre-index enrollment. RESULTS: 63,908 members were matched (31,954 per cohort). Median follow-up time was 3.1 years for RA and 1.9 for non-RA patients. Members with RA had lower rate of disenrollment compared with non-RA members (rate ratio=0.60; 95% CI=[0.59, 0.61]). Within those who disenrolled, time to disenrollment was longer for RA patients (median=806 vs 470 days). Cox regression showed similar results (hazard ratio (HR)=0.61) and results were consistent across sensitivity analyses: (1) including death as disenrollment (HR=0.61) and (2) excluding patients who died during the first 30 days of follow-up (HR=0.63). Results were mainly driven by higher disenrollment during the first 12 months in the non-RA group, with the difference between the cohorts weakening over time. CONCLUSIONS: RA patients stayed with health plan longer and were less likely to disenroll than non-RA patients. This may be due to phenomenon called job-lock in which patients with chronic disease are reluctant to leave their job due to the fear of losing their health insurance.