Background/Purpose .  The relationship of medical noncompliance with poor health outcomes has been described in chronic diseases, including systemic lupus erythematosus (SLE).  These outcomes are amplified in SLE due to its multi-organ involvement and high-risk immunosuppressant therapy.  The association of noncompliance with socio-demographic and disease factors has been poorly studied and inconsistently identified.   In past studies, noncompliance in SLE has been defined as rate of loss to follow-up (LTF) of scheduled outpatient appointments.  Reported LTC rates have ranged from 24% to 48%.   We investigate the relationship of socio-demographic and disease factors with noncompliance and focus on self-reported damage, health status, and depression in a predominantly indigent Black cohort.

Methods .  We selected a sample of indigent SLE patients from the Grady Lupus Clinic who consented into the Georgians Organized Against Lupus (GOAL). GOAL is a longitudinal population-based cohort of patients with validated SLE from Atlanta, Georgia, who are surveyed annually on disease outcomes and access to care.  Grady Lupus Clinic (GLC) is the only lupus clinic in GA that serves a large indigent population with SLE. Data on appointment compliance was abstracted from GLC electronic medical records.  The first outcome, LTF, was defined as no GLC visits for the past 12 months within the past 24-month period.  The second outcome, rate of appointment noncompliance, was calculated as 1 minus the rate of accomplished GLC visits in the past 12-month period.  Logistic regression and multi-way ANOVA analyses examined the association of socio-demographic and disease factors with outcomes.

Results .  The sample consisted of 127 patients, predominantly non-white (94%), females (91%) and living below 100% the poverty threshold (73%).  Depression increased the risk of LTF, while longer disease duration and poverty protected against LTF (Table 1).

The adjusted appointment noncompliance rate was higher in patients who self-reported severe disease activity (34%; 95% CI 19-48%), compared to those with mild disease activity (18%; 95% CI 1-35%), p-value 0.058.  In contrast, patients with organ damage tended to have lower appointment noncompliance (22%; 95% CI 8-26%) than those without organ damage (33%; 95% CI 16-50%), although the difference was not statistically significant (p 0.083).

Conclusion .   Barriers to optimizing SLE outcomes include medical noncompliance defined as loss to follow-up and appointment noncompliance.  The first step in overcoming this barrier is recognizing the associated socio-demographic and disease factors.  Our data show that lower compliance is seen in patients with depression, shorter disease duration, higher income, and higher self-reported disease activity.  With these factors in mind, targeted intervention for improved outcomes can be tailored for those within this vulnerable population.