Persistence of antimalarial therapy and its effect on mortality in systemic lupus erythematosus: A population-based longitudinal multistate study

ObjectiveExamine patients' Antimalarials (AM) use history and identify determinants of transitions among AM persistence, discontinuation, and death in a population-based incident cohort of Systemic Lupus Erythematosus (SLE) patients receiving AM.MethodsWe use population-based health administrative data to follow all British Columbia residents with incident SLE who initiate AM. Persistence is defined as the period from the pharmacy fill date until the supplied quantity is exhausted, with a gap of at least three consecutive months without an AM prescription, marking entry into the discontinuation state; episodes are tracked as transitions between these states and death. Transition-specific hazard ratios (HRs) are estimated using a reversible frailty multistate Cox model, adjusting for baseline demographics, comorbidities, healthcare use, and medications.ResultsThe cohort comprises 7,399 incident SLE patients initiating AMs. On average, patients spent half of their follow-up time in a state of persistence and the other half in discontinuation. The transitions show that while 99.2% of persistence episodes end in a lapse to discontinuation, only 64.5% of discontinuation episodes result in a return to persistence. Crucially, the risk of an episode ending in death is over five times higher during periods of discontinuation compared to persistence. Lapsing from persistence is less likely in rural residents (HR 0.94), users of cardiovascular drugs (0.87), and patients in higher-income areas (0.93), but more likely in those with alcohol-use disorder (1.47). Returning to persistence is inhibited by older age (0.995), depression (0.92), and higher comorbidity (0.97), yet encouraged by NSAID use (1.08). Among persistent patients, death is associated with older age (1.06 per year) and rural residence (1.67). Death during periods of discontinuation increases in men (HR 1.55), with older age (1.09 per year), greater outpatient use (1.014), the use of glucocorticoids (1.24), and cardiovascular drugs (1.30), and decreases with higher income (0.63 for quintile 4).ConclusionAM discontinuation is associated with higher mortality. Lower income, alcohol misuse, depression, multimorbidity, and complex drug regimens undermine AM persistence and heighten mortality. Financial support, counselling to reduce alcohol use, integrated mental-health care, and careful medication review may sustain persistence and improve survival in SLE.