Although adherence to hydroxychloroquine (HCQ) therapy does not adequately predict adherence to other lupus medications in children with systemic lupus erythematosus (SLE), concurrent assessment of immunosuppressant (IS) adherence can provide useful information, according to study findings published in Seminars in Arthritis and Rheumatism.
This retrospective cohort study included 423 children with SLE aged 5 to 18 years identified in the Medicaid and Truven Health MarketScan Commercial databases between 2013 and 2018. Of the study patients, 319 received mycophenolate, 97 received azathioprine, and 84 received methotrexate. Medicaid enrollees accounted for 63% of the study patients.
The proportion of days covered (PDC) over 365 days was calculated to estimate adherence among users of different medications among new users of ISs and HCQ. Full adherence was defined as PDC of at least 80%. Intraclass correlation coefficients (ICC) and kappa statistics were used to assess agreement between estimates of adherence. Correlations between nonadherence and hospitalizations were estimated for HCQ, IS, or concurrent use of both using negative binomial regression models. Baseline disease severity, healthcare utilization, and demographic characteristics were used for adjustments.
Full adherence ranged from 51% to 52% in the HCQ group, and 33% to 45% in the IS group. There was modest agreement between HCQ and IS adherence, with mycophenolate performing better than methotrexate (ICC 0.55 [95% CI, 0.47-0.63] vs 0.27 [95% CI, 0.07-0.46], respectively). Subsequent hospitalizations were 2.9 (95% CI, 1.2-7.1) times higher for HCQ nonadherence than for HCQ adherence; for IS and concurrent nonadherence, subsequent hospitalizations were 5.9 (95% CI, 2.4-14.6) and 5.6 (95% CI, 2.0-15.5) times higher, respectively. Subsequent hospitalizations were better predicted by IS nonadherence compared with HCQ nonadherence alone.
Limitations of the study included the use of pharmacy refill data, an inability to determine accurate adherence or why medication stops and switches occurred, a lack of consideration of other medications or the effects of polypharmacy, the potential for misclassification of SLE cases, and the exclusion of uninsured children.
The study authors concluded, “While HCQ adherence estimates in children and adults with SLE remain a valuable tool, adherence to other immunomodulatory medications can provide different and useful information. Concurrent assessment of adherence to both HCQ and IS medications should be considered whenever possible, and efforts to address medication-specific barriers should be made when adherence to different medications is discordant.”
Chang JC, Costenbader KH. Hydroxychloroquine and immunosuppressant adherence patterns and their association with subsequent hospitalization rates among children with systemic lupus erythematosus. Published online June 8, 2022. Semin Arthritis Rheum. doi:10.1016/j.semarthrit.2022.152042