(P-366) Trajectories of Corticosteroid Dosing Pattern Among Patients with Systemic Lupus Erythematosus

Background: Despite complications associated with long term use, corticosteroid plays a major and essential role in the treatment of systemic lupus erythematosus (SLE).

Objectives: To define the trajectories of corticosteroid dosing in patients with SLE by group-based trajectory modeling (GBTM) using information on longitudinal course of dosing pattern to treatment over time, and to recognize characteristics according to the trajectory groups.

Methods: We performed a cohort study using the Korean Health Insurance Review and Assessment Service-National Patients Sample database during 2017. Individuals with SLE were defined as the patient who had experienced at least one diagnostic for SLE (International Classification of Diseases, 10th revision code M32). The GBTM was applied to assess dosing pattern during a 6-month follow-up in SLE patients starting treatment with a corticosteroid. The optimal GBTM was determined on the basis of the balance between each model's Bayesian information criterion and the percentage of patients in the smallest group in each model. In addition, a multinomial logistic regression model was used to examine the relationship between characteristics and the likelihood of being a member of different trajectory groups.

Results: Among all of 914 patients with SLE, the number of patients starting treatment with a corticosteroid was 428. Three GBTM dosing pattern groups were suggested by the optimal model, and patients were categorized as demonstrating high-then-moderate dosing, consistently moderate dosing, consistently low dosing during follow-up. Hospitalization had the greatest influence on dosing pattern, with hospitalized patients significantly more likely to exhibit any dosing pattern other than consistently moderate dosing (aOR 15.54, 95% CI 5.97‐40.41, aOR 3.55, 95% CI 1.52‐8.25). Patients between 40 and 59 years of age were significantly less likely to show high-then-moderate dosing than 19 years old and under (aOR 0.18, 95% CI 0.03‐0.99, aOR 0.16, 95% CI 0.03‐0.91). Medical aid patients were significantly more likely to show consistently low dosing than Health insurance (aOR 2.96, 95% CI 1.02‐8.59) but were not more likely to exhibit high-then-moderate dosing compared with those with consistently moderate dosing.

Conclusions: Through GBTM in a population of patients initiating corticosteroid therapy for SLE, three-dosing pattern trajectories were identified. In the present study, low-then-high or moderate dosing was not classified, which may represent exacerbation, due to the small sample size. Further research with a larger sample size is needed to clearly elucidate the factors that influence dosing pattern.