Predictors of High-cost Patients With Noninfectious Inflammatory Eye Diseases

Noninfectious inflammatory eye diseases (NIIEDs), such as uveitis, is a general term used to describe a complex mix of acute, chronic, allergic, and inflammatory disorders. Prior literature has established that, in addition to severe clinical burden, NIIEDs is associated with significant economic burden for US payers; however, no literature provides a current estimate of the economic burden associated with patients with high-cost NIIEDs. This study aimed to better understand the cost and resource use distribution and predictors of patients with high-cost NIIEDs.

This retrospective cohort study selected adult patients with NIIEDs from a large US administrative claims database between 2006 and 2015. Among the included patients, total all-cause health care costs were calculated for a randomly selected 12-month period. Patients in the top 20% of total all-cause health care costs were identified as high-cost patients; the remaining patients were identified as lower-cost patients. Patient demographic characteristics, clinical characteristics, cost, and health care resource utilization (HRU) were compared. Logistic regression models were used to determine characteristics associated with high-cost patients.

Patients with NIIEDs (n = 14,879) were categorized into 2976 high-cost and 11,903 lower-cost patients. High-cost patients with NIIEDs were significantly more likely to experience blindness, cataract, cystoid macular degeneration, retinal detachment, and visual disturbances during the follow-up period than the lower-cost patients (all P < 0.05). The high-cost patients accounted for ~77% of the total all-cause health care spend. High-cost patients incurred an average annual total health care cost of $59,873, and the top 1 percentile incurred $349,967 during the follow-up period. Hospitalization was a key cost driver among the high-cost patients, accounting for 50% of the total cost among the top 1 percentile of patients. High-cost patients were more likely to have specific autoimmune diseases, inpatient admission, and use of biologic and immunosuppressant agents.

A small segment of patients with NIIEDs consumed most resources. This study identified several predictors based on patient characteristics and HRU that may help inform the profile of patients with NIIEDs with the highest health care needs. As such, patients with a given profile can be selected for targeted interventions by clinicians to potentially help improve quality of care and to reduce costs.