REAL-WORLD EFFECTIVENESS OF STATIN THERAPY AND PREDICTIVE MODELING OF ADHERENCE AMONG PATIENTS WITH SEVERE HYPERCHOLESTEROLEMIA
Author(s)
Vikash K. Verma, MBA, PharmD1, Louis Brooks Jr, MS2, Marissa Seligman, PharmD3, Abhimanyu Roy, MBA4, Abhinav Nayyar, MBA, MBBS5, Ankitkumar Arora, MPharm6, Nandana Acharjee, Other7, Saba Wajih, BDS8, Anuj Gupta, MSc8, Vishan Khatavkar, MBA9, Gargi Mahashay, BTech5, Arunima Sachdev, MA4, Aamir Bashir, PhD10, Pankaj Bhardwaj, MBA, RPh9;
1Optum Lifesciences, Boston, MA, USA, 2Optum, Bloomsbury, NJ, USA, 3Optum, Winchester, MA, USA, 4Optum, Gurgaon, India, 5Optum Life Sciences, Gurugram, India, 6Optum Global Solutions, Gurgaon, India, 7Optum, Gurugram, India, 8Optum Lifesciences, Noida, India, 9Optum Lifesciences, Gurugram, India, 10Optum Global Solutions, Gurugram, India
1Optum Lifesciences, Boston, MA, USA, 2Optum, Bloomsbury, NJ, USA, 3Optum, Winchester, MA, USA, 4Optum, Gurgaon, India, 5Optum Life Sciences, Gurugram, India, 6Optum Global Solutions, Gurgaon, India, 7Optum, Gurugram, India, 8Optum Lifesciences, Noida, India, 9Optum Lifesciences, Gurugram, India, 10Optum Global Solutions, Gurugram, India
OBJECTIVES: To assess the real-world effectiveness of high- and moderate-intensity statins on low‑density lipoprotein cholesterol (LDL‑C) reduction among statin-naïve adults with severe hypercholesterolemia (LDL-C ≥190 mg/dL) and to develop predictive model for statin adherence.
METHODS: A retrospective new-user cohort study was conducted using Optum® Market Clarity data (01/01/2016-03/31/2025). Eligible adults (≥20 years) had (1) baseline LDL‑C ≥190 mg/dL, (2) no statin use during a 6‑month baseline period, and (3) continuous medical and pharmacy enrollment ≥12 months pre‑index and ≥6 months post‑index. The index date was the first statin fill within 90 days of the qualifying LDL‑C value. Patients were required to have ≥1 LDL‑C measurement at 3, 6, or 12 months post‑index. Outcomes included absolute and percent LDL‑C change, adherence (proportion of days covered [PDC] ≥80%), discontinuation, adverse effects (30-60 days), and cardiovascular (CV) events. Multiple machine‑learning algorithms (logistic regression, random forest, gradient boosting) identified predictors of non‑adherence.
RESULTS: Among 639,861 patients screened, 5,904 met inclusion criteria. High‑intensity statins were initiated in 2,678 patients (45.3%), producing mean LDL‑C reductions of 101.2 mg/dL (43.5%) at 6 months and 100.2 mg/dL (43.0%) at 12 months. Moderate‑intensity statins (n=3,177; 53.8%) achieved smaller reductions: 30.5 mg/dL (18.7%) and 26.7 mg/dL (16.4%) at 6 and 12 months, respectively. Adherence declined over follow‑up (n=5,291 at 1-3 months; 1,920 at 3-6 months; 1,711 at 6-12 months), and ~25% discontinued therapy. A total of 766 CV events and 942 adverse effects occurred within 30-60 days. The predictive models showed strong performance (AUC range 0.75-0.80) and identified key predictors of non‑adherence including demographic, clinical, treatment-related, and behavioral factors.
CONCLUSIONS: High‑intensity statins produced substantial LDL‑C reductions in routine clinical practice, but declining adherence, discontinuation, and early adverse effects may limit long‑term effectiveness. Machine‑learning models accurately identified patients at elevated risk for non‑adherence, supporting opportunities for proactive, personalized lipid‑lowering interventions to optimize outcomes.
METHODS: A retrospective new-user cohort study was conducted using Optum® Market Clarity data (01/01/2016-03/31/2025). Eligible adults (≥20 years) had (1) baseline LDL‑C ≥190 mg/dL, (2) no statin use during a 6‑month baseline period, and (3) continuous medical and pharmacy enrollment ≥12 months pre‑index and ≥6 months post‑index. The index date was the first statin fill within 90 days of the qualifying LDL‑C value. Patients were required to have ≥1 LDL‑C measurement at 3, 6, or 12 months post‑index. Outcomes included absolute and percent LDL‑C change, adherence (proportion of days covered [PDC] ≥80%), discontinuation, adverse effects (30-60 days), and cardiovascular (CV) events. Multiple machine‑learning algorithms (logistic regression, random forest, gradient boosting) identified predictors of non‑adherence.
RESULTS: Among 639,861 patients screened, 5,904 met inclusion criteria. High‑intensity statins were initiated in 2,678 patients (45.3%), producing mean LDL‑C reductions of 101.2 mg/dL (43.5%) at 6 months and 100.2 mg/dL (43.0%) at 12 months. Moderate‑intensity statins (n=3,177; 53.8%) achieved smaller reductions: 30.5 mg/dL (18.7%) and 26.7 mg/dL (16.4%) at 6 and 12 months, respectively. Adherence declined over follow‑up (n=5,291 at 1-3 months; 1,920 at 3-6 months; 1,711 at 6-12 months), and ~25% discontinued therapy. A total of 766 CV events and 942 adverse effects occurred within 30-60 days. The predictive models showed strong performance (AUC range 0.75-0.80) and identified key predictors of non‑adherence including demographic, clinical, treatment-related, and behavioral factors.
CONCLUSIONS: High‑intensity statins produced substantial LDL‑C reductions in routine clinical practice, but declining adherence, discontinuation, and early adverse effects may limit long‑term effectiveness. Machine‑learning models accurately identified patients at elevated risk for non‑adherence, supporting opportunities for proactive, personalized lipid‑lowering interventions to optimize outcomes.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
MSR100
Topic
Methodological & Statistical Research
Topic Subcategory
Artificial Intelligence, Machine Learning, Predictive Analytics
Disease
SDC: Cardiovascular Disorders (including MI, Stroke, Circulatory)