Scientific Poster Session I - Original Research

Original Research
  Saturday, November 11, 2023
  11:30 AM–01:00 PM

Abstract

Introduction: Statin-associated muscle symptoms (SAMS) contribute to statin nonadherence. In a previous study, we have successfully developed and validated a pharmacological SAMS (PSAMS) phenotyping algorithm that distinguishes objective vs nocebo SAMS using structured and unstructured Minnesota Fairview electronic health records (EHRs) data.

Research Question or Hypothesis: We aimed to develop and validate a pharmacological SAMS risk stratification (PSAMS-RS) score using Fairview EHR data.

Study Design: Retrospective cohort study

Methods: Using our PSAMS phenotyping algorithm, we identified PSAMS cases and controls based on Fairview EHR data. We split the data into derivation (1/1/2010 to 12/31/2018) and validation (1/1/2019 to 12/31/2020) cohorts. The derivation cohort was further split into 80% training and 20% testing cohorts. EHR features were screened using Least Absolute Shrinkage and Selection Operator (LASSO). A PSAMS-RS score was constructed based on LASSO coefficients in the training set, with a score cutoff determined by optimizing precision/recall balance in the testing set.

Results: We identified 1.9% (310/16128) PSAMS patients in the derivation and 1.5% (64/4182) in the validation cohort. After fitting the LASSO regression, 4 out of 59 clinical features were determined to be significant predictors for PSAMS risk. A score >26 points is associated with significantly higher hazard of developing SAMS within a year of statin initiation in the derivation (HR, 2.01; 95% CI, 1.62-2.61) and in the validation cohort (HR, 2.03; 95% CI, 1.13-3.74). PSAMS-RS Score= 8 - 8x male gender + 2x concurrent beta-blockers use + 23x prior coronary artery disease + 3x prior peripheral vascular disease

Conclusion: The PSAMS-RS score provides a simple tool to stratify patients’ risk of developing PSAMS after statin initiation. For patients with a score >26, clinicians could take preemptive measures to prevent potential PSAMS-related statin nonadherence.

Presenting Author

Boguang Sun PharmD
University of Minnesota

Authors

Matt Loth Ph.D.
University of Minnesota

Chih-Lin Chi PhD, MBA
Institute for Health Informatics

Pui Ying Yew BS
Institute for Health Informatics

Rui Zhang PhD
University of Minnesota College of Pharmacy

Meijia Song BSc.
University of Minnesota

Robert Straka PharmD, FCCP
University of Minnesota