Topic Modelling for Medical Prescription Fraud and Abuse Detection. Issue 3 (16th December 2018)

• Record Type: Journal Article
• Authors: Zafari, Babak; Ekin, Tahir
• Is Part Of: Journal of the Royal Statistical Society. Volume 68:Issue 3(2019)
• Page Start: 751
• Page End: 769
• Publication Date: 2018-12-16
• Subjects: Fraud analytics -- Fraud detection -- Healthcare fraud -- Hierarchical Bayesian methods -- Medicare part D -- Opioid abuse -- Prescription; Statistics -- Periodicals; 519.5
• DOI: 10.1111/rssc.12332
• Languages: English
• ISSNs: 0035-9254

Abstract

Summary: Medical prescription fraud and abuse have been a pressing issue in the USA, resulting in large financial losses and adverse effects on human health. The size and complexity of the healthcare systems as well as the cost of medical audits make use of statistical methods necessary to generate investigative leads in prescription audits. We analyse prescriber–drug associations by utilizing the real world Medicare part D prescription data from New Hampshire. In particular, we propose the use of topic models to group drugs with respect to the billing patterns and exhibit the potential aberrant behaviours while using medical specialities as a covariate. The prescription patterns of the providers are retrieved with an emphasis on opioids and aggregated into distance-based measures which are visualized by concentration functions. This output can enable healthcare auditors to identify leads for audits of providers prescribing medically unnecessary drugs.