The Program Integrity Data Scientist II is responsible for developing, implementing, managing, and deploying in-depth analyses that meet the information needs associated with payment accuracy, anomaly detection, and Fraud, Waste, and Abuse (FWA).
Build concepts as algorithms that identify claims for pre- or post-pay intervention based on probability of fraud, waste, and abuse. Algorithms are implemented into production workflows for action: medical record request and audit, downcode adjustment, denial and remittance communication, etc.
Analyze and quantify claim payment issues and provide recommendations to mitigate identified program integrity risks.
Identify trends and patterns using standard corporate, processes, tools, reports and databases as well as leveraging other processes and data sources.
Conduct outcome analyses to determine impact and effectiveness of corporate program and payment integrity initiatives.
Collaborate on the examination and explanation of complex data relationships to answer questions identified either within the department or by other departments as it relates to payment accuracy, anomaly detection, and FWA.
Monitoring of and providing explanation of anomalies related to trends associated with the potential for Fraud Waste and Abuse across the corporate enterprise.
Collaborate with the Legal Department, generating data and analyses to support Legal proceedings.
Develop hypothesis tests and extrapolations on statistically valid samples to establish outlier behavior patterns and potential recoupment.
Create, maintain, and communicate an analytical plan for each project.
Mine and analyze large structured and unstructured datasets.
Employ wide range of data sources to develop algorithms for predicting risk and understanding drivers, detecting outliers, etc.
Develop visualizations that demonstrate the efficacy of developed algorithms.
Provide statistical validation and analysis of outcomes associated with clinical programs and interventions.
Collaborate with other teams to integrate with existing solutions.
Communicate results and ideas to key stakeholders.
Prepare code for operationalization of end-to-end model pipeline and deliverable for business consumption.
Perform any other job related duties as requested.
Bachelor's degree in Data Science, Mathematics, Statistics, Engineering, Computer Science, or a related field required
Equivalent years of relevant work experience may be accepted in lieu of required education
Three (3) years data analysis and/or analytic programming required
Healthcare experience required
Proficient in SQL and at least one of the following programming languages: Python / R / RAT STAT
Ability to perform advanced statistical analyses and techniques including t-tests, ANOVAs, z-tests, statistical extrapolations, non-parametric significance testing, and sampling methodologies
Working knowledge of predictive modeling and machine learning algorithms such as generalized linear models, non-linear supervised learning models, clustering, decision trees, dimensionality reduction and natural language processing
Proficient in feature engineering techniques and exploratory data analysis
Familiarity with optimization techniques and artificial intelligence methods
Ability to analyze large quantities of information and identify patterns, irregularities, and deficiencies
Proficient with MS office (Excel, PowerPoint, Word, Access)
Demonstrated critical thinking, verbal communication, presentation and written communication skills
Ability to work independently and within a cross-functional team environment
Experience with cloud services (such as Azure, AWS or GCP) and modern data stack (such as Databricks or Snowflakes) preferred
Familiarity with SAS is preferred
Preferred beginner level of knowledge of developing reports or dashboards in Power BI or other business intelligence applications
Knowledge of healthcare coding and billing processes, including CPT4, HCPCS, ICD-9, DRG and Revenue Codes preferred