Data Scientist - HOS Analytics

GENERAL DUTIES/RESPONSIBILITIES:

• Collaborate with Stars, clinical, and care management leaders to understand HOS performance drivers and translate them into analytical solutions.
• Segment members by likelihood to respond to the HOS survey and by predicted physical or mental health trajectory, to prioritize outreach and intervention.
• Build and fine-tune models predicting member-level risk of HOS composite decline, including Improving/Maintaining Physical Health, Improving/Maintaining Mental Health, Monitoring Physical Activity, and Reducing the Risk of Falling.
• Analyze item-level HOS survey data, not just composite scores, to isolate which specific survey questions are driving measure performance.
• Build pipelines that track HOS survey administration cycles - baseline and follow-up cohorts, sampling, fielding, and response rates - and validate case-mix adjustment logic against CMS methodology.
• Model statistical significance and year-over-year variance to distinguish genuine performance shifts from survey noise or sampling error.
• Coordinate with HOS survey vendors on cohort tracking, sampling methodology, fielding timelines, and case-mix adjustment specifications.
• Collaborate with engineering teams to version, test, and deploy models using Git, CI/CD pipelines, and virtual machine (VM) environments.
• Partner with clinical, care management, complex case management, Compliance, and Legal teams to ensure HOS-related data outputs and interventions align with CMS guidance.
• Build dashboards tracking measure-level HOS performance against Star Rating cut points and prior-year trend for each composite.
• Help standardize definitions, documentation logic, and reporting workflows to scale HOS analytics enterprise-wide.

To perform this job successfully, an individual must be able to perform each essential duty satisfactorily. The requirements listed below are representative of the knowledge, skill, and/or ability required. Reasonable accommodations may be made to enable individuals with disabilities to perform the essential functions.

• Minimum Experience:
  
  2+ years of relevant experience in predictive modeling and analysis, with demonstrated application to longitudinal or outcomes-based member data.

• Required:
  
  PhD in Computer Science, Engineering, Mathematics, Statistics, or related field

• Demonstrated experience with CMS HOS survey methodology, including baseline/follow-up cohort design, case-mix adjustment, and item response theory or other psychometric methods.
• Working knowledge of individual HOS composite measures:
  Improving/Maintaining Physical Health, Improving/Maintaining Mental Health, Monitoring Physical Activity, and Reducing the Risk of Falling.
• Working knowledge of CMS Star Ratings procedures, including how HOS measures are weighted, cut-point determined, and incorporated into the overall Star Rating.

• Excellent communication, analytical, and collaborative problem-solving skills.
• Experience building end-to-end data science solutions and applying machine learning methods to real-world problems with measurable outcomes.
• Solid data structures and algorithms background.
• Strong programming skills in one of the following:
  Python, Java, R, Scala, or C++.
• Demonstrated proficiency in SQL and relational databases.
• Experience with data visualization and presentation, turning complex longitudinal outcomes analysis into clear, actionable insight for non-technical stakeholders.
• Experience setting experimental or analytical frameworks for complex, ambiguous scenarios, including longitudinal/cohort study design.
• Understanding of relevant statistical measures such as confidence intervals, significance of error measurement, and development/evaluation data sets.
• Experience manipulating and analyzing complex, high-volume, high-dimensionality, and unstructured data from varying sources.

• Healthcare experience, particularly within Medicare Advantage.
• Experience with functional status, frailty, or geriatric outcomes modeling.
• Experience in cloud ecosystems such as Azure or AWS.
• Published work in academic conferences or industry circles related to survey methodology, psychometrics, or health outcomes measurement.
• Demonstrable track record of handling ambiguity, prioritizing needs, and delivering results in an agile, dynamic environment.

The work environment characteristics described here are representative of those an employee encounters while performing the essential…