Sharon Zhao

Sharon Zhao

Lead Data Scientist
Pronouns she/her

Xiaohong (Sharon) Zhao is a senior data scientist at Mathematica. She has extensive experience using data science techniques to measure health quality and gleaning insights from health and social data. Her work focuses on the testing and validation of quality measures, including risk adjustment, reliability testing, measure refinement, and endorsement of hospital inpatient and outpatient quality measures and the Agency for Healthcare Research and Quality’s (AHRQ’s) quality indicators. Her expertise includes machine-learning algorithms, data visualization and reporting, and workflow analysis and design.

Currently, Zhao is working on several quality measure projects to support quality improvement for clients in the government and private sector. She serves as module lead in a project to help AHRQ refine its patient safety and pediatric quality indicators. She led multiple statistical analysis to detect the impacts of measure refinements. Zhao also led feature/model selection, and performance validation on improving risk-adjustment models. Furthermore, Zhao led the design and implementation of data processing and analysis workflow for testing the SEP-1 Early Management Bundle: Severe Sepsis/Septic Shock chart-abstracted measure. The modularized workflow enables efficient code development and convenient code review for researchers. Zhao also led the programming implementation of Intraoperative Hypotension measure testing, including tuning the risk-adjustment model and testing for reliability and validity. Results from the initial testing have been published in Anesthesia & Analgesia.

Before joining Mathematica in 2015, Zhao was a research scientist at Siemens Corporate Research. She used machine-learning methods to detect patterns of genomic sequencing data. Zhao has a Ph.D. in computer science from Simon Fraser University.

  • Quality measures: AHRQ patient safety indicators, pediatric quality indicators, chart-abstracted measures, electronic clinical quality measures
  • Groupers/taxonomies: ICD-10-CM-PCS AHRQ Clinical Classifications Software Refined, Elixhauser Comorbidity variables, Medicare Severity Diagnosis-Related Groups
  • Machine learning
  • Risk-adjustment models
  • Data visualization and reports
  • Algorithm and workflow
Focus Area Topics
  • Health
  • Quality Improvement
  • Medicare
  • Health Information Technology and Analytics

See Clearly. Act Quickly.

Our experts can help you make smart, sustainable decisions. From local to global challenges in health, human services, and international development, we’re here to improve public well-being and make progress together.

Work With Us