Skip to main content

Precision Population Analytics: Population Management at the Point-of-care

This on-demand webinar does not offer CE credit.

Tang PC, Miller S, Stavropoulos H, Kartoun U, Zambrano J, Ng K. Precision population analytics: population management at the point-of-care. J Am Med Inform Assoc. 2021;28(3):588-595. doi:10.1093/jamia/ocaa247

Read the article

Watch the Recording



Paul Tang, MD, MS
Palo Alto Medical Foundation
Kenney Ng, PhD
Principal Research Staff Member
Center for Computational Health

Manager and Moderator

Hannah Burkhardt
PhD Candidate
University of Washington School of Medicine
James Rogers
PhD Student
Columbia University


Statement of Purpose

Machine learning methods have been applied to clinical data from electronic health records (EHRs), primarily in the inpatient and intensive-care settings. We applied machine-learning techniques to support outpatient personalized management of three common chronic diseases by creating precision cohorts of similar patients in similar clinical situations. We presented our dynamic population analysis of treatment outcomes for the index patient’s relevant precision cohort at the point-of-care, integrated within the EHR. We also applied the same precision-cohort analysis to the organization’s entire patient population for each disease. The reports calculate the organization’s potential clinical performance if the best-practice personalized treatments were adopted for each patient.

Our methods focused on developing a repeatable workflow that can be applied to each chronic disease. The workflow includes:

  • Defining disease-specific clinical treatment “decision points” that can be automatically extracted from the longitudinal patient EHR data and used for modeling and analysis
  • Identifying appropriate confounding variables for patient similarity matching based on both knowledge-driven clinical guidelines and data-driven feature selection from EHR data
  • Dynamically constructing a precision cohort of patient events that are clinically similar to the current state of the index patient for analysis
  • Presenting the patient-specific analysis to the clinician at the point-of-care within the EHR.

Target Audience

The target audience for this activity is professionals and students interested in health informatics.

Learning Objectives

The general learning objective for all of the JAMIA Journal Club webinars is that participants will

Use a critical appraisal process to assess article validity and to gauge article findings' relevance to practice

After participating in this webinar, the listener should be able to:

  • Understand limitations of applying RCT results in clinical practice
  • Learn a method for analyzing observational data to create precision cohorts of patients similar to an index patient in a similar clinical situation
  • Formulate an approach to visualize and integrate observational data analysis in clinicians’ EHR workflow to inform clinical decisions at the point-of-care
  • Apply analysis of precision cohorts within an organization to derive best practices for common chronic conditions (precision population analytics)

Commercial Support

No commercial support was received for this activity.

Disclosures for this Activity

Kenney Ng discloses that he is an employee of IBM.

The following presenters, planners, and staff who are in a position to control the content of this activity disclose that they and their life partners have no relevant financial relationships with commercial interests:

  • Presenter: Paul Tang
  • JAMIA Journal Club planners: Hannah Burkhardt, Kirk E. Roberts, James Rogers
  • AMIA Staff: Susanne Arnold, Pesha Rubinstein
Dates and Times: -
Type: Webinar
Course Format(s): On Demand