Start Your Search

Yingjun (Victoria) Zhu

Author of

  • OS10 - Disease and Clinical Management with Technology (ID 16)

    • Event: e-Health 2017 Virtual Meeting
    • Type: Oral Session
    • Track: Clinical and Executive
    • Presentations: 1
    • Coordinates: 6/06/2017, 10:30 AM - 12:00 PM, Room 205B
    • OS10.01 - The Concepts of Canadian Population Grouping Methodology and Its Application (ID 173)

      Yingjun (Victoria) Zhu, Case Mix, Canadian Institute for Health Information; Ottawa/CA

      • Abstract
      • Slides

      Purpose/Objectives: CIHI’s newest product – a Canadian based population grouping methodology looks at an individual’s experiences with the healthcare system across many sectors, including inpatient stays, day surgeries, physician visits, ED visits, hospitalizations for mental health illnesses and stays in long-term care (LTC) and complex continuing care facilities. Using data collected over an extensive time period, the grouping methodology classifies individuals, including healthy people, into cohorts with similar clinical characteristics; producing cost weights which compute the population’s current and future burden of morbidity, their number of primary healthcare and ED visits, as well as the likelihood of being admitted to a LTC facility. Using these outputs, users are able to quantify the health burden of the population, identify high cost users, provide inputs into funding models, monitor population health and diseases and profile/predict future healthcare usage.

      Methodology/Approach: The population grouping methodology starts with everyone who is eligible for healthcare and then looks at interactions with the healthcare system and diagnosis information over a 2-year period to describe a person’s health status and predict their future morbidity and system use. This means, everyone, including individuals who haven’t had any interactions with the healthcare system and those with no health conditions are also included in the analysis, providing a true picture of the entire population. By using age, sex, health conditions and the most influent health condition interactions as the predictors, regression models have been applied. These models produce predictive indicators for the concurrent period as well as one year into the future. While the data is produced at the individual level, the power of the model lies in the user’s ability to aggregate the data by population segments and compare healthcare resource utilization by different geographic regions, health sectors, socio-economic status and health status.

      Finding/Results: By comparing with other similar products in the world, CIHI’s population grouping methodology has been proved to have similar or better predictive power.

      Conclusion/Implication/Recommendations: CIHI’s population grouping methodology is a very useful tool for profiling and predicting healthcare burden and future system use, with key applications for health policy makers, planners and funders. The presentation will focus on how these user groups can apply the outputs of the methodology to aid in their decision making and planning processes. For example, we will demonstrate how the cost weights can be used in large-scale funding models, to set physician capitation rates and identify/monitor high system and high-cost users. We will demonstrate how the clinical profiles created across the continuum of care can be used for disease surveillance and monitoring. Finally, we will demonstrate how the outputs from the models can be used to predict future system use patterns, such as visits to a family medicine physician.

      140 Character Summary: A very useful tool for profiling and predicting healthcare burden and future system use, with key applications for health policy makers, planners and funders.

      Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.