Online Program

Small Area Estimation and Validation Framework

Tuesday, November 5, 2013 : 12:50 p.m. - 1:10 p.m.

Ali Mokdad, PhD, Institute for Health Metrics and Evaluation, University of Washington, Seattle, WA
The County Health Rankings ranks US counties based on their overall health outcomes and risk factors, allowing communities to create collaborative action plans to improve the health and well-being of constituents. Rankings data is currently challenged by biases arising from use of self-reported data obtained from national health surveys. These data also have limited sampling among some US subpopulations and regions. The Institute for Health Metrics and Evaluation proposes to address these challenges by creating new estimates of mortality and leading health risk factors for all US counties using our small area estimation and validation framework, based on five national survey datasets. Our small area estimation framework integrates three unique modeling approaches to best fit different types of survey data. Each approach is optimized and validated using a gold-standard dataset to ensure the most accurate and precise estimates possible. Use of multiple data sets creates more robust estimations, and permits adjustment for self-reporting bias. Factors to be reported include: tobacco and alcohol use, obesity, diet, physical inactivity, as well as diabetes, high cholesterol, hypertension and insurance coverage. This work will ensure timely and accurate rankings that allow the public health community to understand and address health inequities and challenges.

Learning Areas:

Biostatistics, economics
Program planning
Public health or related research

Learning Objectives:
Describe three unique modeling approaches to best fit different types of survey data.

Keyword(s): Statistics, Epidemiology

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I am a professor at the University of Washington and have worked on health metrics for over 20 years.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.