This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 562
Type: Topic Contributed
Date/Time: Wednesday, August 4, 2010 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract - #309126
Title: Validity of Statistical and Computational Methods to Ascertain Causes of Death: Can They Outperform Physicians?
Author(s): Christopher J.L. Murray and Jeanette Kurian Birnbaum*+ and Abraham D. Flaxman and Sean Green
Companies: Institute for Health Metrics and Evaluation and Institute for Health Metrics and Evaluation and Institute for Health Metrics and Evaluation and Institute for Health Metrics and Evaluation
Address: 2301 5th Ave, Seattle, WA, 98121,
Keywords: causes of death ; verbal autopsy ; Bayesian statistics ; machine learning ; population health measurement
Abstract:

In areas without vital registration, causes of death are commonly studied using information collected after death from household members, called verbal autopsy (VA). VA's are traditionally analyzed by physicians, an expensive and time-intensive process. Computer-based algorithms offer more efficient analytic alternatives. These algorithms can be "trained" on VA data for which true causes of death are ascertained from hospital records, and then applied to other VA's. The Symptom Pattern method is a Bayesian algorithm that assesses the probability distributions of VA responses to estimate causes of death at both the individual and population levels. Methods from the field of machine learning such as random forests are also promising options for assigning causes of death. We apply the Symptom Pattern method and random forests and compare their performance to traditional methods.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2010 program




2010 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.