This Article Full Text (PDF) References Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Saved Citations Download to citation manager Add to My Marked Citations Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Byass, P. Articles by Hoang Van Minh, Search for Related Content PubMed Articles by Byass, P. Articles by Hoang Van Minh, Social Bookmarking                   What's this?

A probabilistic approach to interpreting verbal autopsies: methodology and preliminary validation in Vietnam

UmeÅ International School of Public Health, UmeÅ University, UmeÅ, Sweden, peterbyass{at}aol.com

Dao Lan Huong

Health Strategy and Policy Institute, Ministry of Health, Hanoi, Vietnam

Hoang Van Minh

Faculty of Public Health, Hanoi Medical University, Hanoi, Vietnam

Aims: Verbal autopsy (VA) has become an important tool in the past 20 years for determining cause of death in communities where there is no routine registration. In many cases, expert physicians have been used to interpret the VA findings and so assign individual causes of death. However, this is time consuming and not always repeatable. Other approaches such as algorithms and neural networks have been developed in some settings. This paper aims to develop a method that is simple, reliable and consistent, which could represent an advance in VA interpretation. Methods: This paper describes the development of a Bayesian probability model for VA interpretation as an attempt to find a better approach. This methodology and a preliminary implementation are described, with an evaluation based on VA material from rural Vietnam. Results: The new model was tested against a series of 189 VA interviews from a rural community in Vietnam. Using this very basic model, over 70% of individual causes of death corresponded with those determined by two physicians, increasing to over 80% if those cases ascribed to old age or as being indeterminate by the physicians were excluded. Discussion: Although there is a clear need to improve the preliminary model and to test more extensively with larger and more varied datasets, these preliminary results suggest that there may be good potential in this probabilistic approach.

Key Words: verbal autopsy • Bayesian • probability • cause of death • Vietnam • mortality.

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

This article has been cited by other articles:

				Nguyen Thi Kim Chuc and V. K. Diwan FilaBavi, a demographic surveillance site, an epidemiological field laboratory in Vietnam Scand J Public Health, January 1, 2003; 31(1): 3 - 7. [Abstract] [PDF]