Bayesian Communication (BC) takes advantage of publication over the World Wide Web and enables the reader of research results to evaluate those results in light of prior knowledge, belief, and values. The goals of this proposal are to establish whether the gains in comprehension, usefulness, and decision making are worth the cognitive effort required; and to establish a paradigm for evaluating in the future BC and other strategies for improving the interaction between Web-based publications and the reader. The specific hypotheses are that academic readers will find BC (1) to provide better comprehension of study results, over Evidence-Based Medicine (EBM) (traditional) methods; (2) to be more useful than traditional statistical reports; and (3) to be more helpful in making decisions based on the research literature.
As specific aims, we shall: (1) Develop a Web-based interface for novel BC and for standard EBM measures; (2) Develop and evaluate an on-line library of research reports yoked to the BC and EBM interfaces; (3) Develop an on-line environment for evaluating the use of BC and EBM interfaces; (4) Develop an on-line environment for evaluating the use of BC and EBM interfaces; (5) Perform a randomized on-line evaluation of the usefulness, comprehension, and decision assistance of BC and EBM measures by up to 200 academic users reading 1 article each. Outcome measures will include questionnaire responses, as well as feature use ad latency of responses. Because the results of research get adapted to local need and knowledge, BC has the potential for speeding the adoption of innovation, and for improving clinical care.

Agency
National Institute of Health (NIH)
Institute
National Library of Medicine (NLM)
Type
Research Project (R01)
Project #
5R01LM006849-03
Application #
6605047
Study Section
Biomedical Library and Informatics Review Committee (BLR)
Program Officer
Florance, Valerie
Project Start
2001-06-01
Project End
2005-05-31
Budget Start
2003-06-01
Budget End
2005-05-31
Support Year
3
Fiscal Year
2003
Total Cost
$410,000
Indirect Cost
Name
Johns Hopkins University
Department
Pediatrics
Type
Schools of Medicine
DUNS #
001910777
City
Baltimore
State
MD
Country
United States
Zip Code
21218
Lehmann, Harold P; Barshay, Alexander; Grimm, L Allan et al. (2005) Web-based Bayesian Communication: the Bayesian z-test. AMIA Annu Symp Proc :1023