
There is a more recent version of this eprint available. Click here to view it.
Strong, M. and Oakley, J.E. (2011) Bayesian Inference for Comorbid Disease Risks Using Marginal Disease Risks and Correlation Information From a Separate Source. Medical Decision Making, 31 (4). 571 - 581. ISSN 0272-989X
Abstract
Background: Public health interventions are increasingly being evaluated for their cost-effectiveness. Such interventions act ‘upstream’ on the determinants of ill health and commonly reduce the incidence of several diseases. Diseases that share determinants are usually correlated at an individual level, which we observe as comorbidity. This paper is motivated by the problem of estimating comorbid disease state risks when only single disease risk estimates are available. Methods: A case study is presented based on a physical activity cost-effectiveness model. The correlation between the risk of coronary heart disease, stroke and diabetes is estimated from cross sectional data using a Bayesian multivariate probit model. This is then combined with disease specific marginal baseline risks and intervention effects to give comorbid disease state risks. The expected numbers of QALYs gained through avoiding the comorbid states is estimated from disease specific utility data under a range of assumptions. Finally, the incremental benefit of physical activity is calculated under these utility assumptions. The difference in incremental benefit due to the intervention’s impact on reducing or increasing the disease risk correlations is explored in a sensitivity analysis. Results: If comorbidity is not taken into account, incremental benefit is overestimated compared with all scenarios in which the comorbidity is included in the model. Overestimation is greatest when physical activity is assumed to reduce disease state co-occurrence as well as disease risk. Conclusions: The proposed method reduces overestimation of benefit and allows the sensitivity to different assumptions about the correlation between disease risks to be determined.
Metadata
Item Type: | Article |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | © 2011 Sage. This is an author produced version of a paper subsequently published in Medical Decision Making. Uploaded in accordance with the publisher's self-archiving policy. |
Keywords: | Public Health; Cost-effectiveness analysis; Bayesian; Multivariate probit model; Correlated binary data; Evidence synthesis; PHYSICAL-ACTIVITY; MONTE-CARLO; STROKE; STATES |
Dates: |
|
Institution: | The University of Sheffield |
Academic Units: | The University of Sheffield > Faculty of Science (Sheffield) > School of Mathematics and Statistics (Sheffield) |
Depositing User: | Symplectic Sheffield |
Date Deposited: | 09 May 2014 09:55 |
Last Modified: | 26 Nov 2016 05:50 |
Published Version: | http://dx.doi.org/10.1177/0272989X10391269 |
Status: | Published |
Publisher: | SAGE Publications |
Refereed: | Yes |
Identification Number: | 10.1177/0272989X10391269 |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:78431 |
Available Versions of this Item
- Bayesian Inference for Comorbid Disease Risks Using Marginal Disease Risks and Correlation Information From a Separate Source. (deposited 09 May 2014 09:55) [Currently Displayed]