About the research
This study explores the cost-effectiveness of three weight-loss medications prescribed to obese patients in primary care settings.
The researchers:
- investigated the relationship between body mass index (BMI) and three health factors, cardiovascular events, mortality and diabetes, using data from the Clinical Practice Research Database
- conduced a systematic review of clinical evidence for three weight-loss drugs, orlistat, sibutramine and rimonaban
- conducted a full synthesis of the clinical evidence using a mixed treatment comparison
- reviewed the existing literature of cost-effectiveness models that evaluate drug interventions in treating obesity
- compared the cost-effectiveness of the three interventions. Drawing on data from the Health Survey for England, the relationship between BMI and health related quality of life controlling for age, cardiovascular events and diabetes were further explored.
The researchers used change in body mass index (BMI) to capture the clinical effects of the interventions. The relationship between quality of life and BMI was captured using EQ-5D quality of life data, a validated generic quality of life instrument collected in the Health Survey for England. Time to cardiovascular and diabetes events were derived from the Clinical Practice Research Database, and health state costs were sourced from the published literature such as medical journals and databases.
Evidence from 41 clinical studies were used to inform treatment-related changes in BMI. Results from seven BMI risk models showed consistent increases in risk as a result of increasing BMI. A curvilinear relationship was found between EQ-5D scores and BMI, with underweight and overweight or obese respondents having lower quality of life scores than those of normal weight.
In short, the economic model found sibutramine to be the most cost-effective intervention. While this treatment has since been withdrawn due to safety issues, the study’s novel approach brought a more rigorous method to understanding the relationship between quality of life and BMI, as previous analyses used simple linear models obtained using ordinary least square regressions and did not take into account the cardiovascular and diabetes effects.
Methodology
The analysis is based on a mixed treatment comparison that was used to estimate treatment-related changes in BMI using summary statistics from 41 studies. The risk of major events (diabetes onset, heart attack, stroke or death) and time to events were derived from a cohort of UK patients using Weibull proportional hazards regression models. The cost-effectiveness of each intervention was compared using a cohort simulation model.
The Health Survey for England datasets were merged to explore the relationship between quality of life and BMI. Covariates included age, gender, comorbidities related to obesity (diabetes, angina, stroke, heart attack) and recent event. The relationship was estimated using a user-written analysis program in GAUSS using maximum likelihood estimation of an adjusted limited dependent variable mixture model.
Publications
This research was featured in the following academic journals:
Ara, R., Blake, L., Gray, L., Hernandez, M., Crowther, M., Dunkley, A., Warren, F., Jackson, R., Rees, A., Stevenson, M., Abrams, K., Cooper, N., Davies, M., Khunti, K., and Sutton, A. (2012) ‘What is the clinical effectiveness and cost-effectiveness of using drugs in treating obese patients in primary care? A systematic review.’ Health Technology Assessment 16(5), pp. 1-202. doi: 10.3310/hta16050 Retrieved 2 September 2013 from http://www.journalslibrary.nihr.ac.uk/__data/assets/pdf_file/0005/64769/FullReport-hta16050.pdf
Gray, L., Cooper, N., Dunkley, A., Warren, F.C., Ara, R., Abrams, K., Davies, M.J., Khunti, K., and Sutton, A. (2012). ‘A systematic review and mixed treatment comparison of pharmacological interventions for the treatment of obesity’. Obesity Reviews 13(6), pp. 483-498. doi: 10.1111/j.1467-789X.2011.00981.x Retrieved 2 September 2013 from http://onlinelibrary.wiley.com/doi/10.1111/j.1467-789X.2011.00981.x/pdf