APPLICATION OF COPULAS IN ECONOMIC EVALUATION

Author(s)

Crespo C*1;Monleon A1;Díaz W2;Rodriguez C1;Brosa M3, Rios M1 1University of Barcelona, Barcelona, Spain, 2Universidad de Antioquia, Medellin, Colombia, 3Oblikue Consulting, Barcelona, Spain

OBJECTIVES: The aim of this study was to analyse the applicability of copulas distribution in economic evaluation. METHODS: We have analyzed data from an observational prospective study of patients with allergic rhinitis in Spain (n=498). Main data were direct cost (€2012) and Health Related Quality of Life (SF-12). We have calculated the goodness of fit for copulas (Gumbel copula, Clayton Copula, Frank Copula, Normal Copula, Plackett Copula and T copula) based on the empirical process comparing the empirical copula with a parametric estimate of the copula derived under the null hypothesis. We have used inversion of Kendall's tau method to fit copulas. A multivariate independence sample was generated to compare with copulas results. This process was replicated for a 100 times to obtain p-values by bootstrap method. RESULTS: Marginal distribution of direct cost was a 3-parameter Gamma distribution (shape=1.856, scale=0.00324, location=10.97). Marginal distribution of Health Related Quality of Life was associated to a 1- gamma (shape 2.9253 and scale 0.16104). P-value range were 0.093 to  0.144 for independent distribution, 0.004 to   0.031  for Gumbel copula, 0.246 to 0.522 for Clayton Copula, 0.545 to  0.814 for Frank Copula, 0.463 to 0.716 for Normal Copula, 0.373 to 0.628 for T Copula and 0.549 to 0.847 for Plackett Copula. Frank Copula and Plackett Copula had the best goodness of fit. Kendall’s Tau for Fran Copula showed a correlation of -0.4212. CONCLUSIONS: Copulas distribution allows us to adjust better the non-lineal relation between cost and effectiveness. Furthermore, this kind of approach could improve probabilistic sensitivity analyses.

Conference/Value in Health Info

2013-11, ISPOR Europe 2013, The Convention Centre Dublin

Value in Health, Vol. 16, No. 7 (November 2013)

Code

PRM195

Topic

Methodological & Statistical Research

Topic Subcategory

Confounding, Selection Bias Correction, Causal Inference

Disease

Respiratory-Related Disorders

Explore Related HEOR by Topic


Your browser is out-of-date

ISPOR recommends that you update your browser for more security, speed and the best experience on ispor.org. Update my browser now

×