Premium
Making Medical Decisions in Dependence of Genetic Background: Estimation of the Utility of DNA Testing in Clinical, Pharmaco‐Epidemiological or Genetic Studies
Author(s) -
Nguyen Thuy Trang,
Schäfer Helmut,
Timmesfeld Nina
Publication year - 2013
Publication title -
genetic epidemiology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.301
H-Index - 98
eISSN - 1098-2272
pISSN - 0741-0395
DOI - 10.1002/gepi.21701
Subject(s) - epidemiology , biobank , medicine , population , confidence interval , genetic epidemiology , framingham heart study , framingham risk score , disease , genetics , environmental health , biology
ABSTRACT An index measuring the utility of testing a DNA marker before deciding between two alternative treatments is proposed which can be estimated from pharmaco‐epidemiological case‐control or cohort studies. In the case‐control design, external estimates of the prevalence of the disease and of the frequency of the genetic risk variant are required for estimating the utility index. Formulas for point and interval estimates are derived. Empirical coverage probabilities of the confidence intervals were estimated under different scenarios of disease prevalence, prevalence of drug use, and population frequency of the genetic variant. To illustrate our method, we re‐analyse pharmaco‐epidemiological case‐control data on oral contraceptive intake and venous thrombosis in carriers and non‐carriers of the factor V L eiden mutation. We also re‐analyse cross‐sectional data from the Framingham study on a gene‐diet interaction between an APOA 2 polymorphism and high saturated fat intake on obesity. We conclude that the utility index may be helpful to evaluate and appraise the potential clinical and public health relevance of gene‐environment interaction effects detected in genomic and candidate gene association studies and may be a valuable decision support for designing prospective studies on the clinical utility.