Feasibility of using population level genomic knowledge for public health purchasing

Although genomic knowledge and personalized medicine have great promise, the impact on public health practice and policy is unclear. This is especially true for policies for vulnerable populations or in developing countries where genetic tests are not widely available or prohibitively expensive. But while individual-level genetic information is often difficult and costly to gather, population-level genetic information is often widely available. This study explores whether ethnically based genetic information can be used to guide treatment strategies at the population level. Specifically, we examine the cost effectiveness of using genetic variation (heterozygote CYP2C19*2 allele) to determine treatment for Acute Coronary Syndrome (ACS) with either a low cost treatment (clopidogrel) or a higher cost, more effective alternative (prasugrel). Our two part, risk-benefit decision analytic model linking direct health care costs and outcomes (QALYs) to rates of adverse events suggests clopidogrel is a cost effective alternative to presugrel ($4,476/ QALY) for the genetic variation in our baseline sample population. Using this baseline model, we identify the thresholds at which presugrel becomes a cost effective population level strategy by varying the rates of genetic variation in the population (15% for Europeans, 24% for Maori, 29% for Asian, and 45% for Pacific Islanders), relative effectiveness of the different treatments, and cost of adverse events. Thus, this study presents rates of genetic variation from a number of countries and discusses whether it is feasibility to use population level genetic information to determine the most effective and cost effective population level treatment strategies when resources are constrained.