It is uncommon to find published clinical trials that measure the health benefits of medical testing. As a consequence, policy makers often have to decide whether access to, or public funding of, medical tests is warranted without knowing the clinical impact of testing on the patient. In the situation where a policy maker is considering a companion genetic test and tailored drug therapy, deficiencies in the evidence base are exacerbated because two technologies need to be assessed and the proposed genetic biomarker needs to be validated. The Linked Evidence Approach (LEA) is a methodology that was developed in 2005 to cope with inadequacies in the evidence supporting medical test evaluations. In 2010 the approach was adapted to the evaluation of pharmacogenetic interventions. This article describes how LEA and similar analytic frameworks are used internationally, highlights particular challenges with the approach, and proposes ways that LEA might be applied to pharmacogenomic interventions.