Evidence-based classification of recommendations on use of genomic tests in clinical practice: Dealing with insufficient evidence

Muin J Khoury, Ralph J Coates, James P Evans

Numerous genomic tests continue to emerge as potential tools in the diagnosis, treatment, prognosis, and prevention for a wide variety of common human diseases. To date, most of these tests have "insufficient evidence" of clinical validity and utility for their use in clinical practice. Explicit and quantitative tools can be used in the evaluation of direct and indirect evidence on the utility of genomic tests. As suggested in an article in this month's issue by Veenstra et al., a recommendation matrix can be developed based on the amount of certainty of the evidence and the assessment of the risk-benefit profile. To supplement the current binary (up or down) evidence-based recommendation for use, it is worthwhile to explore all available data to develop a three-tier evidence-based recommendation classification of genomic tests ("use in practice," "promote informed decision-making," and "discourage use"). Promoting informed decision making may be a valuable recommendation for tests for which there is sufficient information on analytic and clinical validity and for which the risk/benefit analysis on clinical utility is promising but not definitive. This approach could provide interim guidance for clinical practice, while rigorous outcomes research is conducted to assess the impact of such tests on patients, families, and population health outcomes.