How to know when physicians are ready for genomic medicine

See allHide authors and affiliations

Science Translational Medicine  13 May 2015:
Vol. 7, Issue 287, pp. 287fs19
DOI: 10.1126/scitranslmed.aaa2401


Despite perceptions to the contrary, physicians are as prepared for genomic medicine as they are for other medical innovations; educational initiatives and support from genetics specialists can enhance clinical practice.

With the tremendous investment in the genomic sciences over the last two decades, the biomedical community is eager to apply new genomic knowledge to patient care. Genomic testing, including whole-genome and exome sequencing, has demonstrated clinical utility in certain contexts (1). However, the workforce of fewer than 2000 board-certified medical geneticists is insufficient to meet the demand created by the greater prominence of genomics in clinical medicine (2). As a result, physicians without specialized genetics training will be increasingly called upon to order genomic testing and use the results in the care of their patients. At the same time, several studies have found that physicians report “unpreparedness” and low confidence in their ability to apply genomic data to patient care (3). Here we discuss how to assess whether physicians are ready for the genomic revolution and whether previous medical innovations have been held to the same standards.


Despite recent standardization efforts, most molecular laboratories do not currently use a uniform approach to the clinical interpretation of the large number of genetic variants revealed by genome sequencing, most of which have unknown clinical significance. Even for variants known to be pathogenic, often insufficient data exist to predict the likelihood of disease in an individual who is currently asymptomatic. Communicating this uncertainty to the clinician remains a challenge for the clinical laboratory. For example, if genomic sequencing for a 60-year-old patient with no signs, symptoms, or family history of heart disease uncovers a rare genetic variant in a known cardiomyopathy gene, the treating clinician has no data to guide her in interpreting the significance of that finding for her patient. The complexity of clinical sequencing reports might be prone to misinterpretation by nongeneticist physicians, leading to over- or underestimation of the disease risk associated with a given variant. Such results might prompt physicians to order an expensive cascade of follow-up diagnostic tests, each with its own potential complications, risks to the patient, and costs (4).

The concerns voiced above reflect an appropriate respect for key principles in medicine, including the avoidance of patient harm and good stewardship of limited health care resources. But at what point do they create an artificial crisis that distracts from more productive questions that might point the way forward for a new field? If physicians are not ready to use genomic medicine now, how will we know when they are? A useful thought experiment is to ask the same of other medical innovations. When did the biomedical community declare nonradiologist physicians ready to order computerized tomography (CT) scans for their patients and make clinical decisions based on their findings? Introduced in the 1970s, CT carries risks associated with radiation, intravenous contrast agents, and the discovery of radiographic lesions incidental to the test’s primary indication. However, with guidance from their radiologist colleagues, physicians routinely use CT for the standard management of conditions ranging from abdominal pain and headaches to cancer and stroke.

On what grounds would we hold genomic technology to different standards? Similar to the radiologist, the genomics-laboratory director holds primary responsibility for interpreting sequencing results using available data and generating reports that physicians can understand. The receiving physicians may choose to communicate back with the specialist for further guidance in medical decision-making. As genomic medicine grows in prominence, nongeneticist physicians might also seek greater support from genetic counselors than they now do in practice.

Genetic exceptionalism is the idea that genetic technology and information are inherently different from other routine processes in medical care and, by extension, should be handled more cautiously (5). Additional caution might be appropriate for genetic information that is potentially stigmatizing or anxiety provoking. But like any other medical test, most genetic testing is used for diagnosis, prognosis, and therapeutics. Similarly, the degree of physician preparedness for genomic medicine is not exceptional among other complex medical innovations such as myriad types of imaging, microscopic pathology assessment, or targeted therapies.


The concerns about physician preparedness for genomic medicine are also problematic because no universal definition of preparedness exists. To a large extent, these concerns stem from surveys in which physicians have reported little experience with genomic medicine in their practices and low perceived confidence in their ability to order genetic tests and manage the results appropriately (3). But self-reported attitudes and perceptions do not necessarily correlate with skills and behavior. Even objective genomics knowledge assessment might not adequately determine whether a physician can use genomic medicine in his or her practice.

To assess residency education across the spectrum of medical specialties, the Accreditation Council for Graduate Medical Education (ACGME) has identified six core competencies that training programs should target: patient care, knowledge for practice, practice-based learning and improvement, interpersonal and communication skills, professionalism, and systems-based practice; the Association of American Medical Colleges (AAMC) has added two others: interprofessional collaboration and personal and professional development (6). Although these discrete competencies facilitate the evaluation of trainees, they oversimplify what it means to be an effective physician. To better describe the roles a physician plays, medical specialties are now defining entrustable professional activities (EPAs), the “professional activities that together constitute the mass of critical elements that operationally define a profession” (7). Each EPA (for example, triaging ill patients to an appropriate level of medical care) can map to several competencies (for example, the medical knowledge, procedural skills, and professionalism required to triage a patient appropriately).

Efforts are under way to apply this educational framework to genomic medicine for the nongeneticist. In anticipation of the more widespread diffusion of genomics into clinical medicine, genomics education has become a priority for the National Human Genome Research Institute (NHGRI) of the U.S. National Institutes of Health. It has developed a Genetics/Genomics Competency Center for Education (G2C2) to curate high-quality curricula and other resources for physicians and other health care providers, searchable by genomic medicine competencies. Through its Inter-Society Coordinating Committee on Practitioner Education in Genomics (ISCC-PEG), NHGRI is partnering with more than 30 professional organizations representing generalists and specialists, such as the American Medical Association, the American Heart Association, and the American Board of Ophthalmology (8). The ICSS-PEG has developed five genomics EPAs in the areas of (i) family history, (ii) genomic testing, (iii) treatment based on genomic results, (iv) somatic genomics, and (v) microbial genomics, each of which can be mapped to specific genomic medicine competencies (Fig. 1) (3). The ICSS-PEG recognizes that “preparedness” for genomic medicine will have different definitions for different types of physicians; thus the committee addresses common educational needs while supporting individual professional societies and specialty boards in developing their own genomics EPAs for their members. For example, for the primary care physician, the EPA of eliciting, documenting, and acting on relevant family history pertinent to a patient’s clinical status might include the recognition of a potential hereditary colorectal cancer syndrome such as Lynch syndrome. For the ophthalmologist, this EPA would include using a family history assessment in a patient with retinitis pigmentosa to identify the potential implications of its inheritance pattern for family members. It is unknown whether the application of an EPA framework to genomic medicine will improve clinical practice, but defining genomic medicine competencies and EPAs is a first step toward evaluating the degree to which medical learners at all stages of training and practice are meeting them.

Fig. 1 Get ready for genomic medicine.

Shown are three EPAs for physicians and examples of the competencies each requires. Not shown are EPAs for somatic and microbial genomics. CME, continuing medical education.PHOTO



Beyond physician education, there remain several challenges to the clinical integration of genomic medicine. For many clinical contexts, the impact of genomic testing on patient outcomes, such as improved survival or quality of life, remains to be demonstrated (9). The electronic health records of most health systems are not currently prepared to incorporate the complexity of genomic data into clinical care, despite the tremendous opportunities for integrated clinical decision support to help busy physicians use genomic data in medical decision-making (10). However, the preparedness of physicians themselves for genomic medicine is not unique among other diagnostic and therapeutic innovations. The fundamental principles of medicine in genomics are the same as elsewhere in medicine—modest training and support from specialists and health systems will prepare nongeneticist physicians to use genomics in the care of their patients.


  1. Acknowledgments: The authors are supported by the U.S. National Institutes of Health (grant numbers KL2-TR001100, L30-DK089597, R01-HG005092, U01-HG006500, and U19-HD077671).

Stay Connected to Science Translational Medicine

Navigate This Article