Making Sense of Genetic Tests

Today, approximately 2,500 genetic counselors are employed in the U.S., mostly in hospitals or clinics where they counsel patients on a daily basis. Others work in diagnostic laboratories where they serve multiple roles, including writing test interpretations, coordinating research, creating and maintaining genetic databases, educating clients and health care providers, and reviewing test orders. Training for genetic counselors involves a highly specialized masters program in medical genetics and counseling. With the growing availability of genetic tests, one of the greatest benefits that genetic counselors provide to patients, hospitals, and insurers is rigorous review of test orders.

Despite their highly specialized knowledge, genetic counselors are an under-recognized resource within the healthcare system. Many different types of physicians, such as pediatricians, neurologists, and obstetricians, order genetic tests for their patients; however, most of them do not have expertise in genetics. When you consider the limited emphasis on genetics in medical schools, the ever increasing number and complexity of genetic tests, and the limited number of genetic counselors and physician geneticists in the U.S, it is not surprising that errors in genetic test ordering are common.

An Analysis of Genetic Orders

At ARUP Laboratories, genetic counselors review all DNA sequencing and other complex genetic test orders for clinical utility and cost-effectiveness. We contact the ordering physicians to clarify family history, clinical history, and previous diagnostic workup. To assess the impact of our test order review, we retrospectively analyzed all tests ordered in our laboratory in 2010. Our analysis showed that approximately one in three complex test orders was in error and that correcting these orders saved patients and hospitals approximately half a million dollars.

We also examined the types of errors that our review identified. Approximately 40% of errors were correctable by ordering an alternative, more medically appropriate test. Cancelling the test was required in 30% of the errors, because the sample had been collected incorrectly for the new recommended test.

In 10% of test order errors, the physician had ordered complete gene sequencing despite a known familial mutation. In these cases, switching the order to targeted sequencing resulted in significant cost savings. For example, testing for Lynch syndrome ordinarily involves sequencing and duplication/deletion analysis of four different genes. Knowledge of a familial mutation, on the other hand, allows testing of just one exon of one gene. An additional 10% of cases involved orders for complete gene sequencing when a targeted panel would be more appropriate. An example of this type of error involved ordering a complete CFTR gene sequence for a routine obstetric patient rather than the panel of 23 cystic fibrosis (CF) gene mutations recommended by the American Congress of Obstetricians and Gynecologists. The remaining 10% of cases consisted of genetic tests already performed on the patient, which did not need to be repeated.

Improving Orders for Genetic Tests

Our small study showed that physicians often order genetic tests incorrectly. Furthermore, we found that the cost implications of these errors are significant. Although we did not directly measure the quality or the patient safety consequences, they likely are as worrisome as the cost impact.

Performing a genetic test using a methodology that does not match the clinical scenario has major implications for test interpretation and followup care. For example, performing a targeted CF panel on a patient with a family history of a rare disease-causing mutation may lead to a false-negative test result that may inappropriately reassure the physician and family. This patient could go on to have delayed diagnosis and therapy, not to mention being subjected to additional invasive workups seeking alterative diagnoses. The counter example is problematic in other ways. Performing complete gene sequencing for CFTR on a routine obstetric patient may identify variants of uncertain significance that could confuse and worry both the physician and patient, possibly leading to additional diagnostic workups or poorly informed reproductive decisions.

Given these data and potential outcomes, laboratories should consider employing the service of a local genetic counselor to review genetic test orders. In a typical hospital setting, this might involve an hour a day. For laboratories in academic and other high-volume clinical settings, the most practical approach may be to collaborate with departments such as pediatrics, obstetrics, and oncology, which might already have genetic counselors on staff.

Based on ARUP Laboratory's experience, there is a good possibility that such a program would pay for itself by reducing erroneous orders. From a patient quality of care perspective, fewer patients would be left waiting for genetic test results that are not going to be helpful. Rather, patients would receive appropriate, cost-effective testing based on their symptoms and family history.