CLN - Focus on Laboratory Stewardship

More Than a CPT Code

A View on Genetic Testing From Inside a Health Plan

Daniel Anderson, MD, MBA, and Michael Astion, MD, PhD

Daniel Anderson, MD, and Michael Astion, MD, PhD, interviewed medical geneticist and thought leader Matthew Fickie, MD, FACMG, of Highmark Health in Pittsburgh, about the challenges of third-party payer management of laboratory testing.

Fickie is medical director at Highmark Health Blue-Branded Health Plans, which serves more than 4 million members in Pennsylvania, West Virginia, and Delaware. Fickie is also chair of the Economics of Genetic Services Committee at the American College of Medical Genetics and Genomics.

What are the top challenges third-party payers face in managing genetic testing?

There are six key problems in managing genetic testing: 1) the lack of specific Current Procedural Terminology (CPT) codes; 2) shortage of genetics experts inside insurance companies; 3) the quality of genetic assessments and medical necessity policies; 4) the fact that the focus of medical genetics research has not favored comparative analysis of genetic testing; 5) the lack of innovation and innovative partnerships within the insurance industry; and 6) abusive practices by clinical laboratories.

What kinds of challenges do CPT codes present?

Providers use Tier 1 CPT codes to bill for common genetic tests. Though not perfectly specific, these codes have enough specificity to allow insurance companies to understand what is be-ing purchased. For example, the genetic test for the common cystic fibrosis variants is coded as CPT 81220.

Tier 2 CPT codes, which range from CPT 81400 through 81408, are general genetic testing procedures arranged in order of increasing complexity, with 81400 being the least complex and time-consuming (so-called Level 1) and 81408 being the most complex (Level 9).

Tier 2 codes generally are for rarer diseases and less frequent testing than Tier 1. Each Tier 2 code has a long list of specific tests that fall under the same code. For example, more than 50 different specific genetic tests belong under CPT code 81401 (Molecular Pathology Procedure, Level 2).

This complicates insurers’ ability to deal with these tests because they don’t know the specific Tier 2 tests they are paying for, so can’t easily apply medical necessity criteria. However, Tier 2 codes at least suggest how much work is being done, so reasonable prices can be estimated.

The worst problem from a test management perspective is if a genetic test is not covered by a Tier 1 or Tier 2 code. In this case, clinical laboratories code tests as the dreaded CPT code 81479 “unlisted molecular policy procedure code.”

A tremendous amount of spending on ex-pensive cancer genomic profiling and other genetic tests gets buried in this code. Since an insurer can’t be confident about the healthcare service being purchased, it will have difficulty in applying fee sheets or medical necessity policies, both terrible circumstances from a utilization management perspective.

You mentioned a shortage of genetic expertise inside the insurance industry. What expert knowledge is missing and what tasks require it?

The insurance industry lacks medical geneticists and genetic counselors. Currently, most medical directors within insurance companies are generalists, for example family practitioners and general internists, and lack special training in genetic medicine. Yet they are still expected to establish and enforce policies involving genetic testing.

Since it takes expertise to understand, create, and enforce fair and effective policies, this lack of expertise is undesirable for patients and for the medical directors who are trying to serve them. When you take this problem and add the aforementioned nonspecific code problem, you can understand how hard this job can be for professionals without genetics expertise.

Can you comment further on the lack of quality in genetic assessments and medical necessity policies?

By genetic assessments, I am referring to analysis of the evidence underlying the use of a genetic test. Does the test measure what it claims to measure, and is it useful in improving patient outcomes?

The types of trials evaluated, and the outcomes prioritized by some health technology groups, are not appropriate for genetic diseases and tests. Medical necessity policies take the results of assessments and translate them into criteria for using a genetic test.

For example, which patients under what clinical circumstances would benefit from the test? Third-party experts or commercial organizations provide many genetic assessments and medical necessity policies, the quality of which varies. Smaller payers in particular are at risk as they likely have fewer resources to hire in-house genetic experts and to pay for high-quality third-party assistance.

What is the challenge with the direction of medical genetics research, and why does it cause problems for third-party payers?

Medical geneticists, most of whom reside in academic settings, do not focus on comparative effectiveness research. Instead, they migrate toward making gene discoveries.

My impression is that they would rather discover the tenth genetic finding related to a rare disease than perform comparative studies of various testing strategies on a population for which multiple options are available. This is despite the likelihood that the improved testing strategies would positively impact more people.

What sorts of innovative partnerships should insurance companies pursue, and what strategic advantages do these partnerships provide?

In general, payers have not pursued enough innovative partnerships with the laboratory industry. An example of an innovative partnership is the relationship between Illumina and Harvard Pilgrim Health, a health plan that covers approximately 1.2 million people in New England.

This partnership will assess the total costs and clinical outcomes of noninvasive prenatal testing (NIPT) versus traditional screening practices in a risk sharing contract. It will provide open market access of NIPT for average risk pregnancies in a way that limits the extent to which the arrangement increases overall healthcare costs.

The risk and information sharing under this agreement is an example of innovative policymaking that would benefit patients, payers, and diagnostic testing companies and could provide clarity on comparative effectiveness as discussed in my prior point. I would like to see more of this type of collaboration and experimentation around covered benefits for genetic testing.

You mentioned as a top challenge abusive practices by some clinical labs. Could you elaborate on this issue?

I recently reviewed a $40,000 cancer genomic profiling test from a commercial lab that raised several red flags with the way it was billed. Unfortunately, this kind of abuse is not rare. Some abnormal billing practices are innocent mistakes caused by a lack of coding expertise. Coding genetic tests can be a complex process, and the rules aren’t as clear as in other fields.

But some of the abusive billing practices are fraud, like purposefully misusing the coding system to maximize the likelihood of a paid claim. A specific example of this type of fraud is switching from a newer, correct CPT code that the insurance company does not cover, to an in-correct, older code that is still covered.

What predictions do you have, and what opportunities do you see, for how third-party payers will handle genetic tests in the future?

I have a few predictions. First, I think the cost of most genetic testing will continue to decline, to the point that eventually it will not be cost-effective for insurance companies to have medical professionals reviewing the claims. Insurers will find a way to automate the process so they are adjudicating the cases with computer algorithms and using human review for only the most dif-ficult, expensive cases, as well as for grievances.

Automation will continue to improve, especially with advances in artificial intelligence (AI). AI can analyze both insurance claims and electronic medical records to filter the laboratory claims most worthwhile for medical professionals to review. The current process is labor intensive and difficult. For example, I think AI might be able to do a better job in natural language processing of notes and separating initial cases of cancer from relapsed cancer. This type of separation is important because relapsed cancer is an indication for a variety of expensive cancer genomic profiling tests and treatments.

Another prediction is that test utilization and policy implementation will improve as a younger generation of medical professionals, who are more educated about genetic testing, rise through the ranks.

I also believe there are opportunities for a variety of technology companies to develop rapid, reliable, web-based tools to better handle the ordering and preauthorization of genetic tests. These tools could also automatically route tests to in-network labs for the insurance companies paying the claims.

Finally, I think there are going to be many more indications for cancer genomic profiling for risk stratification and treatment selection in a variety of neoplasms. This will benefit cases in which there is indeterminate pathology and will be an improvement over current surveillance strategies. Insurance companies will need to have up-to-date medical necessity policies in this arena and adjudicate cases fairly and consistently, as the research on these technologies is moving fast.

Daniel Anderson, MD, MBA, is chief resident in anatomic pathology and laboratory medicine at the University of Washington in Seattle. +EMAIL: [email protected]

Michael Astion, MD, PhD, is clinical professor of laboratory medicine at the University of Washington department of laboratory medicine, and medical director of the department of laboratories, Seattle Children’s Hospital. +Email: [email protected]


CPT 2018 Professional. Chicago: American Medical Association; 2018.
McCallister EM. De-risking risk reduction. BioCentury 2018. Available from: (Accessed February 2019)

CLN's Laboratory Stewardship Focus is supported by Seattle Children's Patient-Centered Laboratory Utilization Guidance Services

Seattle Children's Patient-Centered Laboratory Utilization Guidance Services