#392 - Genetic testing: when it's valuable, how to choose the right test, and what to do with the results

May 18, 2026 Episode Page ↗
Overview

Peter Attia explores genetic testing, providing a framework to understand its clinical value, limitations, and where it offers meaningful insights versus probabilistic or non-actionable information, emphasizing when to prioritize direct phenotypic measurement.

At a Glance
12 Insights
1h 2m Duration
15 Topics
8 Concepts

Deep Dive Analysis

Introduction to Genetic Testing and its Complexities

Historical Context of the Human Genome Project

Understanding Genetic Variants and Disease Mechanisms

Major Limitations of Current Genetic Testing

Key Questions to Ask Before Genetic Testing

Genetic Testing for Cardiovascular and Metabolic Disease

Genetic Testing for Inherited Cardiac Conditions

Genetic Testing for Cancer Risk and Inherited Syndromes

Genetic Testing for Neurodegenerative Diseases

Critique of Functional Medicine Genetic Testing

Utility of Pharmacogenetics in Treatment Decisions

Framework for Evaluating Genetic Test Utility

Overview of Different Types of Genetic Tests

Interpreting Genetic Test Results and Follow-Up

Summary of Genetic Testing Use Cases and Principles

Phenotype

The observable, measurable output of the body, encompassing everything from lab values to symptoms and diseases, resulting from the interaction of genes, environment, behavior, and aging. It represents a real-time, integrated readout of health.

Mendelian Inheritance

A simplified model of genetics where a single gene mutation has a large and predictable effect on a trait or disease, often taught with dominant and recessive traits. This is an exception rather than the rule for most common diseases.

High-Penetrance Mutations

Genetic mutations where the presence of the variant almost guarantees the development of a specific disease, such as the HTT gene expansion in Huntington's disease. These are rare compared to the polygenic nature of most common diseases.

Somatic Mutations

Genetic mutations that are acquired during a person's lifetime and are not inherited from parents. These mutations are the cause of the vast majority (about 95%) of cancers and are not detectable by standard inherited genetic tests.

Germline Mutations

Genetic mutations that are inherited from parents and are present in every cell of the body. These are responsible for about 5% of cancers and are the focus of hereditary genetic testing.

Single Nucleotide Polymorphism (SNP)

A common variation in a single nucleotide that occurs at a specific position in the genome. Genotyping arrays (used in consumer tests) scan for these common variants, but they often miss rarer, more clinically significant mutations.

Polygenic Risk Score (PRS)

A composite score that aggregates the effects of thousands of common genetic variants across the genome to reflect an individual's overall genetic predisposition to a given disease relative to the population. While promising at a population level, individual utility is still early stage.

Pharmacogenetics

The study of how an individual's genetic makeup affects their response to drugs. This field uses genetic testing to guide medication selection, dosage, and safety, especially in cases where drug metabolism varies significantly between people.

?
What are the fundamental questions to ask before considering genetic testing?

Before genetic testing, you should ask what you are trying to learn, if genetics is the best tool for that question, what you will do differently with the results, and if you are mentally prepared for the answer.

?
Why did the Human Genome Project not immediately solve major disease mysteries?

The project revealed the immense scale and complexity of the human genome, with only 1.5% coding for proteins and the vast majority being regulatory "junk DNA," making accurate interpretation and understanding of function a far greater challenge than sequencing.

?
Are genetic test results deterministic or probabilistic for most diseases?

For the vast majority of common conditions like heart disease, cancer, and diabetes, genetic tests provide probabilistic risk, meaning they shift the likelihood but do not guarantee disease development, as these conditions are shaped by many genes, environment, and behavior.

?
Why is directly measuring phenotype often more useful than genetic testing?

Directly measuring phenotype (e.g., blood pressure, cholesterol, imaging) provides a real-time, integrated readout of all contributing factors (genetic, environmental, behavioral, age) that is immediately actionable, unlike genetic testing which provides source code without knowing how the program will run.

?
When is genetic testing most useful for cardiovascular and metabolic diseases?

Routine genetic testing is generally weak for these conditions, but it can be useful in exceptions like familial hypercholesterolemia for diagnosis and family screening, or for rare variants like SCARB1 that can misrepresent risk from standard lipid panels.

?
How do consumer genetic tests differ from clinical-grade gene panels for cancer risk?

Consumer genetic tests typically scan for only a few common variants (e.g., three BRCA mutations), providing false reassurance, whereas clinical-grade gene panels sequence a defined set of genes in depth to detect rare, high-impact mutations relevant to cancer risk.

?
What is the utility of knowing one's ApoE status for Alzheimer's disease?

While not deterministic, knowing ApoE status can be useful for informing emerging therapeutic strategies, prompting more aggressive management of other modifiable risk factors, and facilitating long-term planning for finances and care.

?
Why should one be skeptical of functional medicine genetic panels?

These panels often inflate the importance of common genetic variants (e.g., MTHFR, COMT, detox genes) to justify supplement protocols that lack strong clinical evidence and do not outperform standard clinical care.

?
How does pharmacogenetics provide clinical value?

Pharmacogenetics helps guide medication selection and dosing by predicting an individual's response to drugs, which is particularly useful in situations where treatment is trial-and-error, side effects are severe, or drug metabolism varies significantly.

1. Ask Specific Questions Before Testing

Before undergoing genetic testing, define precisely what you aim to learn, as vague questions often lead to confusion rather than clarity. This ensures the test is targeted and potentially useful.

2. Prioritize Phenotypic Measurement First

For common conditions like cardiovascular or metabolic disease, directly measure biological outputs (e.g., cholesterol, blood pressure, insulin resistance) rather than relying on genetic predispositions, as phenotype offers real-time, actionable information.

3. Evaluate Test Utility by Effect & Actionability

Assess genetic tests based on how large the genetic variant’s effect is on risk and how much knowing about it will actually change your clinical management or behavior. Prioritize tests with high effect size and high actionability.

4. Consider Psychological Impact of Results

Before ordering a genetic test, honestly evaluate if you are mentally prepared for the answer, as information, even if true, can cause significant anxiety without leading to constructive action.

5. Use Clinical-Grade Panels for Cancer Risk

If assessing cancer risk, opt for clinical-grade gene panel testing (e.g., for BRCA1/2, Lynch syndrome genes) rather than consumer SNP tests, which only cover a narrow, potentially misleading, set of variants.

6. Test for Pharmacogenetic Variants

If struggling with medication tolerability or choosing between treatment options, consider pharmacogenetic testing to guide medication selection and dosing, as it can predict drug response and safety.

7. Choose Test Type Based on Question

Match the genetic test type (e.g., single gene, panel, whole exome) to the specific clinical question you are trying to answer, aiming for the narrowest test that reliably provides the necessary information without excessive, confusing data.

8. Consult Experts Before Testing

Before ordering any genetic test, especially those informing medical decisions, discuss with a clinician or genetic counselor to understand what the test covers, what it misses, and what a negative result truly means.

9. Use CLIA-Certified Labs for Clinical Tests

For any genetic test that will inform meaningful medical decisions, ensure the testing is performed by a CLIA-certified laboratory with demonstrated expertise in the relevant genetic area.

10. Understand Negative Results

A negative genetic test result does not automatically mean a clean bill of health; it only indicates no pathogenic variant was found on that specific test and does not override strong phenotypic evidence or family history.

11. Have an Action Plan for Results

For any genetic test, consider “what now?” before ordering it. The clinical value comes from whether the results confirm a diagnosis, identify actionable risk, add context, or inform planning, rather than simply documenting a finding.

12. Be Skeptical of Functional Medicine Panels

Approach functional medicine genetic panels (e.g., MTHFR, COMT, detox, nutrigenomics) with skepticism, as they often inflate the importance of common variants to justify supplement protocols lacking strong clinical evidence.

If we truly had a test that could reliably tell you which diseases you were most likely to develop and exactly how to prevent them, that would be a genuine game changer for medicine.

Peter Attia

The assumption was that once we knew this sequence, we would quickly understand function. That reading the code would tell us, more or less directly, how disease worked and how to prevent or treat it. This is not exactly what happened.

Peter Attia

The genetics shift the probability distribution. It doesn't write the ending.

Peter Attia

The phenotype tells me something that is happening right now, something integrated across all the contributing factors, genetic, environment, behavior, age, everything. It's a real-time readout that can be acted upon.

Peter Attia

A result that is likely to produce fear or confusion without changing screening treatment or planning in a constructive way has real costs.

Peter Attia

A mutation can be biologically interesting without being clinically actionable.

Peter Attia

The biggest mistake I see people making is treating a consumer SNP test as though it were a clinical-grade gene panel. These are fundamentally different tools.

Peter Attia

The test is just the information gathering step. The clinical value comes entirely from what happens next.

Peter Attia

Test with intention. Know what you're looking for, know what you'll do when you find it out, and know what you will do if you don't. Everything else follows from that.

Peter Attia

Genetic Testing Decision Framework

Peter Attia
  1. Determine what exactly you are trying to learn, making the question as specific as possible.
  2. Assess if genetic testing is the best tool for this question, or if measuring the phenotype directly would be easier and more informative.
  3. Consider what you will do differently if you get an answer, evaluating how the test will change your behavior.
  4. Reflect on whether you are mentally prepared for the answer, whether positive or negative, to ensure the information is empowering and not just frightening.
2003
Human Genome Project completion year The first draft was published in 2001, and the project was essentially complete in 2003.
1.5%
Percentage of genome that is protein-coding The vast majority of the genome is non-coding, often referred to as 'junk DNA'.
5%
Percentage of cancers attributable to inherited germline mutations The vast majority (95%) of cancers arise from acquired somatic mutations.
Up to 15 times higher
Increased Alzheimer's disease risk for individuals with two copies of ApoE4 Compared to someone without the mutation, but it is still not destiny.
Roughly 50%
Percentage of Alzheimer's disease patients who do not carry an ApoE gene This highlights that ApoE is a risk factor, not a sole cause.
About 10%
Percentage of Parkinson's disease and ALS cases due to known genetic mutations Most cases are not due to a single clear common genetic risk factor.
Up to 40%
Percentage of population carrying MTHFR variants The high prevalence suggests their average effect is small and often clinically irrelevant.
About 10%
Percentage of population with non-functional CYP2C19 enzyme affecting Plavix activation These individuals cannot convert Plavix into a usable compound and require alternative medication.