Overview of The Drive Episode #392: Genetic Testing — When It’s Valuable, How to Choose the Right Test, and What to Do With the Results
Peter Attia lays out a practical framework for thinking about genetic testing: when it is genuinely useful, when it adds little beyond standard clinical data, how to choose the right type of test, and how to interpret results without overreacting. The main message is that genetics can be powerful, but only when the question is specific, the test matches the question, and the result will change something meaningful.
Core Framework for Genetic Testing
Start with the question
Before ordering any test, ask:
- What exactly am I trying to learn?
- Is genetics the best tool, or would phenotype be more informative?
- If I get the answer, what will I do differently?
- Am I prepared for either a positive or negative result?
Genetics is usually probabilistic, not deterministic
Most genetic findings shift risk rather than determine destiny.
- High-penetrance mutations can strongly predict disease in rare cases.
- For most common conditions, genes influence probability, not certainty.
- Environmental factors, behavior, aging, and chance still matter greatly.
Phenotype often beats genotype
When possible, direct measurement is usually more actionable than inferred risk:
- Blood pressure for hypertension
- ApoB / LDL for lipid risk
- Imaging for coronary disease
- Clinical history and exam for many other questions
More data does not always mean more clarity
Broad genetic testing can create:
- False reassurance
- Incidental findings
- Variants of uncertain significance
- More ambiguity than actionable insight
Where Genetic Testing Is Most Useful
Hereditary cancer
One of the clearest high-value uses of germline testing.
Key examples
- BRCA1 / BRCA2
- Lynch syndrome
- Other high-penetrance cancer predisposition genes
Why it matters
- Can change screening frequency and intensity
- May lead to chemoprevention or prophylactic surgery
- Can guide testing for relatives
Important caution
Consumer tests like 23andMe are not equivalent to clinical hereditary cancer panels. A negative consumer result does not rule out a clinically important mutation.
Inherited cardiac disorders
More useful than in routine ASCVD risk assessment, especially when family history suggests an inherited syndrome.
Most relevant uses
- Arrhythmias
- Cardiomyopathies
- Sudden cardiac death in the family
- Structural heart disease syndromes
Why it matters
Genetic risk may justify:
- More regular screening
- Earlier detection
- Targeted cardiac evaluation even if routine tests were previously normal
Neurodegenerative disease
Useful in selected cases, but emotionally and clinically more complex.
APOE
- Strong common risk factor for Alzheimer’s disease
- Two copies of APOE4 can substantially raise risk
- Still not deterministic
When it may help
- Risk stratification
- Planning and long-term decision-making
- Motivating more aggressive control of other modifiable risks
Rare, high-penetrance cases
Testing may be appropriate in families with:
- Early-onset Alzheimer’s
- Familial Parkinson’s disease
- ALS
- Huntington’s disease
Where Genetic Testing Is Less Useful
ASCVD and metabolic disease
Genetics matter, but phenotype usually gives more actionable information.
Examples
- Cholesterol
- Blood pressure
- Insulin resistance
- Obesity
- Lipoprotein(a)
Main point
For most people, measure the trait directly rather than infer it from DNA.
Exceptions
- Familial hypercholesterolemia
- Rare variants that distort interpretation, such as SCARB1
- Situations where genetic confirmation changes family screening or patient behavior
Mental health and complex chronic conditions
Genetic contributions exist, but current testing usually doesn’t guide treatment well.
- Risk estimates are modest
- Predictive power is limited
- Standard clinical care remains more useful than most genetic panels
The Problem With “Functional” and Direct-to-Consumer Genetic Panels
Attia is highly skeptical of many tests marketed around detox, methylation, neurotransmitters, or “personalized” nutrition.
Common examples of overhyped genes
- MTHFR
- COMT
- Cytochrome P450 “detox” genes
Why he считает these weak
- Variants are often very common in the population
- Biological relevance is overstated
- Clinical actionability is usually poor
- Claims often outpace evidence
Bottom line
A variant can be biologically interesting without being clinically useful.
Pharmacogenetics: A Stronger Use Case
Pharmacogenetics is one of the more defensible areas for genetic testing because the question is practical: How will a patient respond to a medication?
Examples
- CYP2C19 and Plavix (clopidogrel)
- Some variants prevent proper drug activation
- Alternative antiplatelet therapy may be needed
- HLA-B*58:01 and allopurinol
- Strongly associated with severe hypersensitivity risk
- Testing can be standard of care in appropriate settings
Why it works better
- Specific clinical question
- Clear action threshold
- Often directly changes treatment choice or dosing
How to Choose the Right Genetic Test
Match the test to the question
Not all tests are interchangeable.
Single-gene or single-variant testing
Best when you already know the exact familial mutation.
- Example: testing for a known BRCA mutation in a relative
SNP genotyping / consumer tests
Good for:
- Ancestry
- Some trait estimates
Poor for:
- Clinical diagnosis
- Rare disease mutations
- Cancer risk assessment
Polygenic risk scores
Interesting, but still early.
- Can capture population-level signal
- Often not very useful alone for individual decision-making
Gene panels
Often the best choice for a focused clinical question.
- Hereditary cancer panels
- Cardiac panels
- Pharmacogenetic panels
Whole exome / whole genome sequencing
Best for:
- Rare, unexplained, complex cases
- When a panel is negative but suspicion remains high
Downside:
- Much more data
- More incidental findings
- More uncertain interpretation
How to Interpret Results
A negative result is not always “all clear”
A negative test only means no pathogenic variant was found on the specific test ordered.
It does not:
- Override strong family history
- Exclude variants the test did not cover
- Eliminate disease risk from non-genetic causes
Four broad result categories
-
Confirms a suspected diagnosis
- Helps solidify management and family screening
-
Reveals a new actionable risk
- May change screening or prevention
-
Adds context without changing management
- Useful, but not necessarily practice-changing
-
Identifies a risk without a clear action plan
- Common in dementia risk and similar areas
- May still help with planning and psychological preparation
The key follow-up question
After any result, ask: What now?
If there is no change in screening, treatment, family planning, or behavior, the test may have limited value.
Practical Takeaways
Best uses of genetic testing
- Hereditary cancer syndromes
- Selected inherited cardiac conditions
- Specific pharmacogenetic decisions
- High-penetrance familial neurodegenerative syndromes
Weakest uses of genetic testing
- Broad “health optimization” SNP panels
- Detox/methylation personality-style interpretations
- Overpromised diet or supplement guidance
- Routine prediction of common chronic disease when phenotype is already measurable
Best mindset
- Test with intention
- Choose the narrowest valid test for the question
- Use clinical-grade, preferably CLIA-certified labs
- Understand privacy implications
- Know in advance what a positive or negative result would change
Final Message
Peter Attia’s central point is that genetic testing is a tool, not a blueprint. It can be life-changing when used for the right question with the right test, but it can also create confusion, anxiety, or false reassurance when used indiscriminately. The most important principle is simple:
Use genetics only when it improves clarity and changes what you do next.