In this “Ask Me Anything” (AMA) episode, Peter explores how to think critically about medications and supplements by focusing not on whether an intervention is inherently “good” or “bad,” but on whether it makes sense for a specific person with a specific problem. He explains why clearly defining the problem matters more than choosing the intervention itself, how the intended purpose of a medication or supplement should influence the standard of evidence required, and why mechanistic reasoning alone is rarely enough to justify taking something. Peter also examines how baseline risk shapes the true benefit of an intervention, why relative risk statistics can be misleading without proper context, and how to weigh not only side effects, but also cost, inconvenience, and opportunity cost when deciding whether something is worth taking. Additionally, he discusses practical ways to evaluate whether a supplement is actually having a meaningful effect, how to think about discontinuing therapies, why supplements deserve far more skepticism than they often receive, and the small group of over-the-counter supplements he believes may offer a reasonable risk-reward trade-off.
If you’re not a subscriber and listening on a podcast player, you’ll only be able to hear a preview of the AMA. If you’re a subscriber, you can now listen to this full episode on your private RSS feed or on our website at the AMA #85 show notes page. If you are not a subscriber, you can learn more about the subscriber benefits here.
Would you like access to this entire podcast episode, its extensive show notes, and more premium content?
We discuss:
- How to properly define health problems before considering medications or supplements [1:45];
- How the intended purpose of an intervention should determine evidence standards and risk tolerance [5:00];
- Understanding the hierarchy of evidence for medications and supplements and avoiding the mistake of treating weak evidence as clinical proof [9:00];
- Why mechanistic explanations can be misleading when evaluating longevity interventions [13:15];
- How baseline risk—and the distinction between relative and absolute risk reduction—changes the real-world benefit of medications and supplements [18:15];
- Thinking beyond side effects: the many forms of downside associated with medications and supplements [22:45];
- Why medications and supplements require different standards of trust and evidence [26:00];
- How to structure meaningful self-experiments with medications and supplements to determine if it’s working [30:30];
- How to monitor the effects of medications and supplements without fooling yourself [32:30];
- How to periodically reevaluate and potentially discontinue medications and supplements [35:15];
- The biggest risks and failure modes of over-the-counter supplements: efficacy, poor quality control, contamination, interactions, toxicity, and marketing-driven overuse [38:30];
- Why the US supplement regulatory system creates unreliable products [41:45];
- A practical framework for evaluating medications and supplements [46:30];
- Over-the-counter supplements with the best balance of evidence and low downside risk [48:00]; and
- More.
Show Notes
How to properly define health problems before considering medications or supplements [1:45]
Defining the problem before choosing an intervention
Most people define the problem too vaguely
- Peter argues that one of the biggest mistakes people make with supplements and pharmaceuticals is defining problems at the wrong level of abstraction.
- Common examples of vague problem definitions:
- “I want to be healthier”
- “I want more energy”
- “I want better longevity”
- These goals are:
- difficult to measure
- difficult to falsify
- not actionable
- prone to subjective interpretation
A better framework: define actionable metrics
- Peter recommends framing health problems around three specific elements:
- 1) A measurable metric
- Something objective that can actually be tracked
- Ideally something that can be repeatedly measured over time
- 2) A threshold or target
- A specific number or outcome that defines success
- Clarifies what level represents a meaningful problem or improvement
- 3) A time horizon
- Defines when improvement should reasonably occur
- Prevents endless, open-ended experimentation
- Example: cholesterol / ApoB
- Bad framing:
- “My cholesterol is bad”
- Better framing:
- “My ApoB is 130 mg/dL”
- “I want it below 60 mg/dL”
- “I want to accomplish that within six months”
- Bad framing:
- Example: sleep
- Bad framing:
- “My sleep is bad”
- Better framing:
- “It takes me 60 minutes to fall asleep on most nights”
- “I want sleep onset to be under 10 minutes”
- “I want that improvement within two months”
- Bad framing:
The importance of the counterfactual
Ask: “What happens if I do nothing?”
- Peter emphasizes evaluating whether the problem:
- meaningfully increases risk
- reduces quality of life
- creates downstream consequences
- This matters because it helps:
- separate genuine problems from things that simply feel actionable
- prevent unnecessary interventions
Poor problem definition leads to false positives
“Poor problem definition almost guarantees some sort of false positive.” —Peter Attia
Humans are “storytelling machines”
- Peter explains that vague goals make it easy to falsely believe something worked.
- Examples:
- slightly better energy on a random day
- one improved lab value
- one better night of sleep
- Consequences
- People may:
- continue supplements for years
- spend large amounts of money
- expose themselves to unnecessary risk
- without ever knowing whether the intervention truly helped
- People may:
Core principle: start with the problem, not the molecule
Peter’s main rule: Do not start with the molecule, start with the problem.
- A proper intervention decision should include:
- a metric
- a threshold
- a timeline
- consequences of doing nothing
- If those elements are absent, the person is likely:
- making impulse-driven decisions
- responding to marketing
- engaging in “supplement shopping” rather than making evidence-based intervention choices
How the intended purpose of an intervention should determine evidence standards and risk tolerance [5:00]
Classifying the “job” of an intervention
Why classification matters
- Peter argues that once the health problem is clearly defined, the next question becomes:
- “What is the purpose of the intervention?”
{end of show notes preview}
