Peter Attia's Guide to Reading Your Lab Results

Picture a dashboard with no warning lights on. Most drivers take that as proof the car is fine. Peter Attia says patients do something similar with blood work: if nothing on the printout is flagged in red, they assume, in his words, 'all systems are go' (04:40). It is an understandable leap. Reference ranges look authoritative. Numbers feel objective. And for most adults, a lab draw is the only routine contact they have with the medical system, so it absorbs all the weight of a much bigger question. That question, Attia says, is rarely spoken aloud, but it sits underneath nearly every panel a patient hands him: am I healthy?

The trouble, he argues in this Ask Me Anything episode, is that bloodwork was never built to answer that question completely, and treating it as a report card oversells what it can do. Reference ranges mark statistical boundaries, not guarantees of wellness. A value inside the range is not proof that an organ system is thriving, just that it has not crossed a threshold that labs happen to flag. Attia says this gap has only widened as direct-to-consumer testing companies have multiplied the number of markers on offer. More markers can look like more insight, but volume is not the same as meaning. A large direct-to-consumer panel testing lots of labs does not give more clarity if none of those numbers can be placed in context (05:02).

So Attia puts a number on the gap. At the end of a first lab review with a new patient, he says he now tells them something close to this:

We will know roughly a third of what we care to know about you, maybe, maybe, maybe a bit less, maybe twenty-five percent. [05:29] — Peter Attia

That twenty-five to thirty percent is real and worth having. But three-quarters of the picture, by his own accounting, comes from somewhere else entirely. Figuring out where requires stepping back from the lab printout and asking a more basic question: what does healthy even mean?

What Health Actually Means

Attia's answer depends on who is asking. A patient who has just finished chemotherapy defines health narrowly. The relevant question for that person is whether they are NED, meaning no evidence of disease: has the cancer been cleared, have blood counts recovered from a treatment that can devastate the hematopoietic system, the network of bone marrow and blood-forming tissue that chemotherapy often damages (06:50). For that patient, a handful of numbers might answer nearly everything that matters in the moment.

But most of the people Attia sees are not managing an acute illness. They are trying to live long and live well, and for that population he proposes a cleaner definition: longevity is the product of lifespan and healthspan. Lifespan is how long someone lives. Healthspan is the quality of that life while it lasts. Neither one alone is sufficient.

Living a long time in poor health, that's, that's not optimal. And having a short life in high health is also not optimal. So we, we want both of those things. [07:55] — Peter Attia

On the lifespan side, Attia says the threats are a short and fairly stable list: cardiovascular disease, cancer, neurodegenerative and dementing diseases, metabolic disease, and accidental death. He adds that organ-specific conditions, chronic obstructive pulmonary disease in longtime smokers, kidney disease, liver disease unrelated to metabolic dysfunction, matter too, but the five big categories cover most of the risk (08:18).

Healthspan, in his framework, has three components, and this is where the real blind spot in lab testing shows up. Physical health means freedom from pain, strength, endurance, and mobility, essentially the ability to do whatever someone wants to do with their body, with the bar set differently for different people's goals (09:03). Cognitive health covers processing speed, short-term and long-term memory, and executive function, the brain's capacity to plan, focus, and follow through. Emotional health includes a sense of purpose, happiness, the strength of relationships, and what Attia calls distress tolerance, the capacity to absorb hardship without being derailed by it (09:26). None of that shows up in a vial of blood. A test can flag inflammation or cholesterol, but it cannot measure whether someone can still hike a mountain, hold a thought, or recover from a bad week. That is the other seventy-five percent, and it has to be assessed by entirely different means, through physical function testing, cognitive assessment, and honest conversation about mood and purpose.

The Trouble with One Number

Even within the slice that labs do cover, Attia says a single snapshot is a weak foundation for a decision. The first problem is reliability. Without longitudinal data, meaning a history of results over time, there is no way to rule out a plain laboratory error. The second problem, which persists even after an abnormal result is confirmed by a repeat test, is duration: how long has this actually been true?

He illustrates this with high-sensitivity C-reactive protein, a blood marker of inflammation. A reading of three is, in his words, quite elevated (11:49). But a single number cannot say why. Maybe the person is sick that week. Maybe they were vaccinated recently and the marker is still settling, or maybe that elevated three has been their baseline for the better part of a year while their true resting level is closer to half a point. Without the history, a clinician cannot tell whether this is chronic inflammation that demands a treatment plan or a temporary blip that will resolve on its own. Attia says the same logic applies to LDL cholesterol, the lipoprotein that carries cholesterol through the blood and, at high levels, drives plaque buildup in arteries. A value of 160 mg/dL puts someone above the 80th percentile, clearly high by population standards (12:31). But whether that number reflects three months of a saturated-fat-heavy diet or a lifelong genetic pattern changes the entire clinical approach. One suggests a dietary conversation. The other suggests a search for an inherited lipid disorder.

Not every marker needs this kind of repeated scrutiny, though. Attia notes that certain genetic or near-genetic markers, such as Lp(a), a cholesterol particle largely set by genetics, and ApoE, a gene variant linked to cardiovascular and neurodegenerative risk, typically only need to be measured once, assuming the initial test was run correctly (13:01). These are fixed traits, not moving targets, so they do not require the same longitudinal tracking that something like CRP or LDL does.

The through-line across all of this is a kind of discipline about what a single data point can and cannot support. A lab result is a flashlight beam, not a floodlight. It illuminates a narrow patch clearly, but it says nothing about everything outside that beam, and even within the beam, a second look from a different night sometimes changes what is visible. Attia's practical suggestion, implicit throughout the conversation, is to ask three questions of any result: does the pattern repeat, has the context around it been considered, and does it actually speak to one of the things, cardiovascular disease, cancer, cognitive decline, physical capacity, emotional resilience, that determines whether a life is both long and good. Most individual numbers answer a sliver of that. Knowing which sliver, and trusting it accordingly, is the actual skill, more than any single test result ever could be.