Death Bars Backcasting
Start from what actually kills people your age, then work backwards to today.
- Difficulty
- Easy
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 90%
Attia calls it the death bars analysis. Rather than starting from interventions, you start from the mortality table. For a non-smoker over 40, roughly 80 percent of deaths come from atherosclerotic cardiovascular disease, cancer, neurodegenerative disease, or metabolic disease such as type 2 diabetes, plus accidental death from falls and automotive accidents. Once the bars are named, the strategy question becomes concrete: how do I avoid or delay these specific things? A second binary sharpens it further. Is it better to live longer with a chronic disease or to live longer without one? The data are unambiguous that extending the disease-free period wins, so the goal becomes pushing onset later — a first heart attack very late in life rather than an early one followed by decades on the verge of the next.
Origin
Extracted from The Tim Ferriss Show
Core principles
- 01You cannot lower a risk you have not enumerated.
- 02A small number of causes account for the overwhelming majority of outcomes.
- 03Delaying onset beats extending survival with the disease.
- 04Prevention has to begin long before the disease is diagnosable.
- 05Accidental death belongs in the list even though it is not a chronic disease.
How to run it
- 1
Pull the mortality data for your cohort
Look at what actually ends life for people of your age and smoking status rather than at what feels dangerous or gets media coverage.
Watch out Perceived risk and actual risk diverge badly; Attia notes heart disease kills 8 to 12 women for every one who dies of breast cancer.
- 2
Name the bars
For an over-40 non-smoker the bars are atherosclerotic cardiovascular disease, cancer, neurodegenerative disease, metabolic disease, and accidental death. Those five cover roughly 80 percent of outcomes.
Pro tip Write the list somewhere you will see it; it is the filter for everything downstream.
- 3
Resolve the onset-versus-survival binary
Ask whether you want a longer life containing the disease or a longer life before the disease. The data favour delaying onset, which reframes every intervention as a delay mechanism.
- 4
Work backwards to today's levers
For each bar, identify the modifiable inputs that move it — ApoB and blood pressure for cardiovascular, early screening for cancer, exercise for neurodegenerative, glucose regulation for metabolic.
Pro tip One lever often serves several bars at once, which is how exercise ends up dominating.
Watch out Do not wait for a randomized trial on each lever; most of these trials will never be run.
- 5
Add the cheap accident controls
Accidental death is a bar too. Attia cites Ed Thorp, in excellent shape past 90, who plans his travel specifically to minimize time spent in automobiles.
Pro tip Look for the low-cost controls first; they are almost free relative to their effect.
In the wild
Ferriss recounts having the investor Ed Thorp on the podcast — around 90 years old and in excellent shape. Thorp had run the numbers on automobile risk and planned his travel specifically to minimize time in cars. Attia calls automotive time a very underappreciated cause of morbidity and mortality and says he thinks about it a lot. It is death-bars reasoning applied to the one bar most longevity content ignores entirely.
→ A meaningful mortality risk gets reduced by an itinerary decision rather than a medical intervention.
Common mistakes
Optimizing for the wrong killer
Attention follows publicity rather than mortality. Attia's pink-ribbon versus red-ribbon comparison makes the point that the largest bar often gets the least attention.
Ignoring accidents
Falls and car time are excluded from most longevity plans even though they sit in the same 80 percent as the chronic diseases.
Aiming to survive rather than delay
Framing the goal as surviving the disease longer rather than getting it later inverts the strategy and pushes effort into late-stage treatment.
Is it for you?
Best for
Adults over 40 who want their prevention effort concentrated where the mortality actually is.
Not ideal for
Acute or already-diagnosed conditions, where treatment rather than population-level backcasting drives decisions.
From the transcript
“we call this the death bars analysis so what are the death bars well it turns out if you're over 40 you don't smoke we…”
“it's better to get cancer later in life than to live longer with cancer as one example it's better to have your first heart attack…”
From the episode
#661: Dr. Peter Attia — The Science and Art of Longevity, Optimizing Protein, Alcohol Rules, Lessons from Glucose Monitoring with CGMs, Boosting Your VO2 Max, Preventing Alzheimer's Disease, Early Cancer Detection, How to Use DEXA Scans, Nature’s Longevity Drug, and More