Does Being Overweight Shorten Your Life? What the Research Says
In this article
This is one of the most searched health questions there is, and it's also one of the easiest to answer badly. The headlines swing between "obesity is a death sentence" and "BMI is meaningless, ignore it." Neither is true, and the gap between them is where most people live.
The honest version takes a little more patience. Weight and lifespan are related, the relationship is real and well-measured, and it is not a straight line. Very high body weight is linked to shorter life expectancy in the large studies. So, at the extremes, is very low body weight. And BMI (the number most of this research is built on) is a decent tool for populations and a blunt one for any single person.
None of this is about blame. It's about what the data can and can't tell you, so you can make a calm decision instead of a panicked one.
The short answer
Yes, on average, carrying substantially excess weight is associated with a shorter life. In the largest pooled studies, mortality is lowest at a BMI around 22.5 to 25, and rises as BMI climbs above that. The effect is modest in the overweight range and grows steadily into the obese range. At the high end (BMI 40 to 45) the loss of life expectancy is large, roughly comparable to smoking. Very low BMI also carries higher mortality, though much of that reflects underlying illness rather than thinness itself. The encouraging part: the curve is gradual, so modest, sustainable changes move you in the right direction.
Want to see how your weight factors into your own estimate, alongside everything else? Take the free assessment →
What the biggest studies found
Two large collaborative analyses do most of the heavy lifting here, and they agree with each other.
The Prospective Studies Collaboration (Whitlock et al., The Lancet, 2009) pooled 57 studies covering about 900,000 adults, followed for a mean of 13 years. Its central finding: mortality was lowest at a BMI of about 22.5 to 25, and above that range, each 5 kg/m² higher BMI was associated with roughly 30% higher overall mortality.
The more recent Global BMI Mortality Collaboration (Di Angelantonio et al., The Lancet, 2016) went bigger still. To strip out the two things that muddy this research (smoking and pre-existing illness), the authors restricted their main analysis to never-smokers who were free of chronic disease and survived at least five years of follow-up. That left about 3.9 million adults across four continents. Using a BMI of 22.5 to 25 as the reference, they found each 5 kg/m² above the healthy range raised the risk of death by roughly a third (a hazard ratio of about 1.39 in Europe, with similar figures elsewhere).
Higher overall mortality per 5-unit rise in BMI above the healthy range (roughly 22.5 to 25)
Prospective Studies Collaboration, The Lancet, 2009, ~900,000 adults; consistent with Global BMI Mortality Collaboration, The Lancet, 2016, ~3.9 million adults
Broken out by category, the 2016 analysis found the risk climbs gently at first and then more steeply:
- BMI 25 to 27.5 (lower overweight): ~7% higher mortality
- BMI 27.5 to 30 (upper overweight): ~20% higher
- BMI 30 to 35 (obesity grade 1): ~45% higher
- BMI 35 to 40 (obesity grade 2): ~94% higher
- BMI 40 to 60 (obesity grade 3): roughly 2.8x higher
The shape matters as much as the numbers. The lower overweight range carries a small risk increase. The steep part of the curve is in the obese range, not just above "ideal."
How many years are we talking about?
Relative risks are hard to feel. The 2009 collaboration translated its findings into something more concrete: estimated loss of life expectancy.
At a BMI of 30 to 35, median survival was reduced by an estimated 2 to 4 years. At a BMI of 40 to 45, it was reduced by an estimated 8 to 10 years, which the authors noted is comparable to the effect of smoking.
Estimated reduction in median survival at BMI 30–35; roughly 8–10 years at BMI 40–45 (comparable to smoking)
Prospective Studies Collaboration, The Lancet, 2009
Two things to hold alongside those numbers. First, they are population averages, not a verdict on any one person. Second, the gradient runs the other way too: because the curve is gradual, moving from grade 2 obesity toward grade 1, or from grade 1 toward the overweight range, recovers part of the risk. You don't have to reach an "ideal" number to benefit.
The other side of the curve: why very low BMI also carries risk
The BMI-mortality relationship is often described as J-shaped or U-shaped, and that second upswing is real. In the 2009 data, below the 22.5 to 25 range, lower BMI was associated with higher mortality.
But the reason is important, and it's easy to misread. The collaboration found that most of the excess risk at low BMI was driven by strong inverse associations with respiratory disease and lung cancer. In plain terms, a lot of thin people in these studies were thin because they were ill or were smokers, not the other way around. This is called reverse causation, and it's exactly why the 2016 analysis excluded smokers and people with existing disease before drawing its conclusions.
So the takeaway isn't "being underweight is as risky as obesity for everyone." It's that unexplained low body weight can be a signal worth paying attention to, and that the healthy zone is a range, not a single point.
BMI is a population tool, not a personal diagnosis
Here's the part the scary headlines skip. Every number above is built on BMI, and BMI only measures weight against height. It doesn't know the difference between muscle and fat, and it doesn't know where fat sits on your body.
That limitation is measurable, not hypothetical. A NHANES study of over 13,000 U.S. adults (Romero-Corral et al., International Journal of Obesity, 2008) compared BMI to directly estimated body fat. A BMI of 30 or above was highly specific (few false alarms) but had poor sensitivity: it missed a large share of people who had excess body fat by direct measurement (catching only about 36% of affected men and 49% of affected women). BMI can flag a muscular person as "obese" and clear a person with low muscle and high fat as "healthy."
Where fat sits matters too. In the EPIC study of about 359,000 Europeans (Pischon et al., New England Journal of Medicine, 2008), waist circumference and waist-to-hip ratio predicted mortality even after accounting for BMI. People in the highest waist-circumference group had roughly double the risk of death of those in the lowest, and abdominal fat carried risk independent of overall weight. This is why a tape measure around your waist can tell you something your bathroom scale can't.
What this means for you (without the diet-culture noise)
The research supports a few calm conclusions, none of which involve punishing yourself.
Direction beats destination. Because the curve is gradual, you capture real benefit from partial progress. You do not need to hit a "perfect" BMI to lower your risk. Moving one category in the right direction is a genuine win the data recognizes.
Modest and sustainable outperforms dramatic and temporary. The mortality studies measure long-run body weight, not a number you touched for three weeks. A crash diet you can't maintain doesn't show up as a benefit in this literature. A change you can hold for years does.
Weight is one lever, and rarely the first one. For many people, the highest-leverage moves aren't about the scale at all. Regular movement, better sleep, and food quality improve health markers even when weight barely shifts. Our longevity diet guide and the minimum effective exercise dose are both better starting points than any weight target, and both tend to move weight as a side effect anyway.
Measure your waist, not just your weight. Given what the EPIC data showed, waist circumference is a cheap, useful complement to BMI. It's one of the few home measurements that adds real signal.
If you want to see how your weight interacts with everything else (your activity, sleep, diet, and the rest), that's exactly what the free assessment is for. It uses these same published effect sizes, and you can read exactly how in our methodology.
See how weight fits into your full picture
Weight is one of several factors in your estimate. Take the 2-minute assessment to see where it ranks for you, and which changes would move the needle most.
Take the Assessment →Frequently asked questions
How many years does obesity take off your life?
It depends heavily on the degree. In the Prospective Studies Collaboration analysis of ~900,000 adults, a BMI of 30 to 35 was associated with an estimated 2 to 4 years of reduced median survival, while a BMI of 40 to 45 was associated with 8 to 10 years, comparable to the effect of smoking. These are population averages, not predictions for any individual, and the loss is smaller nearer the healthy range.
What is the BMI with the lowest mortality?
Across both large collaborative analyses, all-cause mortality was lowest at a BMI of roughly 22.5 to 25. Risk rises as BMI climbs above that range, and also rises below it, though the low-BMI increase is largely explained by underlying illness and smoking rather than thinness itself.
Is being slightly overweight dangerous?
The risk in the lower overweight range is small. In the 2016 Global BMI Mortality Collaboration, a BMI of 25 to 27.5 was associated with about 7% higher mortality versus the healthy reference range, rising to about 20% at 27.5 to 30. The steep increases begin in the obese range. Slightly overweight is a gentle nudge on the curve, not a cliff.
Is BMI a reliable measure of my personal health?
Not on its own. BMI measures weight against height and can't distinguish muscle from fat or account for where fat is stored. A 2008 NHANES analysis found BMI missed a large share of people who had excess body fat by direct measurement. Waist circumference adds independent information, and in the EPIC study it predicted mortality even after adjusting for BMI. Use BMI as one signal among several, not a diagnosis.
Do I have to reach a "normal" BMI to benefit?
No. Because the relationship between BMI and mortality is a gradual curve rather than a step, partial progress captures real benefit. Moving from a higher obesity category toward a lower one, or from obese toward overweight, is associated with meaningfully lower risk. Sustainable direction matters more than hitting a specific target.
Can you be too thin?
Very low BMI is associated with higher mortality in the large studies, but most of that excess risk traces back to respiratory disease, lung cancer, and smoking rather than low weight being harmful in itself. That said, unexplained low body weight can be a signal worth discussing with a doctor. The healthiest zone is a range, not a single low number.
Citations
- Global BMI Mortality Collaboration; Di Angelantonio E, Bhupathiraju SN, Wormser D, et al. Body-mass index and all-cause mortality: individual-participant-data meta-analysis of 239 prospective studies in four continents. The Lancet. 2016;388(10046):776-786.
- Prospective Studies Collaboration; Whitlock G, Lewington S, Sherliker P, et al. Body-mass index and cause-specific mortality in 900,000 adults: collaborative analyses of 57 prospective studies. The Lancet. 2009;373(9669):1083-1096.
- Romero-Corral A, Somers VK, Sierra-Johnson J, et al. Accuracy of body mass index in diagnosing obesity in the adult general population. International Journal of Obesity. 2008;32(6):959-966.
- Pischon T, Boeing H, Hoffmann K, et al. General and abdominal adiposity and risk of death in Europe. New England Journal of Medicine. 2008;359(20):2105-2120.