Danish study finds being very thin may be deadlier than being overweight

Being modestly overweight may not reduce life expectancy, whereas being extremely thin could do so.

A major Danish study that followed more than 85,000 adults found that people with a BMI under 18.5 were almost three times as likely to die prematurely as people at the middle to upper end of the so-called "healthy" range.

The relationship between weight and health is less straightforward than is often presumed. This new, not-yet peer-reviewed research indicates that the lowest mortality risk may not fall precisely within the conventional "healthy" body mass index (BMI) category.

Instead, the results indicate that people whose BMIs would usually be labelled "overweight" may have outcomes as favourable as, or better than, those of people with lower BMIs.

BMI and mortality risk in the Danish study

When the researchers mapped BMI against mortality, they identified a U-shaped pattern: people with both the lowest and highest BMIs faced the greatest risk of death.

The data, presented in a conference paper at the Annual Meeting of the European Association for the Study of Diabetes, showed that being underweight posed the greatest risk. Those with a BMI below 18.5 were almost three times more likely to die early than participants with a BMI of 22.5 to 24.9.

People at the bottom of the "healthy" range were also at greater risk. A BMI of 18.5 to 19.9 was linked to twice the likelihood of death, while a BMI between 20 and 22.4 was associated with a 27% higher risk of early death than the reference group.

This may appear unexpected, as a BMI from 18.5 to 24.9 is generally regarded as ideal.

At the other end of the range, however, additional weight did not consistently mean a higher risk. Participants with BMIs from 25 to 35, usually classified as "overweight" or "obese", had no significant increase in mortality relative to the reference group.

A substantial increase was seen only among people with a BMI of 40 or above, whose risk of death more than doubled (2.1 times).

These results provide further evidence against the widespread social assumption that thinness automatically means health. Research has shown, however, that underweight can endanger health, particularly in later life.

Why being underweight can affect health

Some fat stores can support the body during illness. For instance, people receiving cancer treatment, including chemotherapy, often lose weight because of issues such as reduced appetite and altered taste.

People who begin treatment with greater fat reserves can use those stores, allowing their bodies to maintain essential processes. By contrast, a person with very little body fat may exhaust their reserves rapidly, reducing the body's capacity to recover.

Unplanned weight loss can also signal illness. Conditions including cancer and type 1 diabetes commonly cause weight loss before they are diagnosed. Consequently, a low BMI may sometimes indicate an underlying disease.

Not surprising

In response to the researchers' conference paper, headlines have appeared including: Being too thin can be deadlier than being overweight, Danish study reveals.

That may seem startling, but it should not. Humans need food to live, and without it we die. This has been understood for hundreds, if not thousands, of years.

When food is unavailable, the body enters a catabolic state and breaks down tissue to obtain the energy required to keep the brain working. During this process, other vital functions, including immune function, are deprioritised so that energy can be directed to the brain.

It is important to note that every Danish participant had received a body scan for health-related reasons. As these scans are expensive, they are generally performed for a clear reason – when a health problem is suspected.

The researchers recognise that participants may have been losing weight because of an underlying condition. Therefore, it may be the illness itself, rather than the weight loss associated with it, that raises the risk of death.

Nevertheless, the findings support suggestions from other studies: thinness is not invariably protective, and extra weight is not invariably damaging. The idea that a person can be "fat but fit" is gaining further scientific support.

Should the "healthy" BMI range therefore be moved upwards? The researchers raise this possibility, arguing that modern medical advances enabling people to manage obesity-related conditions, including diabetes and heart disease, may be moving the safest weight range higher than it once was.

For the Danish population examined, a BMI of 22.5 to 30 may now be associated with the lowest mortality risk.

A blunt tool

The difficulty is that BMI has always been an imprecise tool, as I have argued previously. It does not account for several factors relevant to health, including diet, lifestyle and fat distribution.

BMI may also give a misleading picture for people from different racial, ethnic and cultural backgrounds. Critics argue that standard thresholds are based on white body types, meaning entirely healthy bodies in other groups may be judged "unhealthy".

Indeed, BMI was created almost two centuries ago from data drawn from a small group of white European men.

Although attempts have been made to adjust ranges for some ethnic groups – for instance, NHS guidance sets lower BMI thresholds for increased diabetes risk in Asian and black groups – BMI still does not capture differences in body composition, fat distribution and baseline risk between people in our diverse society.

Where major healthcare decisions, including eligibility for fertility treatment and some operations, depend on BMI, it should be an accurate and equitable measure that has been developed and validated in populations genuinely representative of those to whom it is applied.

Ideally, healthcare professionals would be able to use more detailed assessments, such as blood tests, imaging scans and comprehensive lifestyle information. These approaches are expensive and time-consuming, but they provide far more insight than a height-to-weight ratio can offer.

BMI will remain in use until better measures are broadly accessible, but studies such as this highlight the need to improve its interpretation.

The Danish findings remain preliminary. Further information and research are required before firm conclusions can be reached. Yet the central message remains: being extremely thin is dangerous, while carrying some additional weight may not shorten life. The key point is not that thin is bad and fat is good, but that BMI on its own is a fragile indicator of health.

Rachel Woods, Senior Lecturer in Physiology, University of Lincoln

This article is republished from The Conversation under a Creative Commons license. Read the original article.