Body shape stories fail to kill off myths

"Pear-shaped women aren't protected from heart disease," the Daily Mail warns, going on to say that a new study "overturns 'myth' that it's better to have wobbly thighs than a fat stomach."

Despite the headline, the study they are reporting on did not look directly at women's (or men's) body shapes, but was in fact designed to describe the levels of specialised proteins called adipokines.

These are important for controlling fat cell development and breakdown, and are of interest to researchers trying to understand why obesity and diabetes are so closely linked.

The researchers compared levels of these proteins in blood and fat taken from the buttocks of people divided into two groups – those who were at risk of diabetes and heart disease, and those who were not at risk of developing these conditions.

One of the main findings of the study was that people with higher levels of fat in their buttocks (gluteal fat) were more likely to be at risk of diabetes and heart disease.

But, based on this study alone, it is too soon to say whether the pear-shaped myth has been shown to be "pear-shaped", as the papers put it, because only a potential link was demonstrated in the study.

For the moment, aiming to lose excess weight, however it is distributed throughout the body, and developing healthier eating habits is the best way to reduce your risk of developing these conditions.

Where did the story come from?

The study was carried out by researchers from the Laboratory of Atherosclerosis and Metabolic Research at the University of California, Davis and other academic institutions in Texas and Tennessee. It was supported by a grant from the American Diabetes Association.

The study was published in The Journal of Clinical Endocrinology and Metabolism, a peer-reviewed medical journal.

Aside from the misleading headline and picture of a pear, the Daily Mail reported and explained the background and findings of the study accurately.

But as the researchers didn't take fat samples from anywhere other than around the bottom, it is not possible to overturn a "myth" from this single study, as the papers claim.

The reporting of the pear-shaped health "myth" also seems to be slightly skewed by the Mail. Having a pear-shaped body is not said to protect you against heart disease, but is apparently less likely to put you at risk of heart disease than people who are apple-shaped (fat around the waist) – but the risk is still there.

What kind of research was this?

This was a cross-sectional study that aimed to measure a selection of signalling proteins secreted by fat cells that circulate in the blood in two groups of patients. These groups were people with metabolic syndrome and a control group of people without the syndrome.

The signalling proteins the researchers were interested in – adipokines, cytokines and chemokines – belong to a family of small molecules that regulate the development of a variety of cells, including fat cells. For example, one protein called chemerin has been linked to both obesity and diabetes.

Cross-sectional studies are good for looking at new theories for the development of disease, but as they do not follow people up over time they cannot prove that one thing leads to another. For example, it is not clear from this study alone whether there are some risk factors (circulating fats, for example) that determine cytokine production, or, conversely, are themselves controlled by it. Other studies are needed to look at how this study could be translated into new treatments.

What did the research involve?

Two groups of patients were formed. One group consisted of 45 participants with metabolic syndrome, as defined by US National Cholesterol Education Program criteria (see box).

The other group was a control group of 30 individuals with two or fewer features of metabolic syndrome, who were not taking any blood pressure medication and did not have high fasting glucose levels or fats (triglycerides).

Neither group had diabetes or were on any anti-inflammatory, hypolipidaemic or hypoglycaemic drugs that could affect blood test results.

The researchers selected participants (matched them) who were similar in terms of gender and age within a range of 10 years.

Researchers took bloods to measure a range of routine tests, including lipid profiles. They also estimated insulin resistance calculated from glucose and insulin levels, and took more blood to record baseline levels of the signalling proteins they were interested in.

They then took small samples (biopsies) of subcutaneous fat cells and fluid (about 4-6mls) from the gluteal or buttock region, a relatively easy place to get the sample.

The circulating levels of cytokines and those found in the fat samples were compared to see if these were different among people with or without metabolic syndrome.

The researchers adjusted for age, body mass index and waist circumference, all things that might independently have influenced the results.

What were the basic results?

More women than men were recruited to the metabolic syndrome group (23 women and 7 men). This was similar to the proportion in the control group (36 women and 9 men). The average age was about 50 years in both groups. Waist circumference was higher in the metabolic syndrome group (108cm) compared with the control group (92cm).

Circulating chemerin was higher in the blood of people in the metabolic syndrome group than in those in the control group, and was also higher in subcutaneous fat samples taken from the buttocks. The significant difference was still apparent after adjustment for body mass index, waist circumference and age.

In contrast, the levels of another chemical – omentin-1 – were lower in both groups. No significant differences were observed in the levels of other circulating proteins, such as visfatin and resistin (proteins found in fat).

How did the researchers interpret the results?

The researchers say that they detected abnormal levels of circulating and gluteal fat-secreted
chemerin and omentin-1 levels in the sub-set of patients with metabolic syndrome.

They go on to say that these abnormal levels could explain a higher risk of developing diabetes and cardiovascular disease in these patients.


This study was a well-conducted cross-sectional study that has been over-interpreted by the media. This was not a study that aimed to compare women with different distributions of body fat and their risk of developing heart disease.

The study has its strengths, in that it was carefully conducted and designed to test and describe a particular link of scientific interest. However, it is hard to see how the design or results can say anything about the risk of diabetes or vascular disease in women who are apple-shaped (fat distributed around the waist) compared with those who are pear-shaped (fat distributed around the hips) for several reasons:

  • The participants were selected if they had a higher waist circumference (apple-shaped) and then had the fat around their hips and buttocks sampled. The fat was sampled from the same place in both groups.
  • Participants were not followed-up over time, and so it is not possible to conclude from this study that the links shown will lead to the development of risk factors or disease in the future.
  • There may be other chemicals that were not measured in this study that explain part of the association seen. Also, the researchers themselves describe other sources of chimerin unrelated to fat that may have influenced the results. 

Overall, this study has highlighted the role of chemerin as a biomarker for other risk factors or vascular disease, but its usefulness in comparison with other measures of risk will need further study.

This study does not provide any compelling evidence that one type of body shape is better than another – most experts would maintain that excess fat is bad for your health, whether it is in your belly, bum or thighs. 

NHS Attribution