select search filters
roundups & rapid reactions
factsheets & briefing notes
before the headlines
Fiona fox's blog

expert reaction to study looking at body mass index, health and death (investigating ‘fat but fit’)

In a Norwegian population-based health cohort study, published in the International Journal of Epidemiology, scientists measured the effects of high and low BMI in 60,000 parents and children.


Prof. Tom Sanders, Professor emeritus of Nutrition and Dietetics, King’s College London, said:

“Overweight is currently defined as a BMI of greater than 25, and obesity greater than 30.  There has been much debate about where the nadir is for BMI in relation to all cause mortality.

“This study was conducted in Norway, an affluent country with low rates of obesity and with relatively little ethnic diversity.  The study confirms the known trend for increasing risk of death from cardiovascular disease (CVD) with increasing BMI especially in men – the study split the range of BMI into quintiles with the averages ranging from about 21-29.  Previous studies show that risk of CVD above a BMI of 22.5 in men and above about 26 in women.  What is novel about this study is that they have adjusted the risk estimates in the parents by using their offspring’s BMI as a proxy for inherited effects.  However, environment in early life (poor nutrition and childhood illness) can a have major impact on attained height and it is not possible to rule out residual effects of poverty.  A further limitation is that there are differences in risk according to ethnic groups.  For example, men of black African descent, who tend to be more muscle bound, have a lower risk of coronary heart disease compared to white European descent with similar BMI.

“There is plenty of evidence to show that even modest increases in BMI from about 21 increase serum cholesterol, blood pressure and risk of developing type 2 diabetes.  My main concern is that this paper should not be interpreted as suggesting that being lighter in weight than currently recommended is healthy.

“From previous evidence, the relationship between BMI and mortality has been shown to be J-shaped and not linear as this study shows.  A low body mass index (less than 20) in middle-aged and older adults is associated with a substantial increase in non-cardiovascular disease mortality.  This effect is also seen in health conscious groups such as vegans and vegetarians.  While we have a major epidemic of obesity, we have a parallel epidemic of younger people trying to achieve unhealthily low body weights.”


Prof. Nick Finer, Honorary Clinical Professor, National Centre for Cardiovascular Prevention and Outcomes, UCL, said:

“There continues to be a scientific argument as to whether modest overweight may be protective against ill health, and even death, based upon a number of epidemiological studies.  One of the main arguments against this association is that ill health can result in weight loss so favouring those who have not lost weight and may be modestly overweight – so called reverse causation.  Carslake and his colleagues have used a novel technique to try and get around this problem.

“In a large prospective Norwegian cohort, they have intriguingly explored the impact of children’s body mass index (BMI) on their parents’ health.  There is known to be a strong correlation of body weight and BMI between parents and their offspring due to heredity and a shared environment, so using the child’s BMI as a proxy for their parents’ BMI can be used as an instrument unbiased by reverse causality.  There was a clear relationship between an increased BMI in the parents, and their offspring BMI with death from all causes, cardiovascular disease, diabetes and cancer and argue that this supports a direct cause and effect relationship.  For parents with a low BMI there was an inverse relationship with these diseases and mortality, but this connection was lost if the children’s BMI was used as a proxy for parental BMI.

“The conclusion that higher mortality at higher BMI is a cause and effect relationship, while the higher mortality at lower BMI is exaggerated by confounding factors such as ill health is supported by other recent studies.  The optimal weight in this study was fully in line with health recommendations to maintain a BMI between 21-25 kg/m2 and further refutes the mistaken idea that it is healthy to be overweight.

“The study uses complex statistical techniques and relies on several assumptions that the authors acknowledge are limitations.  However, the size of the large prospective dataset of more than 60,000 parent-offspring pairs gives the findings strength and robustness.”


* ‘Confounding by ill health in the observed association between BMI and mortality: evidence from the HUNT Study using offspring BMI as an instrument’ by David Carslake et al. published in the International Journal of Epidemiology on Friday 1 December 2017.


Declared interests

Prof. Nick Finer: “I am an employee of Novo Nordisk but have no conflicts of interest in relation to this research; the views expressed are personal.”

in this section

filter RoundUps by year

search by tag