April 23, 2018

expert reaction to study looking at caffeine in pregnancy and weight gain in children

A new study, published in BMJ Open, researchers study the association between maternal caffeine intake during pregnancy and the child’s weight gain and overweigh risk up to 8 years.

Dr Thomas House, Reader in Applied Mathematics, University of Manchester, said:

“This study makes the headline claim that in a population of over 50,000 mother-child pairs, caffeine intake is positively associated with risk of childhood obesity.  Let us make a ‘back of the envelope’ calculation on one of the outcome measures using the cup of instant coffee per day as a measure of the mother’s caffeine intake.

“If we take 100 children whose mother group had the lowest caffeine intake (1 cup or less) then at age three, around 11 will be obese; in the next group (1-4 cups) there will be one additional obese child (12 in total); in the next (4-6 cups) there will be an additional two (14 in total) and in the group with highest caffeine intake (6 or more cups) there will be an additional three (17 in total).

“The main problem with the study is that maternal caffeine intake is strongly correlated with other relevant factors, and so these should be controlled for.  In the authors’ statistical analyses, the ‘best fit’ models they’ve used suggest around half of the additional cases of obesity in children of age 3 are explained by caffeine rather than, say, calorie intake of mothers.  But since the caffeine and calorie intake are correlated, the existence of a ‘best fit’ model with equal weighting on both means we actually can’t say whether a model looking at just calorie intake of mothers would be an equally good explanation of child obesity.  It also says nothing about whether including a new explanatory measure, such as the child’s calorie intake, would change the model.

“As such, the evidence provided in the study for a causal effect is extremely weak and the statement from the authors that “complete avoidance might actually be advisable” seems unjustified, particularly when we consider the effects that such a restriction might have on wellbeing of mothers.”

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

“It is well established that accelerated growth rate in early life has long-term consequences for health.  However, this study was not able to determine whether the increase in growth rate in the first year of life was due to an increase in bone and muscle mass, which tends to occur earlier, or body fat which occurs later.

“This study only shows a weak association between caffeine intake during pregnancy and an increased growth rate in the first year of life.  A major limitation is that the study failed to adjust for bottle-feeding or breastfeeding.  This is important because breast-fed babies grow at a slower rate than bottle-fed infants.

“The women with the highest caffeine intakes were older, more likely to be poorly educated or obese prior to pregnancy, and to smoke during pregnancy – all factors associated with a reduced likelihood of breastfeeding their infants.  It may well be that the relationship between caffeine intake in pregnancy and infant growth in the first year of life is spurious.”

Prof Jean Golding, Emeritus Professor of Paediatric and Perinatal Epidemiology, University of Bristol, said:

“This study uses information provided by over 50,000 mothers taking part in the world famous Norwegian MoBA cohort study.  These women provided information on the foods and drinks they were consuming in mid-pregnancy, and on the heights and weights of their children up until 8 years of age.  The researchers have used these data to calculate the amount of caffeine consumed, and compared this with the growth of the children.

“After allowing for other features that can affect the growth of the child, including whether the mother smoked, they have shown that the children of women with high levels of caffeine intake had put on more weight (but not grown taller) than the children of women who had relatively low levels of caffeine.  The findings were particularly associated with excess growth in the first year of life.

“Although this study repeats the findings of two other studies, there are several caveats to be born in mind – the measures of height and weight were reported by the mother and so did not use validated methods.  Nor were there measures that would indicate whether the excess weight found was due to increases in fat as opposed to increases in muscle.  The excess weight found had reduced by age 8, with the exception of the growth of the very high caffeine group.  It will be important to determine whether any effects of high maternal caffeine intake in pregnancy are apparent at later ages, or whether any effects are confined to the prepuberty ages.”

* ‘Maternal caffeine intake during pregnancy and childhood growth and overweight: results from a large Norwegian prospective observational cohort study’ by Eleni Papadopoulou et al. published in BMJ Open on Monday 23 April 2018. 

Declared interests

Dr Thomas House: “In terms of interests, I don’t have any professional or business connections.”

Prof Tom Sanders: “Scientific Governor of the British Nutrition Foundation; I chaired this report: ‘British Nutrition Foundation Task Force on Nutrition and Development – short- and long-term consequences for Health’, Wiley-Blackwell, Chichester 2013.”

Prof Jean Golding: “I do have a conflict of interest in having received funding for research from COSIC (The International Coffee Organisation) and from Coca Cola over 20 years ago.”