The risk of obesity by assessing infant growth against the UK-WHO charts compared to the UK90 reference: findings from the Born in Bradford birth cohort study

BACKGROUND The new growth charts in the UK, the UK-WHO charts, comprise prescriptive data from the WHO standard between two weeks and four years of age. Little is known about the development of obesity risk in normal UK infants, who are necessarily not fed according to the WHO recommendations and do not live in constraint-free environments (the selection criteria of the WHO standard source sample), using the new charts. Here, we investigated infant growth trajectories and traits indicative of childhood obesity using the UK-WHO charts, with the aim to clearly document the implications of adopting the new charts on UK growth monitoring practice. METHODS Mixed effects models were applied to serial weight and length data from 2181 infants (1187 White; 994 Pakistani) in the Born in Bradford birth cohort study to produce curves from 10 days to 15 months of age. Individual monthly estimates were converted to Z-scores and were plotted by sex and ethnic group. The relative risks (RR) of traits indicative of childhood obesity, including high BMI and rapid weight gain, using the UK-WHO charts compared to the previously used UK90 reference were calculated for all infants together and for White and Pakistani infants separately. RESULTS Both ethnic groups demonstrated patterns of growth similar to the UK-WHO charts in length but not in weight. The resulting pattern for BMI was remarkable, with an average gain of 1.0 Z-score between two and 12 months of age. The UK-WHO charts were significantly (p < 0.05) more likely than the UK90 reference to classify BMI above the 91st centile after age six months (RR 1.427-2.151) and weight and BMI gain between birth (one month for BMI) and 12 months of age greater than two centile bands (RR 1.214 and 1.470, respectively). CONCLUSIONS The change to the UK-WHO charts means that normal UK infants risk being diagnosed as being on a trajectory toward childhood obesity. National estimates of obesity will have to be recalculated for previous years to allow longitudinal comparison. The new charts do not allow a focused prevention effort for targeting programmes at infants most at risk of becoming obese, because the use of the 91st or 98th centile on the UK-WHO charts will identify many more infants as being at risk than the same centiles on the UK90 reference. Now more than ever, research is needed to develop a large scale childhood obesity prevention programme which could ideally be integrated with routine infant growth monitoring practice.