Predictors of aberrant eating and child BMI percentile

Childhood obesity is a particular concern because obese children
tend to remain obese in adulthood. Therefore, identifying relevant
environmental pressures should be given high priority. In the
present study we explored the relative importance of four contributory
factors: (i) the proportion (%) of meals eaten in combination
with TV, (ii) preference for high energy-dense snack foods, (iii) selection
of larger portion sizes and eating rate, and (iv) variability
in meal patterning and snacking. In this context we introduced novel
computer-based assessments and extended previous studies by considering
reports from parent–child dyads (N = 218). Our sample
(range 5–11 years) were recruited from an NIHR randomised controlled
trial (n = 69) and from a local interactive science centre
(n = 149). Despite the selection of a large sample with a wide range
of BMI percentiles (SD = 27.54, range = 1–99) we failed to replicate
a number of previous observations. For example, we failed to establish
a clear relationship between degree of distracted eating and
BMI percentile. However, children who never ate in combination
with distraction were leaner, suggesting that distraction may not
play a direct causal role. Similarly, eating rate did not predict BMI
percentile but was associated with parent estimates of child’s portion
sizes. These and related findings will be discussed, together with
the relative merits of obtaining parent and child reports using our
tools and techniques. This research was funded by a BBSRC grant
(ref: BB/I012370/1) and an NIHR HTA award (ref: 09/127/04). Recruitment
was supported by At-Bristol science centre.