Background Although childhood overweight and obesity rates have been falling in England in recent years, they remain an important public health concern because of the association between adolescent obesity and increased ill health risks in adult life including an increased risk of mortality from all causes independent of adult weight. Early intervention is important since behaviours contributing to obesity, higher energy diet and increased sedentary behaviour, are also prevalent during early childhood. We have conducted an epidemiological analysis of routine data collected at birth to examine the factors associated with overweight and obesity in children aged 10-11 years.

Methods Administrative records of children in aged 10-11 years, measured between September 2006 and July 2010, as part of the National Child Measurement Programme in a large area on the South East coast of England were examined. Of the 19,924 children in the study with a valid height and weight measurement, 29% were overweight/obese. Factors identified from the literature as potential predictors of childhood overweight or obesity were extracted, where recorded. These are age; birth weight; gender; maternal smoking; previous live births; multiple birth; mother’s age; and Index of Multiple Deprivation quintile.

Results Factors independently associated with an increased risk of overweight or obesity included deprivation (OR 1.6 [95% CI 1.4, 1.9]) (children born in most deprived areas compared with least deprived), regular maternal smoking during pregnancy (1.3 [1.2, 1.4]) and high brith weight (1.5 [1.2, 1.9]). Low birth weight (0.8 [0.7, 0.9]), being female (0.8 [0.8, 0.9]) and maternal age <20 years (0.7 [0.6, 0.9]) were independently associated with a reduced risk of childhood overweight/obesity. There was a significant participation bias by gender, with girls less likely to be measured which may explain the apparent protective effect for girls. This finding could impact on the results of the NCMP nationally if also found elsewhere. The predictive model, whilst significant, did not discriminate well at the population level. The limited set of data recorded before 2003 restricted the factors which could be included in the model. With better recording of factors such as breastfeeding and ethnicity there is potential to enhance the model in the future.

Conclusion Factors associated with childhood overweight and obesity at age 10-11 were neighbourhood deprivation, regular maternal smoking during pregnancy and high birth weight. Potential protective factors include low birth weight and young maternal age. Early intervention in tackling inequalities and maternal smoking in pregnancy are important strategies to help reduce childhood overweight and obesity.