Objective: To investigate whether risk factors for childhood obesity have changed over time, either in their strength of influence or prevalence. Design: Prospective cohort study. Setting: Great Britain. Participants: The 1958 birth cohort and a sub-sample of their offspring. Height and weight measures, and information on prenatal and early life factors were collected on cohort members aged 7 years in 1965 (n = 8552) and on offspring aged 4-9 years in 1991 (n = 1889). Main Outcome Measure: Body

more »

... mass index (BMI) standard deviation scores (SDS) and overweight/obesity, derived using international standards. Results: Prevalence of childhood overweight/obesity increased by more than 50% between generations. Higher parental BMI was associated with increased childhood BMI; for example, cohort members and offspring of obese mothers had more than double the risk of overweight/obesity, with adjusted odds ratios of 2.29 (95% CI 1.76 to 2.97) and 3.20 (1.85 to 5.54) respectively. Maternal smoking and smaller family size were associated with higher childhood BMI in both generations. There was evidence suggesting that social gap in childhood obesity was widening: indicators of lower socio-economic position showed either no association or a protective effect in cohort members, but tended to be associated with increased BMI in offspring. Full-time maternal employment was associated with greater BMI among offspring, for example by 0.42 kg/m 2 in boys, an effect that had strengthened between generations (p,0.05). Several risk factors had altered in prevalence: parental obesity and maternal employment had increased; socioeconomic factors had improved; family size and maternal smoking had reduced. Conclusions: Risk factors for childhood obesity had altered across two generations, either in terms of their strength of association, their prevalence or both. Parental obesity, maternal employment and socio-economic factors may be playing an increasingly important role in the childhood obesity epidemic.