Background Lower socioeconomic position (SEP) is a risk factor for obesity in children and adolescents in high-income countries. Since diet and physical activity are the main drivers of obesity, the socioeconomic differential in obesity is likely to arise through SEP differences in one or both of these.

Methods We explored SEP differences in trajectories of fat mass, energy intake and physical activity (PA) across childhood and early adolescence, using maternal education as a measure of SEP, in the Avon Longitudinal Study of Parents and Children. Fat mass was measured by DXA scans at mean ages 10, 12, 14, 15 years. Energy intake was assessed using food frequency questionnaires (FFQs) at 3, 4.5, 7, 8.5 years and 3-day diet diaries at 3.5, 5, 7.5, 10, 13 years. PA was assessed by accelerometers worn for 7 days at 12, 14, 16 years, used to calculate mean total counts per minute (CPM) and mean minutes of moderate to vigorous PA (MVPA). Energy intake (adjusted for FFQ vs diary), % fat mass, CPM and MVPA were all modelled using linear spline multilevel models.

Results 5739 individuals had data on maternal education and at least one measure for each of fat mass, energy intake and PA. The sons of women educated to degree level (highest maternal education category) had consistently lower % fat mass across childhood and early adolescence, but differences between the 3 lower SEP categories only began to emerge from age 14. In females there was a stepwise increase in percent fat mass for each decreasing category of maternal education, and the SEP gradient remained largely stable between 10 and 15 years. The SEP patterning in trajectories of energy intake and PA (CPM and MVPA) did not resemble the SEP patterning in trajectories of fat mass. For energy intake, no SEP differences were observed. For PA in males, there was little SEP patterning in MVPA, but for CPM there were higher levels of activity in lower maternal education groups, i.e. the reverse of the fat mass gradient. For females, CPM and MVPA were both higher in the highest maternal education category, but contrary to the fat mass trajectories, differences between the lower three SEP categories were not consistent.

Conclusion Socioeconomic differences in trajectories of energy intake and PA do not resemble the patterns observed for fat mass. Potential explanations for this finding, including measurement error, will be discussed.