Childhood obesity and device-measured sedentary behavior : an instrumental variable analysis of 3,864 mother-offspring pairs

Objective Intergenerational data on mother-offspring pairs were utilized in an instrumental variable analysis to examine the longitudinal association between BMI and sedentary behavior. Methods The sample included 3,864 mother-offspring pairs from the 1970 British Cohort Study. Height and weight were recorded in mothers (age 31 [5.4] years) and offspring (age 10 years) and repeated in offspring during adulthood. Offspring provided objective data on sedentary behavior (7-day thigh-worn activPAL) in adulthood at age 46 to 47 years. Results Maternal BMI, the instrumental variable, was associated with offspring BMI at age 10 (change per kg/m(2), beta = 0.11; 95% CI: 0.09 to 0.12), satisfying a key assumption of instrumental variable analyses. Offspring (change per kg/m(2), beta = 0.010; 95% CI: -0.02 to 0.03 h/d) and maternal BMI (beta = 0.017; 95% CI: 0.001 to 0.03 h/d) was related to offspring sedentary time, suggestive of a causal impact of BMI on sedentary behavior (two-stage least squares analysis, beta = 0.18 [SE 0.08], P = 0.015). For moderate-vigorous physical activity, there were associations with offspring BMI (beta = -0.010; 95% CI: -0.017 to -0.004) and maternal BMI (beta = -0.007; 95% CI: -0.010 to -0.003), with evidence for causality (two-stage least squares analysis, beta = -0.060 [SE 0.02], P = 0.001). Conclusions There is strong evidence for a causal pathway linking childhood obesity to greater sedentary behavior