Physical activity is widely linked with better metabolic functioning, but whether these links reflect distinct causal effects is unclear. In the absence of robust genetic instruments for physical activity, we aim to examine habitual levels of physical activity across adolescence in relation to future metabolic profiles in young adulthood. To do this, we aim to use 3 repeated measures of objectively assessed physical activity (total, light, moderate-to-vigorous, and sedentary time), along with 2 repeated measures of objectively assessed fitness and lean body mass, and to relate these to a wide array of blood-based metabolic markers at follow-up (and to change in these from historical assessments) in the form of clinically relevant traits including fasting insulin and over 80 metabolites including branched chain amino acids. The ALSPAC cohort of children is an ideal resource for carrying out this study given availability of repeated objective measures and the young age of participants which allows associations to be estimated with minimal bias due to subclinical disease.