Physical activity substitution: An overlooked constraint on energy expenditure during exercise and physical activity interventions

Aims
When exercise is prescribed, the new exercise could ‘substitute’ for pre-existing physical activity such that the net effect on energy expenditure is less than predicted. However, the impact of substitution has not been examined to date because of methodological spatiotemporal challenges. To overcome these challenges, we use mathematical modelling to examine the impact of substitution during prescribed exercise interventions.

Materials and Methods
We modelled the impact of substitution during two prescribed exercise interventions (LOW and HIGH) on energy expenditure in 242 patients aged 63 ± 6 years recruited in the south-west of England. The increase in net daily energy expenditure after subtraction of pre-existing physical activity at the time of the new modelled prescribed interventional exercise was derived to account for the potential effect of substitution.

Results
After accounting for substitution, the increase in daily energy expenditure was 38% ± 32% and 73% ± 12% of the potential increase under additive models for LOW and HIGH prescribed exercise scenarios, respectively. Furthermore, substitution introduced considerable heterogeneity in the predicted change in daily energy expenditure. This heterogeneity was most evident in the LOW prescribed exercise scenario, where the predicted change ranged from +91% to −93% of the anticipated increase under an additive model.

Conclusions
Physical activity substitution has the potential to erode the increase in daily energy expenditure during prescribed exercise interventions, and to introduce marked heterogeneity in the treatment response (change in energy expenditure). Successful mitigation of physical activity substitution could potentially increase the effectiveness of physical activity interventions.