produced very small to small changes in body composition Subgroup analyses typically resulted in small influences on overall intervention effects and the differences between subgroups were generally small. This was particularly the case for controlled trials versus uncontrolled trials