Limitations of our study are the small study sample and
the relatively late first measurement (only at 15 wk of
gestation). Strength is the objective measurement of PA
and the longitudinal aspect of the study, enabling us to
look at differential effects of PA at different moments in
pregnancy. The results presented here derived from secondary
analysis of data from a randomized controlled trial
evaluating the effectiveness of a training program. The
intervention did not have an effect on glucose and insulin
metabolism, as reported previously (11). This apparent
contradictory finding that the training program did not
have an effect can be attributed to the very low compliance
of the women to the intervention. Therefore, we felt justified
in analyzing the data as a longitudinal cohort, combining
intervention and control groups.