ABSTRACT
Moderate-to-high levels of physical activity (PA) are now established as conditional lifestyle factors to have a healthy, rewarding, and long life. It is also widely acknowledged that people vary, worldwide, in their PA habits and sports participation. This inter-individual heterogeneity is very well captured by a simple statistic – the variance. The answer to the question – why is there such a wealth of variation within and between populations in their PA levels and patterns? – is not as straightforward as it seems at first glance. Fortunately, the variance can be partitioned and its components “explained” by so-called determinants or correlates studied by Physical Activity Epidemiologists or, by contrast, the labours of Genetic Epidemiologists. The first camp mainly works with individual data where the assumption of independence of observations is mandatory; the second one samples twins and/or nuclear/extended families in which the non-independence due to familial relationships is modeled rather than assuming independence. Fortunately, recent reviews are available within each camp (de Vilhena e Santos et al., 2012). In this brief summary, we will address the topic as follows: firstly, the stepwise approach offered by Genetic Epidemiology (GE) to study PA variation and covariation within and between family members will be presented; secondly, a brief summary about familial data will be done using what we might call a “psychology oriented agenda”; thirdly, GE data will be offered; finally, we will briefly address the Portuguese Healthy Families Study.
