The most interesting result from recent research into exercise is this:<p><a href="http://www.alphagalileo.org/ViewItem.aspx?ItemId=159410&CultureCode=en" rel="nofollow">http://www.alphagalileo.org/ViewItem.aspx?ItemId=159410&Cult...</a><p><a href="http://dx.doi.org/10.1038/srep18259" rel="nofollow">http://dx.doi.org/10.1038/srep18259</a><p>An open access paper on exercise in identical and non-identical twin pairs was recently published, the data suggesting that long-term differences in physical activity between identical twins don't result in any significant difference in longevity, even though other differences in health outcomes are observed. We might draw parallels between this and similar results observed in a mouse study from a few years back, in which the exercising mice had better health but no increase in maximum life span. The researchers here theorize that the well-known epidemiological association between exercise and increased life expectancy is perhaps as much a matter of genetics as of choice.<p>For any observed statistical relationship in humans there are always questions of causation. This is especially true in the web of associations related to aging and mortality in population data, in which life expectancy, wealth, social status, intelligence, education, exercise, diet, and culture all have ties to one another. That we pay great attention to these relationships is a function of having no good way to treat aging, I've long thought: we care about trivial differences in life expectancy of a few years here and a few years there because this is all that is in our power to change right now, and that will continue until the development of rejuvenation therapies. Life expectancy and exercise are linked robustly in many data sets, and even more so now that accelerometers are so cheap and ubiquitous that even large studies can use them to obtain actual rather than self-reported data on physical activity. There are studies to demonstrate longer life expectancy in athletes, longer life expectancy in those who exercise modestly versus those who are sedentary, and so forth. What are these studies measuring, however? For example, what if people who are more robust and would live longer regardless of exercise tend to exercise more? Or perhaps exercise levels are a good proxy for lower levels of visceral fat tissue and consequent chronic inflammation - themselves linked to greater risk of age-related disease and mortality.<p>The results of this study definitely muddy the waters in the search for causation and mechanism in exercise and mortality reduction, providing evidence to support a state of considerable complexity in the relationship between exercise, genetics, and outcomes in health. Nothing in biology is ever as simple as we'd like it to be, so this should perhaps be expected. Regardless he data presented below should be added to the many past studies on exercise and mortality, and its weight balanced accordingly - never take any single set of data and interpretations as gospel in science. This doesn't change the consensus, which is that you should exercise, and that you are expected to obtain benefits by doing so. It does add subtlety to the picture, however.