It is about as proven as it gets in human studies that moderate exercise is good for you and a sedentary lifestyle is bad for you. The difference in life expectancy is probably somewhere in the low end of the five to ten year range, though it is very challenging to pin down how much of that is due to more direct mechanisms of exercise versus indirectly related items such as the amount of visceral fat tissue present in the body. Human studies turn out correlations only: it isn't as though you can structure your study population up front to determine cause and effect. That is possible in animal studies, however, and those robustly demonstrate that moderate exercise is good for you while lazing around being fat isn't. It is entirely reasonable, I believe, to live life under the assumption that the causation for exercise and health shown in animal studies is the reason for the correlation shown in human studies.
Things become somewhat more challenging when you move beyond moderate exercise in humans, however. Whether twice as much exercise is better and why it is that professional athletes live longer than the general population are harder questions to answer. The athlete correlation may well exist because people who are capable of success in that field of endeavor are simply more robust than the rest of us, and would have lived longer than their peers regardless of career. Similar issues exist with most past data on levels of exercise beyond merely moderate and the resulting benefits.
Here is another interesting question: is there such a thing as too much exercise? Is exercise like any drug in that the dose-response curve is most favorable in the middle, and going too far in either direction means losing the benefits? This becomes a more pressing issue in later life, when an individual is damaged by aging and suffering the consequences of a lifetime's accumulation of cellular and molecular defects. We are machines, and a damaged machine is more prone to breakage at any given level of activity. So when we learn that people who are old and damaged tend to have a raised rate of cardiovascular failure at greater levels of exercise, that is the lesson we should take away: damaged machines break. We shouldn't accept this fate for ourselves and restrict our activities, but rather eagerly support research into ways to repair the damage that so greatly impacts health and survival in later life.
There is strong epidemiological evidence of the importance of regular physical activity, such as brisk walking and jogging, in the management and rehabilitation of cardiovascular disease and in lowering the risk of death from other diseases such as hypertension, stroke, and type 2 diabetes. The Physical Activity Guidelines for Americans recommends about 150 minutes per week of moderate-intensity exercise or about 75 minutes of vigorous-intensity exercise. But there is clear evidence of an increase in cardiovascular deaths in heart attack survivors who exercise to excess, according to a new study.
[Researchers] studied the relationship between exercise and cardiovascular disease-related deaths in about 2,400 physically active heart attack survivors. They conducted a prospective long-term study using the National Walkers' and Runners' Health Studies databases. This study confirmed previous reports indicating that the cardiovascular benefits for walking and running were equivalent, as long as the energy expenditures were the same (although when walking, as compared to running, it will take about twice as long to burn the same number of calories). Remarkable dose-dependent reductions in deaths from cardiovascular events of up to 65% were seen among patients who were running less than 30 miles or walking less than 46 miles per week. Beyond this point however much of the benefit of exercise was lost, in what is described as a reverse J-curve pattern.
"These analyses provide what is to our knowledge the first data in humans demonstrating a statistically significant increase in cardiovascular risk with the highest levels of exercise. Results suggest that the benefits of running or walking do not accrue indefinitely and that above some level, perhaps 30 miles per week of running, there is a significant increase in risk. Competitive running events also appear to increase the risk of an acute event. However, our study population consisted of heart attack survivors and so the findings cannot be readily generalized to the entire population of heavy exercisers."
Habitual physical activity and exercise have repetitively been shown to reduce the risk of sudden cardiac death and acute myocardial infarction. Several articles suggest that there is a U-shaped relationship between running dose and reduced all-cause mortality, but these studies lack the statistical power to formally test for a nonlinear dose-response relationship. We therefore used the National Walkers' and Runners' Health studies to examine the dose-response relationship between exercise energy expenditure and cardiovascular disease (CVD)-related mortality in heart attack survivors. The large number of highly active subjects, and the high risk for CVD mortality in persons having had a previous heart attack, provides the statistical power required for testing whether there is an increased CVD risk in the most active heart attack survivors.
Accumulating evidence suggests that exercise practices that are ideal for promoting health and longevity may differ from the high-volume, high-intensity endurance training programs used for developing peak cardiac performance and superb cardiorespiratory fitness (CRF). Studies consistently show that regular moderate-intensity physical activity (PA) is highly beneficial for long-term cardiovascular (CV) health. Improving the CRF from low to moderate to high will progressively improve CV prognosis and overall survival. However, the survival benefits from improvements in the CRF plateau at about 10 metabolic equivalents (with 1 metabolic equivalent equal to an oxygen consumption of 3.5 mL O2/kg body weight per minute), with no additional survival benefit accruing from higher CRF levels. Clearly, 30 minutes of regular vigorous PA enhances health and well-being, but performing 3-hour bouts of strenuous PA does not multiply the health benefits.