Even controlling for exercize wouldn't save this study.. Genetic and environmental factors will correlate with a sit-down lifestyle, confounding any possible conclusions.

You could rule out genetic and family environment factors by comparing identical twins who have different sitting habits. My bet is that a big chunk, if not all, of the effect here could be explained by that. For the remaining "effect", you'd still have the issue of non-family environmental collinearity, though.

Maybe you could find a job where people were randomly assigned to sitting vs. standing tasks. That'd be a nifty little natural experiment