Re:More exercise better in long run
by DaniFong on Monday February 01, 2010 @01:54PM
(#1093092)
Almost every interesting and serious study regarding lifestyles is ethically and practically unable to be instantiated as a controlled experiment over the long term. Like many things, from studies of the economy, to social structures, to the climate. You can try to do a multivariate analysis, but ultimately you're not going to have much to go on. You have to take everything with a grain of salt, but it doesn't mean there isn't some valuable data in there. In addition, long term results are probably more interesting and important than short-term results; so much so that I would argue longitudinal studies are probably more useful than controlled studies, so long as one is careful and rigorous and vigorous with multivariate analysis after the fact.
[ Parent
| on HN ]
Re:More exercise better in long run
by kingkongreveng_ on Monday February 01, 2010 @02:03PM
(#1093125)
There are many controlled intervention studies into diet and exercise. It is done. They provide the only results worth any consideration. More often than not regression analyses and "risk-adjustment" models are invalidated.