Weiye Loh's Library tagged Overload →  View Popular, Search in Google

long-term urban planning “requires leadership, and standing up, and saying, ‘You know, you elected me, this is what we’re going to do,’ and not take a referendum on every single thing.”

the mayor had been speaking narrowly about how social media can shift the public discourse away from long-term thinking.

An article in the May Consumer Reports Health shed some light on this question.  It discussed the widely publicized medical study that showed, contrary to expectations, that “raising HDL (good) cholesterol with drugs did nothing to protect against heart attacks.”  This, the article said, was surprising because observational studies had shown that people with lower levels of HDL had more heart attacks than those with higher levels of HDL.   An observational study, however, shows only a correlation between two variables (e.g., level of HDL and number of heart attacks).  The new result came from a randomized clinical study, using one group of patients who receive a given treatment and a “control group” of patients who do not.  Unlike an observational study, such a study can show whether or not, for example, higher HDL actually prevents heart attacks.  The article went on to emphasize that “we should almost never rely on the results of observational studies, which can only suggest associations with disease but not prove them.”

implicit in most media reports, is that acting on the results of the unreliable observational studies  “couldn’t hurt and might help.”   This makes sense if I have a medical problem for which there is no reliable remedy.   If nothing else has helped my arthritis, insomnia or back pain, it would make perfect sense to try a remedy that will not do serious harm and has some probability of working.