a somewhat complex study in BMJ comparing the mortality benefit of exercise vs drugs in 4 diseases: secondary prevention of coronary heart dz, rehab of stroke, treatment of heart failure, and prevention of diabetes. looked at 300 trials and 340K patients. most trials either compared randomized controlled trials of exercise for these conditions, then compared to medication trials for these conditions (ie, almost all of the trials were not direct comparisons of exercise vs meds). see BMJ 2013;347:f5577 doi: 10.1136/bmj.f5577 (Published 1 October 2013). results:
–the types of exercise interventions varied widely in the trials
–some significant differences in trials with the baseline severity of disease (not so much for secondary prevention of MI, or CHF and impaired glucose tolerance. more so with stroke)
–for secondary prevention of CAD, mortality odds ratios for meds: statins (OR 0.82), ACE-I (OR 0.83), b-blocker (OR 0.85), antiplatelet med (OR 0.83). for exercise OR of 0.89, was not significantly different from the drugs
–for prediabetes, no diff between exercise and using a-glucosidases, biguanides, ACE-I, or glinides
–stroke. mortality significantly lower with exercise (OR 0.83)
–CHF: diuretics were more effective than exercise (OR 0.19, 4x lower mortality), though b-blockers (OR .71) only slightly better than exercise (OR .79), which was slightly better than ACE-I (OR .88) and ARB (OR .92). comparing ARBs to b-blockers, there was a 30% increased mortality with ARBs — again these were typically not direct comparisons of meds
so, more evidence (though not perfect, since most studies are not directly comparing exercise to meds) that exercise is as effective as meds. brings up the (perhaps obvious) situation that much of what we treat medically is to compensate for negative social trends (poor diet, lack of exercise, obesity). brings to mind an old (1879) quote from virchow: “don’t crowd diseases (ie, epidemics) point everwhere to deficiencies of society?” also points out that more medical studies are preferentially done on drugs (???related to the fact that the studies are largely sponsored by drug companies???)
geoff