|Home | About | Journals | Submit | Contact Us | Français|
Von Elm et al provide a welcome set of criteria to judge prospective observational studies.1 What they do not include is the government health warning that should accompany such publications. The accompanying editorial highlights the usefulness of these studies to examine rare diseases,2 but such studies are often used for common illnesses like cardiovascular disease and cancer. End points are frequent and prevalence high enough to make randomised controlled trials more reliable for assessing these illnesses.
The quoted examples seem to prove the importance of cohorts, but a list from cardiology alone shows that results of observational studies are often seriously flawed. Observational studies of the cardioprotective effects of female sex hormones, the usefulness of antioxidants or homocysteine lowering strategies, and rhythm control for atrial fibrillation suggested a clear treatment effect and greatly influenced practice. But subsequent randomised trials refuted each hypothesis.
The main problem is interacting factors that cannot all be statistically accounted for. For example, in general, overweight people do less exercise, have a high saturated fat intake, smoke, and do not attend to their insulin therapy or take their blood pressure tablets. So, the results of cohort studies are often wrong if cohorts are considered in isolation. This would not be a problem if cohort studies were not acted upon until a randomised trial is conducted. Glasziou et al suggested that a combined rates ratio of at least 10 and a P value of <0.01 should be used to distinguish between a true effect and background population “noise.”3 Few of our current favourite targets—mild obesity, salt intake, or passive smoking—would pass this test. The findings of cohort studies should start rather than close the debate. Experts are too hasty to present a hypothesis as a proven fact, and the medical profession is too willing to accept such findings uncritically.
Competing interests: None declared.