Approved anti-depressant deemed unsafe:
Studies suggest that patients, clinicians, and society often hold unrealistic expectations about the effectiveness of tests and treatments. Two articles in this issue add to that literature. In New Zealand, Hudson et al3 surveyed 977 primary care patients and found that many overestimated the benefits of cancer screening and chemopreventive medications. The minimum benefit from screening that respondents deemed acceptable was less than their known benefit. The survey had a modest sample size and low response rate (36%), and its findings might not be fully applicable to other countries, but US studies have reported a similar problem. For example, a variety of studies document Americans' appetite for procedures of dubious effectiveness and their overestimation of benefits.4,5
Physicians are not immune to false beliefs about clinical efficacy or complication rates.8Correcting such misperceptions has always been part of the impetus for the evidence-based medicine movement and its promulgation of systematic evidence reviews, practice guidelines, and other tools that present the facts on benefits, safety, and scientific uncertainties. Even these tools, however, can reflect the misconceptions of those who produce them. The specialists who serve on expert panels derive much of their clinical case knowledge from the patients with advanced disease who fill their clinics. Having seen the worst of the worst, they are less sympathetic to expressions of concern about the potential harms of interventions or imperfections in efficacy studies.9Whereas epidemiologists consider the population denominator to put the numerator in perspective, the world of specialists is confined to the numerator, giving them a skewed basis for judging the population prevalence of diseases or benefit-risk ratios. Were this not enough, the preeminent scientists who often serve on guideline panels bring additional biases, such as being the authors of key studies under review or having financial ties to industry.10
http://www.annfammed.org/content/10/6/491.full
