The media is often bashed for misleading reporting on medical studies. However, a look at the NHS's very useful site Hitting the Headlines reveals that newspaper reports of scientific studies are, more often than not, accurate reflections of the study. More problematic is the assessment of reliability of the conclusions of the studies themselves. Researchers tend to speculate beyond the scope of their study, or put insufficient information in the public domain for those hard-working analysts at the Centre for Reviews and Dissemination (who provide synopses for the site) to evaluate the findings. A recent article in the BMJ urges us to 'Read only the Methods and Results sections; bypass the Discussion section' to avoid being misled by biased interpretations of the data -- interpretations influenced, they claim, by funding from for-profit companies. All this hardly engenders trust in the peer review process. Surely this kind of chicanery is what peer review is supposed to deter? It is saddening to see that the BMJ itself will no longer be open-access from January. Dare I suggest that they cut costs by omitting those speculative and misleading 'discussions' from their articles?