Shared healthcare decisions result from the relationship between a patient and one or more healthcare professionals, and are based on the available evaluation data, also referred to as "evidence". This evidence base is constantly evolving, through a process of continuous construction and reconstruction, adding to and strengthening scientific knowledge. This process is a mark of vitality, not weakness: it ensures that all knowledge, even when considered a matter of fact, is open to debate, which in some cases proves productive and improves patient care.
Data published in peer-reviewed journals are not set in stone. The authors may correct or add to them, either on their own initiative or in response to readers’ comments. A subsequent analysis of the data, perhaps more critical, systematic or rigorous than the first, may also alter results that have already been reported and, in turn, the conclusions that can be drawn from them.
Published articles are sometimes later retracted, due to a glaring error for example, or if fraud is proven or strongly suspected. This renders the reported data worthless, and unusable as a basis for healthcare decisions. Data are also sometimes released prematurely in preprint form, without peer review, but are never formally published in revised form. Whatever the reasons, this raises serious doubts about their quality or even their authenticity.
Scientific fraud and the publication of poor-quality data are failings fuelled by the pressure to publish more in order to achieve more, especially promotion or funding. Other approaches are possible, however, as evidenced by choices certain universities have made in line with the international San Francisco Declaration (see "Universities challenge the impact factor" in this issue > HERE).
Prescrire plays its part in this process, for the benefit of those who work to improve the quality of health care and put patients’ interests first, by valuing debate and by taking a stand itself. Every comment received from a reviewer represents an opportunity for reflection; some readers’ feedback gives rise to a clarification or a correction and sometimes to an article in the "Queries and Comments" section; we re-analyse the harm-benefit balance of drugs "with more follow-up" when new data emerge; and so on.
We can all play our part, in our own way, in building this ever-evolving evidence base.
©Prescrire 1 February 2023
Source: "Building the evidence base" Prescrire International 2023; 32 (245): 31. Free
Enjoy full access to Prescrire International, and support independent information
|