Review of EBM framework for rational decision making explosion:estimates in 2000 sug sted that more thar ion of care which ing the RCIs ar o ormation explosion and the EBM of patient pretere ide. igh-quam ague tion ng EB Develo t of standards for conduct and disciplines. need tobe of only about 20 new articles er yea pt date.and.to sta proving design and reporting of research based topicsneed be re this ke appears in the medical literatur emE0UATORet the ideal setting? versus pragmatic (also known as based summarics and GRADE re mmendations (g evaluation of 26 existing point-ofcare nformat have products e earch instance h the s than others latfor pgdsabe,btorehae re ng is GRADE fran report or suppres hur d storage likely major and to the body of scie facilitatin on and dynamic atier are and 4 and dei aids Such .Despite e pro rapid updating of sy view rmats or smart phones and othe only haphazard adherence in 2016. support systems.Formal re earch ha optimise Evidence dissemination and access endations in thes 6736(163592-Review www.thelancet.com Published online February 16, 2017 http://dx.doi.org/10.1016/S0140-6736(16)31592-6 5 presented diff erently, in terms of gains vs losses). Through explicit consideration of judgment regarding the balance of benefi ts and harms, resource use, and issues of feasibility and equity, EBM has articulated a framework for rational decision making.56,57,61 The recognition that values and preferences vary widely among individuals has an important implication: the standardisation of care, which was one of the original reasons for the introduction of guidelines, and is still considered a key rationale for assessing the quality of care initiatives,62 is neither possible nor desirable63 for the many value and preference-sensitive decisions that clinicians and patients face. GRADE acknowledges the intrinsic variability of patient preferences in its classifi cation of recommendations as strong (right for all, or almost all) and weak (conditional, contingent—right for most but not all, and requiring presentation of evidence that facilitates shared decision making). Development of standards for conduct and reporting of clinical research Scientifi c communities have embraced EBM-related initiatives to develop guidance and checklists for improving design, conduct, and reporting of research. Numerous such initiatives have occurred during the past 25 years, including checklists and statements on how to develop a research protocol and report randomised trials, observational studies, diagnostic test studies, predictive models, and genetic testing studies; they can be accessed via the EQUATOR website. Along the way, researchers have increasingly differentiated between explanatory (also known as mechanistic or proof-of-concept effi cacy) trials that address the question “can intervention work in the ideal setting?” versus pragmatic (also known as practical, eff ectiveness) trials that address the question “does it work in real-world settings?” and “is it worth it and should it be paid for? (effi ciency)”.64,65 There is some evidence that these initiatives have resulted in improvement of the quality of reporting of research—for instance, the reporting of RCTs has improved as a result of the CONSORT checklist.66 Optimal reporting is desirable, but worse than poor reporting is failure to report or suppression of clinical research. Currently, investigators report only 50% of their trials,67,68 a major and avoidable threat to the body of scientifi c knowledge. When half of studies are unreported, both patient care and new research initiatives will often be fl awed. Despite a longstanding awareness of the problem of publication bias,69 the only possible solution to the problem—registration of all trial protocols before research is actually undertaken, and full reporting of the results in a timely manner after the study is completed—received only haphazard adherence in 2016.70 Evidence dissemination and access Practising EBM at what David Sackett called “the coal￾face”71 of clinical care requires rapid access to the best available evidence, suitably fi ltered to ensure effi cient use. Provision of that access is challenging and has been one of the most important academic endeavours of EBM. The process is complicated by the ongoing information explosion: estimates in 2000 suggested that more than 6 million articles are published in more than 20 000 biomedical journals every year.72 MEDLINE alone contains over 22 million indexed citations from more than 5600 journals,73 and 75 RCTs and 11 systematic reviews are published every day.74 Dealing with both the information explosion and inherent human brain limitations in processing evidence has required the application of the EBM principle of critical appraisal to identify high-quality research on the clinicians’ behalf. Haynes and colleagues75 have developed a model service that uses EBM critical appraisal techniques to systematically evaluate more than 3000 articles per year from all medical disciplines. Using EBM information processing and fi ltering, they reported that, on average, clinicians need to be aware of only about 20 new articles per year (99·96% noise reduction) to keep up to date, and, to stay up to date in their area of expertise, authors of evidence￾based topics need be aware of only fi ve to 50 new articles per year.75 Similarly, practising oncologists need be cognisant of only 1–2% of published evidence that is valid and relevant to their practice.76 Information services provide clinicians with alerts when this key new information appears in the medical literature, as well as providing fi ltered search systems that prioritise processed evidence (including clinical practice guidelines and systematic reviews). Electronic textbooks also provide valuable pre-processed information, including evidence￾based summaries and GRADE recommendations (eg, Dynamed and UpToDate), as do other evidence summaries (eg, Best Evidence in Emergency Medicine).45 However, an evaluation of 26 existing point-of-care information summaries found uneven quality across the products, with some products scoring higher on evidence-based dimensions than others.77 Nevertheless, electronic platforms based on the GRADE framework (eg, Making GRADE the Irresistible Choice78,79) that allow digitally structured storage of information are likely to play an important role in facilitating the creation, dissemination, and dynamic updating of trustworthy evidence summaries, guidelines, and decision aids. Such platforms also facilitate rapid updating of systematic reviews and guideline adaptation,79 automated publication of multi-layered presentation formats on smart phones and other devices, and integration of evidence (summaries) and recommendations into electronic medical records as decision support systems. Formal research has optimised and will continue to improve presentation formats of evidence summaries and recommendations in these applications, helping to ensure maximal uptake by front￾line clinicians. For more on EQUATOR see http://www.equator-network. org/index.aspx?o=1032