Chief knowledge Officer of the NHS, Director of The Oxford Centre for Healthcare Transformation
Director, The Nordic Cochrane Centre
Mammography screening is immensely harmful, just like screening for prostate cancer is. A systematic review of countries with organised screening programmes found an overdiagnosis of 52%. This means that a woman can lower her risk of becoming a breast cancer patient by one-third by avoiding going to screening. It has never been shown that screening makes women live longer, and it's not even likely that it does. Declines in breast cancer mortality have coincided with the introduction of effective treatments in all countries and have been totally unrelated to the introduction of screening. Treatment of healthy overdiagnosed women with radiotherapy increases their risk of dying from heart disease and cancer, which is likely to neutralise any possible small effect screening might have on breast cancer mortality. It is time to stop mammography screening.
Professor of Clinical Epidemiology, University of Amsterdam
Improved diagnostic testing can transform healthcare, but diagnostic evidence is limited in scope and often poorly understood. Patrick Bossuyt leads the Biomarker and Test Evaluation Program, to develop the methodology for evaluating medical tests and biomarkers. He spearheaded the STARD initiative for the improved reporting of diagnostic test accuracy studies.
Chair of Council of the RCGP
One of modern medicine’s greatest strengths is its evidence base. It’s the jewel in our crown.
Modern medicine – though far from perfect – has nevertheless flourished through the systematic application of rigorous, evidence-based research.
I will be discussing how far evidence in the political arena compares to that used to manage patients and what this has meant for the delivery of our health care and the NHS
Professor, Dalla Lana School of Public Health, University of Toronto
Human life expectancy improved more in the last 50 years than in the preceding 5000 years. Much of this recent progress arose from declines in childhood mortality, and most of this decline was due to scientific knowledge and technologies (defined widely as drugs, diagnostics, policies, strategies, and epidemiological knowledge). The dominant challenge of the 21st century is to apply scientific knowledge to reduce premature adult mortality, in particular from vascular and neoplastic disease but also from persistent infectious disease such as malaria. Reliable quantification of the causes of death is a key starting point for control of adult diseases, as shown by the early results from India's Million Death Study. A key global priority is the control of tobacco use. Scientific research on adult mortality, paired with specific action, might well halve premature adult mortality worldwide in the next few decades.
Editor-in-Chief of JAMA
It is challenging for editors and publishers to ensure that what they publish is valid. This issue is relevant for opinion pieces, reviews, and research regardless of authorship and funding. Over the past decade various approaches have been developed to ensure that results are valid – will any succeed?
Chairman of the National Institute of Health & Clinical Excellence (NICE)
Sir Michael Rawlins recently stepped down as chairman of the National Institute of Health & Clinical Excellence (NICE), having been at its helm since its formation in 1999. He has a wealth o f experience and was appointed Knight Bachelor in 1999 for services to the improvement of patient protection from the side effects of medicinal drugs. He is the current President of the Royal Society of Medicine.
Director of Clinical Evaluative Clinical Sciences (CECS), Dartmouth Medical School
Medical practice occurs within a local context. Per capita expenditures, resource use and utilization rates vary extensively among regions, communities and health care organizations, and raise serious policy questions concerning the effectiveness and efficiency of health care delivery. Much of the variation is unwarranted: It isn’t explained by illness, evidence-based medicine or patient preferences. Variation leads naturally to a question: Can better care cost less? Examples of research discussed in this lecture provide benchmarks that suggest the answer is Yes! But to reduce variation and realize savings of a magnitude that seems possible, it will take a concerted effort to address the weaknesses in the scientific and ethical basis of clinical decision making uncovered by variation research.
Professor of Epidemiology, Wolfson Institute of Preventive Medicine
In analysing data a balance needs to be struck between general statistical principles and an understanding of the subject matter. Of late more and more prescriptive requirements are being drawn up as to how different sorts of studies should be presented. For overviews the PRISMA checklist of 27 items is now the standard and for RCTs the 25 item CONSORT checklist is used. While helpful in general, these checklists are no replacement for a good understanding the subject matter of the overview and judgement about the quality of the studies which often has a subjective component. In this lecture this tension is explored and some examples are given where the blind following of prespecified rules has led to reviews which defy common sense.
Honorary Consultant Cardiologist, University Hospital of North Staffordshire
Critical steps in Evidence Based Medicine are systematic retrieval and critical analysis of the best available evidence. Greater value is given to randomised controlled trials and meta-analysis than to case studies. This talk will discuss personal experience and case studies showing industry’s use of bribes and legal tactics, including gagging contracts and libel actions, to prevent publication of unfavourable results; and the role of undisclosed conflicts in the misreporting of research. From analysis of case studies showing widespread organisational and individual research misconduct it is concluded that in the era of Evidence Based Medicine much of the “higher quality evidence” cannot be trusted.
Editor, ACP Journal Club
In an ideal world, sound evidence from research would be enough to change clinical practice and patient-important outcomes. In real life, we need to shepherd evidence through information retrieval and appraisal processes, systematic reviews, guidelines, health care policy and resource allocation decisions, training programs, incentives and disincentives. The infrastructure to support most of these steps is increasingly robust and includes knowledge refineries, more efficient systematic reviews, higher quality guideline processes, more timely health policy initiatives, and the emergence of effective computerized point-of-care decision support. Clinician compliance and patient adherence remain major barriers to success and there are many opportunities for trailblazers and innovators.
Director of the Centre of Public Health Excellence, NICE
In 2005 NICE acquired the responsibility for the production of public health guidance on cost effective interventions to prevent disease and promote good health. The guidance was for the NHS and the wider public health community. The ambition was to build on the platform of evidence based medicine, and the experience NICE had acquired in the preparation of clinical guidelines and conducting health technology assessments. From the outset it was acknowledged that applying the principles of evidence based medicine to public health would require the methods and processes to be developed and to be allowed to evolve. This paper describes the scientific, methodological and practical challenges that have arisen in developing guidance in this way. Among other things the nature of the evidence base, the dynamics involved in complex public health interventions and the challenges presented by health inequalities, will be discussed. The problems associated with the strong emphasis on internal validity in evidence based medicine and the importance of external validity in public health will be explored. In conclusion, the significance of the judgement process in involved in interpreting the evidence will be considered.
Thirty years ago, journalists William Broad and Nicholas Wade sounded the alarm over research fraud with their book Betrayers of the Truth. Today, journalists Adam Marcus and Ivan Oransky carry this forward with the website Retraction Watch. Journalists remain a leading force for integrity.
Responses from the biomedical community include indifference and denial. Opinion leaders often justify inaction with conceits, such as “reproducibility” and “peer review”. Journal editors and institutions fail to deal with fraud, misconduct and bogus findings, reluctant to commit resources and fearful of retaliatory litigation.
In high-profile areas, research claims impacting safety and health are increasingly evaluated and disseminated by dubious authorities, unqualified consumers and even cranks. To protect the public, the need to remove false and misleading information from the literature has become urgent. I present examples, and suggest amendment to the ICMJE Uniform Requirements so as to facilitate forensic post-publication audit and rapid no-fault retractions.