Evidence Live Blog
First of all: thank you to everyone who came to Evidence Live 2013 and made it such a success. A conference like this only happens because people come from around the world, taking time out of busy schedules and money out of already-stretched budgets, to participate and to share their passion for evidence-based healthcare.
It’s no secret that healthcare is at somewhat of a crossroads these days. People are waking up- not just those ‘in the know’, but the general public too- and realizing that if healthcare is going to be sustainable and continue to improve people’s lives, that radical changes are necessary in research and policy. Jack Wennberg’s work on unwarranted variation in healthcare can rightly be considered one of the first great revolutions in health service research, and when looking back at what was discussed this week in Oxford, his description that there continue to be large variations in the quality and quantity of care provided across the U.S. highlights that there is still a long way to go. Ensuring equitable access to an appropriate level of service- not too much, not too little- is essential. So too is the publication of all clinical trials- if you haven’t already done so, sign up now to the AllTrials campaign.
The Cochrane Collaboration has been one of the greatest developments in healthcare ever, ensuring that high quality evidence is generated and disseminated for more than two decades. As ever, it was a privilege to have one of its founders, Sir Iain Chalmers, participating in the conference. What was most exciting, though, is that even with Sir Iain, the Editor-in-Chief of The Cochrane Library David Tovey, and the Director of the UK Cochrane Centre Martin Burton sitting in the room, speakers were unafraid to provide poignant, thoughtful criticism of Cochrane’s work in order to push it forward and make it even better. Speakers like Jack Cuzick, Peter Gøtzsche, and Tom Jefferson, among others, highlighted areas where they believe Cochrane can do better, and set out challenges for the Collaboration to refine and evolve their methods in keeping with the latest evidence and methods. It is a testament to the professionalism and scientific integrity of all delegates that Evidence Live was a safe space in which these critiques could be raised.
While the ‘good and the great’ of evidence-based medicine were all there at the conference (indeed, prompting Chairman of the National Institute for Health and Clinical Excellence Sir Michael Rawlins to answer a question telling a delegate to ignore the good and the great and do whatever they thought was right!), it was also about the future of EBM- those younger delegates to whom the reins are rapidly being handed. Students and early career researchers came from all over the world- from places like Iran, and Bangladesh, and China- to participate in the debate, present posters on cutting-edge research, and to hear from distinguished speakers. A rapid-fire session on the second day witnessed some of the most original (even controversial) presentations of the conference. Five speakers took their 'favourite paper of all time' as the starting point, and used this to discuss the future of EBM and how students can contribute. It was an effective reminder that we have come a long way, but there is still much to do.
“But, by the words we speak and the faces we show the world, we force the spring.”
The beginning of Bill Clinton’s first inaugural address seems an odd place to start a discussion about epidemiology, I admit. For us, though, it reflects the developments over the past few months that have changed how evidence-based medicine is practiced, and how it’s going to look in the near future. And we’re excited about all of it.
It’s no secret that there is massive pressure on drug companies to fundamentally change how they operate. This push for a new era of accountability is due to the efforts of many people, including Ben Goldacre, whose Bad Pharma has become an international phenomenon, and Tom Jefferson and Peter Doshi, whose campaign to obtain all the data on Tamiflu has been a major driver towards exposing withheld data. The current state of pharmaceutical research is a little like when you were in school doing an experiment, and something didn’t quite turn out right, and you ‘forgot’ to write down the results for that one part of the experiment. It may have worked in high school, when the worst that could happen is that you would have to stay after class. But when billions of pounds and thousands of lives are at stake, the stakes are quite different.
The All Trials campaign- an online petition that’s rapidly gaining publicity and support- now has over 25000 signatures, including major medical journals (our EvidenceLive partner, the BMJ, is one of the instigators of the campaign). Last week, pharmaceutical giant GSK signed on - an incredible step in the right direction for an industry that continues to be, in large, stuck in the dark ages when it comes to transparency.
This week, another campaign has started to heat up, this time questioning how guidelines are developed and where the incentives lie in their creation. Bad Guidelines has initially targeted the “Guidance on collaboration between healthcare professionals and the pharmaceutical industry” as a particularly egregious example of a document that august institutions, such as the UK Department of Health, have signed on to and probably shouldn’t have. At the very least, one wonders if they read it through before signing up.
The philosopher Thomas Kuhn suggested that the history of scientific progress has been one of paradigm shifts, in which research continues along in a certain paradigm until someone comes along to break out and move on to another radically different way of looking at phenomena. We can’t help but wonder if we are in the midst of a paradigm shift in healthcare- one that values transparency, scientific progress, and responsibility to patients above all else. If you haven’t already signed up, come to EvidenceLive, and help “force the spring” towards a new era in evidence-based medicine.
The Government of Canada reported last week that they are working with pharmaceutical manufacturer Roche to release the neuraminidase inhibitor Tamiflu (oseltamivir) from Canadian stockpiles to ensure availability of the drug for a particularly severe flu season in Canada this year. This move was facilitated by the Public Health Agency of Canada (PHAC) and Health Canada in response to higher than expected demand for the drug.
The agency’s Director of Pandemic Preparedness stated that “given that we are still in the midst of influenza season and given that we certainly do not want to have any kind of supply disruption at this point in time, the Public Health Agency Canada ... has elected to work with Roche to ensure that there will be no supply disruption."
The primary maxim of Evidence-Based Medicine- the whole point of EBM, really- is that clinical decisions should be made according to the best available evidence. We don’t profess to know all the evidence behind what should be stockpiled at a national level in Canada or elsewhere. This complex practice requires an enormous amount of evidence review and surveillance data to determine what will (and could) be needed to safeguard a country.
But we do know something about the evidence base for neuraminidase inhibitors, and the recent controversy about Roche’s withholding of clinical trial data on the use of these drugs to treat influenza. Tom Jefferson and colleagues’ 2009 systematic review and meta-analysis of available data is a great starting point, and notes that publicly available studies, taken together, suggest “modest effectiveness of neuraminidase inhibitors against the symptoms of influenza in otherwise healthy adults”. More importantly, though, their study goes on to note that a “paucity of good data has undermined previous findings for oseltamivir’s prevention of complications from influenza” and that “independent randomized trials to resolve these uncertainties are needed.”
When they reviewed the evidence again in 2012, they were unable to access full clinical study reports despite requesting them from the manufacturer. They concluded that "a high risk of publication and reporting biases in the trial programme of oseltamivir" precluded being able to properly assess the evidence of its effectiveness.
Without this data being available, then, the evidence to support using Tamiflu is questionable at best, and certainly not strong enough to justify stockpiling hundreds of millions of dollars of it for national pandemic preparedness.
What has the PHAC done about this? In 2010, the CMAJ reported that the PHAC reviewed the results of the aforementioned systematic review but “the articles published in the BMJ do not change [the] Public Health Agency of Canada’s advice with respect to the use of antivirals with the H1N1 flu virus.”
The BMJ has taken up this issue of missing clinical trial data, and is pressing Roche to release all its data on Tamiflu trials, and clinical epidemiologists are trying to obtain this data to properly analyze it. Only then, will we know if this drug is truly effective, and if the enormous sum spent on stockpiles around the world was worth it.
The Alltrials petition we wrote about last week is another example of an attempt to remove this critical impediment to scientific progress, and ensure all the necessary evidence is available for decision-making.
Ultimately the decision to prescribe is made by a practitioner acting in patient’s best interest. Beyond understanding relevant guidelines and public health recommendations, we encourage individual practitioners to familiarize themselves with the best available evidence to be the best for their patients. As it stands, we’re slightly bewildered as to PHAC’s decision to promote Tamiflu so strongly, and that’s why we encourage individual practitioners to make their own decisions using the best, most recent evidence.
“Thousands of clinical trials have not reported their results; some have not even been registered.” All trials registered | All results reported
This is a problem.
A petition was launched today that calls on governments, regulators, and research bodies to put measures in place to register and report the methods and results of clinical trials. This initiative, led by Bad Science, Sense About Science, BMJ, James Lind Initiative and CEBM is important. This issue effects all of us: patients, researchers, clinicians, politicians, scientists, and industry.
The petition was followed with a rip roaring editorial by Iain Chalmers, Paul Glasziou and Fiona Godlee in the BMJ that calls for all trials to be registered and their results published. This excellent piece details the consequences of our collective abstention from action and provides advice to patients whom are invited to participate in clinical trials; namely:
“Agree to participate in a clinical trial only if: (1) the study protocol has been registered and made publicly available; (2) the protocol refers to systematic reviews of existing evidence showing that the trial is justified; and (3) you receive a written assurance that the full study results will be published and sent to all participants who indicate that they wish to receive them.”
Don’t wait, sign the petition now.
After signing you can automatically share the message “I've just signed the #AllTrials petition for all trials registered and all results reported” on Twitter or Facebook. Be proud you are taking a step for transparency and improving patient care.
It’s no coincidence that the mentioned speakers and organisations’ are taking part in Evidence Live 2013. We are particularly excited to have Sense About Science join the programme.
Join us from 25-26 March 2013 when the world comes to Oxford to participate in the debate.
Analysis of treatment descriptions, published in BMJ Open, shows that for over half (57%) of the papers we reviewed, published treatment descriptions were not considered sufficient to allow replication.
This is fairly simple to fix, during the publication of randomized trials, but it shows an important issue that significantly limits uptake of interventions in practice.
The key messages are:
- The majority of published trials in our study lacked important details describing the treatment required for healthcare professionals to undertake these treatments in practice.
- Although the majority of problems were not picked up by peer reviewers and editors, when they were detected only about two-thirds were fixed before publication.
- The incomplete treatment descriptions we found represent a substantial waste of the research budget, trial participants’ time and an opportunity cost for clinicians and patients.
Our conclusions are straightforward and state, ‘journals wanting to publish the research of use to practising healthcare professionals need to pay more attention to descriptions of treatments.’
Read the full paper at BMJ Open
Sara Schroter, Paul Glasziou, Carl Heneghan Quality of descriptions of treatments: a review of published randomised controlled trials BMJ Open 2012;2:e001978 doi:10.1136/bmjopen-2012-001978
Guest blog post by CEBM Director Dr Carl Heneghan, also posted on http://www.carlheneghan.com