Rethinking the communication of clinical trial results
If clinical trial results were more quickly and openly accessible, medical improvements would be implemented more swiftly too. So what can be done to ease this process? What isn’t working and how can we change it?
It’s clear that people are continually adapting and evolving how they consume information, and businesses need to continually evolve to reflect this – just look at how social media and websites such as Craigslist have changed the newspaper business. But is scientific publishing keeping up with this change? And, if not, what other options can be used in parallel to communicate clinical trial results data?
Results from clinical trials are shared for two fundamental reasons – to enable clinicians to use the knowledge gained to advance clinical care, and because patients take part in clinical trials with the expectation that the results will be used to improve treatments for themselves and others.
Clinical data therefore need to be communicated quickly and openly. Publishing them in scientific journals will continue to play a key part in this process, but it is important to consider any limitations and how to overcome them.
Challenges with the current model
Frustrating delays: Authors often comment that the communication of trial data, based on publication in conventional, peer-reviewed scientific journals, can be slow. The time taken from submission to publication of a manuscript, typically 6 to 12 months, means that important scientific research is not shared quickly enough in today’s era of rapid communication.
Can’t see the wood for the trees: Publication of results in one of many thousands of journals, that readers often have to pay to access, also means that data are difficult to find and may not reach the relevant audience.
Lack of innovation: The limited format of standard journal articles, often restricted to a certain number of pages and figures, makes it difficult for readers to gain a complete picture of the trial’s findings, while the static nature of these publications discourages iterations, creativity and interactivity. Crucially, this also means that journals may not reflect how readers prefer to consume information in an increasingly connected, and socially integrated, world.
Bias: As humans we naturally prioritise exciting, positive or surprising findings over the negative or mundane, and it’s no surprise that this results in bias, unconscious or conscious, in the study data that are selected for publication. Everyone contributes towards this bias: authors, reviewers, journals and even readers. Although vehicles do exist to publish less ‘exciting’
data, it’s almost impossible to eliminate bias completely and so we need alternative ways of sharing the totality of data.
‘Although the publication of clinical trial data in scientific journals will remain the gold standard, it is important to consider how this can be improved to ensure all audiences have quick, easy and open access’
How can the model change?
The International Committee of Medical Journal Editors (ICMJE) published the following requirements last year, showing that there may be light at the end of the tunnel:
- From 1 July 2018, clinical trial manuscripts submitted to ICMJE journals must contain a data-sharing statement that explains what patient data will be shared (e.g. patient-level data), what additional documents will be made available (e.g. clinical study reports or protocols), how these can be accessed, and when this will happen.
- Clinical trials that start enrolling patients after 2019 need to include a data-sharing plan in the trial’s registration.
These requirements are not as stringent as might have been expected, and do not make data sharing mandatory, but they certainly signal a shift towards greater openness.
Of course the ClinicalTrials.gov website mandates interventional trials are registered there to report results on the site within one year of completion. However, an audit in 2012 established that a remarkable 88% of studies registered on ClinicalTrials.gov had failed to do so.
Some journals have developed digital platforms in the hope of overcoming the challenges of traditional scientific publications. They offer features such as very fast publication, no length restrictions, links with social media, open peer review and pre-peer review publication. These are gaining some recognition, as shown by the Wellcome Trust’s launch of an in-house journal to ensure rapid publication of data from its funded research, to encourage transparency and reproducibility.
There is also the option of using a pre-print server such as bioRxiv to share publications and gain comments before submission to peer-reviewed journals. However, although these are being more widely used, they have yet to gain broad acceptance for sharing clinical trial manuscripts.
Platforms such as these are starting to improve the sharing of clinical data, but have yet to significantly change the behaviour of most investigators and authors.
‘Advances towards more transparent and effective data sharing include proposals by the ICMJE and new digital publishing platforms, but more needs to be done’
What should clinical trial data communication look like in the future?
Groups such as Alltrials have been working to ensure that results from clinical trials are fully reported, which is obviously a great aim. The downside, though, is that the pharmaceutical industry has sometimes been portrayed as being reluctant, over-cautious and opaque when it comes to communicating clinical data.
There is, however, a clear opportunity for pharmaceutical companies, and forward-thinking medical communications agencies, to be leaders in shaping how clinical data are shared with the wider medical community. We strive for a future where trial investigators have open access to all clinical data, and use this to prepare compelling educational communications that provide an overview of a clinical development programme, or place new data into more relevant context that reflects the real world in which patients live and HCPs practice.
These communications would be published rapidly in openly accessible scientific journals that provide opportunities for innovation and interaction, backed by easily available full datasets from every study, including completed trials. Crucially, this also enables the development of innovative approaches to harness, analyse and activate the wealth of clinical data that are currently hidden to provide new advances into diagnosis and treatment.
Making changes in how clinical data are communicated and shared will not necessarily be easy, but the need to incorporate a transparent, efficient and consistent method of sharing trial data is increasingly recognised across academia, industry and public organisations. Together, pharmaceutical companies and medical communications agencies should take the opportunity to be at the forefront of this move towards excellence in sharing data with the wider medical community.
This will ensure that the right information is quickly and efficiently communicated to the right audience using the most appropriate format and channel, and with valuable context, so building trust and, ultimately, making a positive change to patients’ lives.
About the author:
Mike Thompson is a Senior Scientific Director at Complete HealthVizion. He is an experienced medical communications professional who provides editorial, scientific and strategic insight into communications and publications across a range of therapy areas and clients.