Clinical trials are only as credible as their outcomes (Guest blog)
Every year, hundreds of thousands of members of the public are recruited into tens of thousands of clinical trials in health and social care, and thousands of researchers prepare systematic reviews to bring together the findings of existing trials. Much of this effort is wasted if the outcomes measured and reported are not those needed by health service users, practitioners, health technology assessment (HTA) agencies, policy makers and others, making choices between or reimbursement decisions about different interventions, actions and strategies. Considering Peter Tugwell’s view that “Clinical trials are only as credible as their outcomes”, one can see this is equally true for systematic reviews, clinical guidelines, health technology assessments, and healthcare organisations.
This problem could be addressed with the development and application of agreed standardised sets of outcomes, known as ‘core outcome sets’ (COS). A COS should be measured and reported, as a minimum, in all trials undertaken in a specific health condition. On a daily basis, HTA bodies around the world face the consequences of less than optimal adoption of COS in primary studies.
The uptake of COS will be improved through greater dissemination and accessibility to relevant information such as that provided by the COMET Initiative (http://www.comet-initiative.org). COMET hosts a publically available, searchable database containing details of 300 published and 146 ongoing studies. Since inception, there have been nearly 49,000 unique visitors to the website from 168 countries, with 13,616 database searches undertaken.
COS research faces certain challenges. The credibility of a COS depends on the use of sound methodology in its development and transparent reporting of the process. We have recently published a reporting guideline for COS studies (http://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002148) but to date there is no guidance on the minimum standards for developing a COS. We have an ongoing study to define these standards – please access the online survey here if you are interested in participating: http://ctrc.liv.ac.uk/DelphiManager/COS-minimum-standards/, open until the end of December.
On the 10-11th November, 170 people gathered at Rode Hoed in Amsterdam for the sixth meeting of the COMET Initiative. The session on people and patient participation, involvement and engagement was particularly well-received. Previous work by CADTH has shown that trials do not always measure outcomes that patients, carers, patient support group representatives, and service users consider important (1). Just over a quarter of the published COS in the COMET database included patient participants, increasing to 93% in ongoing studies. Thus the question is no longer whether to include patients views, but rather how. The COMET PoPPIE (People and Patient Participation, Involvement and Engagement) Group is developing resources to help, http://www.comet-initiative.org/resources/publicinvolvement.
Maarten de Witt (VU University Medical Centre) argued that we still have a long way to go. He reported that patients will often change their opinions to agree with clinicians on outcomes of importance, but not vice versa, highlighting the importance of preserving patient perspective and challenging power imbalances and disproportional representation when seeking the patient’s voice. Rebecca Craven, a public contributor, tweeted from the meeting that “COMET speakers are great at using plain language to get across their ideas. It really helps all audience to understand and follow.”
Another session that drew attention was focused on COS uptake. Implementation will depend on effective dissemination and its acceptance by the research community including trialists, patients and the public, trial funders, industry, regulators, systematic reviewers, guideline developers, journal editors, HTA bodies and payers. Alison Cuff (ISRCTN Registry) discussed the potential role a clinical trials registry could play in promoting COS. Rhonda Facile (Clinical Data Interchange Standards Consortium) discussed why clinical data standards are needed and how COS in the COMET database have been incorporated into CFAST projects to ensure alignment. Richard Veselý (European Medicines Agency) provided examples of situations when COS are recommended for evaluation of efficacy by the European Medicines Agency (2).
COS may have relevance early on in the research pathway. An awareness of COS will be useful to horizon scanning agencies and networks when they are filtering, prioritising and assessing emerging health technologies. Identifying the absence of core outcomes in clinical trials at an early stage of product development could be useful to organisations involved in early dialogues, to prevent trials being conducted that do not address relevant questions.
COS may also have implications beyond comparative effectiveness trials. Considering the big data sets now being assembled, for example using multiple electronic health records to undertake real world studies or bringing together national research registries, collecting the same core outcomes will facilitate comparing and pooling.
Once again, the COMET meeting was successful in bringing people together with interests in COS, including students, researchers, clinical experts, patients and COS developers. The slides from COMET VI presentations will be available on the website shortly.
- Berglas S et al (2016) Patients’ perspectives can be integrated in health technology assessments: an exploratory analysis of CADTH Common Drug Review. Research Involvement and Engagement 2:21
- Note for Guidance on Clinical Investigation of Medicinal 4 Products for Treatment of Asthma (EMA/CHMP/EWP/2922/01 Rev.1)