All participants in research are important. What patients in clinical trials tell us about treatments – patient-reported outcomes (PROs) such as quality of life and symptoms – is being used more and more to improve decisions about patient-centered care and health policy, but poor reporting of these outcomes risks research effort being wasted. Today, PLOS ONE publishes two important papers concerning patient-reported outcomes.[1,2] Professor Melanie Calvert at the University of Birmingham and her colleagues across the UK, Canada and Australia found that not enough attention has been devoted to PROs in clinical trial protocols, the blueprints for how a trial is run and reported.
Patient-reported outcomes are now getting the focused attention they deserve. In the United States the Patient-Centered Outcome Research Institute was set up in 2010 to conduct “research that answers patients’ questions”, spending nearly $700 million USD on 360 projects. The Canadian Institutes of Health Research recently established the Strategy for Patient-Oriented Research. However, PROs must be collected rigorously from the beginning to minimise loss of research information. CONSORT, the Consolidated Standards of Reporting Trials, has long been the gold standard for trial reporting. In 2013 the CONSORT-PRO Extension was introduced, promoting the transparent reporting of PRO data from trials. However, this guidance is aimed at the end of the research process and may be too late to make researchers think more about patients earlier on – at the protocol stage. Guidance on preparing trial protocols themselves has been developed – the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) checklist was also published in 2013. However, this does not offer PRO-specific guidance.
In the first paper, Calvert and colleagues review the existing guidance for protocol writers on PROs. They found that guidance was hard to access, lacked consistency and may be challenging to put into practice. In total they uncovered a confusing mix of 162 recommendations across 54 publications.
In the second paper, Kyte and colleagues studied 75 protocols funded by the UK National Institutes of Health Research’s Health Technology Assessment program in 2012-3. About two-thirds of the protocols complied with reporting SPIRIT items. Most troubling was that two of items about randomization – the unique characteristic of randomized trials – were missing in about a fifth of the protocols. Anything less than 100% is avoidable waste and the funder is not getting value for money. More than a third of the sample selected was from 2012, prior to the introduction of SPIRIT, however the NIHR website’s guidance for prospective protocol developers still only suggests using SPIRIT (in section 4.4). This is in contrast to a stronger endorsement of PROSPERO, a registry for systematic review protocols. Clearly there is an opportunity to more strongly recommend using SPIRIT, incorporating PRO issues at the same time.
These papers point to a clear need for more attention to PROs when developing trial protocols, and Prof. Calvert and colleagues used rigorous evidence-based methods that were explicitly reported in sufficient detail to allow interested readers to replicate them.
One suggestion to improve reporting is the development of a SPIRIT PRO extension. The EQUATOR Network already has a library of more than 200 reporting guidelines, with lots more in development. Reporting guideline developers have worked hard to help authors more completely report their research, but the marketplace is getting full. The SPIRIT-PRO extension would need to be both accessible and acceptable to researchers who wish to incorporate PROs in their trials and we need to ensure we don’t make using guidelines, such as a SPIRIT PRO extension, so complicated that it becomes a barrier to using them. We need to consider how to best ease using reporting guidelines: technology might be helpful here.
Another way to increase the use of SPIRIT and PRO guidance would be to build them into teaching about trials in university graduate programs, such as epidemiology and biostatistics. However this is done, researchers need to improve the reporting of patient-reported outcomes for the sake of both the trial participants and the patients who rely on this evidence to improve their health.
Citations:  Calvert M et al. (2014) Patient-Reported Outcome (PRO) Assessment in Clinical Trials: A Systematic Review of Guidance for Trial Protocol Writers. PLOS ONE 9(10): e110216 doi:10.1371/journal.pone.0110216
 Kyte D et al. (2014) Systematic evaluation of the patient-reported outcome (PRO) content of clinical trial protocols. PLOS ONE 9(10): e110299. doi:10.1371/journal.pone.0110229
Image: Hands_II – Oleg Afonin
David Moher is a senior scientist in the Clinical Epidemiology Program, Ottawa Hospital Research Institute, and Associate Professor, Department of Epidemiology and Community Medicine, University of Ottawa, where he holds a University Research Chair. He has a special interest in journalology (publication science), including ways to improve the quality of reporting health research. He spearheaded the development of the CONSORT Statement for reporting randomized trials, the PRISMA Statement for reporting systematic reviews and meta-analyses, and has been involved in many other reporting guideline initiatives, including SPIRIT and CONSORT PRO. Dr. Moher is associated with many journals: an editor-in-chief of Systematic Reviews; an editorial board member of several journals, including PLOS Medicine, a member of PLOS ONE’s Human Research Advisory Group, the advisory board for the International Congress on Peer Review and Biomedical Publication, and a member of the EQUATOR Network’s steering group.