SMOOTH (Self-Management of Open Online Trials in Health) analysis found improvements were needed for reporting methods of internet-based trials.
Price A., Vasanthan L., Clarke M., Liew SM., Brice A., Burls A.
BACKGROUND: The growth of trials conducted over the internet has increased, but with little practical guidance for their conduct and it is sometimes challenging for researchers to adapt the conventions used in face-to-face trials and maintain the validity of the work. AIM: To systematically explore existing self-recruited online randomized trials of self-management interventions and analyze the trials to assess their strengths and weaknesses, the quality of reporting and the involvement of lay persons as collaborators in the research process. METHODS: The Online Randomized Controlled Trials of Health Information Database (ORCHID) was used as the sampling frame to identify a subset of self-recruited online trials of self-management interventions. The authors cataloged what these online trials were assessing, appraised study quality, extracted information on how trials were run and assessed the potential for bias. We searched out how public and patient participation was integrated into online trial design and how this was reported. We recorded patterns of use for registration, reporting, settings, informed consent, public involvement, supplementary materials, and dissemination planning. RESULTS: The sample included 41 online trials published from 2002-2015. The barriers to replicability and risk of bias in online trials included inadequate reporting of blinding in 28/41 (68%) studies; high attrition rates with incomplete or unreported data in 30/41 (73%) of trials; and 26/41 (63%) of studies were at high risk for selection bias as trial registrations were unreported. The methods for (23/41, 56%) trials contained insufficient information to replicate the trial, 19/41 did not report piloting the intervention. Only 2/41 studies were cross-platform compatible. Public involvement was most common for advisory roles (n=9, 22%), and in the design, usability testing and piloting of user materials (n=9, 22%) CONCLUSIONS: This study catalogs the state of online trials of self-management in the early 21st century and provides insights for online trials development as early as the protocol planning stage. Reporting of trials was generally poor and, in addition to recommending that authors report their trials in accordance with CONSORT guidelines, we make recommendations for researchers writing protocols, reporting on and evaluating online trials. The research highlights considerable room for improvement in trial registration, reporting of methods, data management plans, and public and patient involvement in self-recruited online trials of self-management interventions.