Think twice before starting a new trial; what is the impact of recommendations to stop doing new trials?

Abstract

In evidence-based medicine, we base our conclusions on the effectiveness of interventions on the results of high-quality meta-analysis. If a new randomized controlled trial (RCT) is unlikely to change the pooled effect estimate, conducting the new trial is a waste of resources. We evaluated whether recommendations not to conduct further RCTs reduced the number of trials registered for two scenarios. Analysis of registered trials on the World Health Organisation (WHO) International Clinical Trials Registry Platform (ICTRP). We regarded trial protocols relevant if they evaluated the effectiveness of (1) exercise for chronic low back pain (LBP) and (2) cognitive behavioural therapy (CBT) for chronic pain. We calculated absolute and relative numbers and change of registered trials in a pre-set time window before and after publication of the recommendations, both published in 2012. We found 1,574 trials registered in the WHO trial registry for exercise in LBP (459 before 2012; 1,115 after) and 5,037 trials on chronic pain (1,564 before 2012; 3,473 after). Before 2012, 13 trials on exercise for LBP (out of 459) fit the selection criteria, compared to 42 trials (out of 1,115) after, which represents a relative increase of 33%. Twelve trials (out of 1,564) regarding CBT for chronic pain, fit the selection criteria before 2012 and 18 trials (out of 3,473) after, representing a relative decrease of 32%. We found that visibility, media exposure and strength of the recommendation were related to a decrease in registered trials. Recommendations not to conduct further RCTs might reduce the number of trials registered if these recommendations are strongly worded and combined with social media attention.