Skip to main content

Table 1 Opinions and practices of participants regarding evidence-based planning of future trials

From: Synthesizing existing evidence to design future trials: survey of methodologists from European institutions

Question Possible answers Responses (%)
What is your primary affiliation? Clinical trials unit 29 (38%)
A funding body 3 (4%)
Pharmaceutical industry 15 (20%)
HTA/Cochrane/WHO 28 (37%)
Missing 1 (1%)
How do you judge whether a summary treatment effect provides conclusive evidence or whether further research is needed (more than one choice allowed)? I examine the statistical significance of the summary effect and its CI 31 (46%)
I examine the clinical importance of the summary effect and its CI 39 (57%)
I test whether future studies could change the statistical significance of the summary effect 7 (10%)
I follow the GRADE guidelines for judging imprecision 19 (28%)
Not involved in interpretation of meta-analysis results/other/missing 29 (43%)
Do you think that network meta-analysis should be considered as the preferred evidence synthesis method instead of pairwise meta-analysis? Yes, network meta-analysis should always be preferred 15 (22%)
No, network meta-analysis should not be considered 5 (7%)
It should be considered only if there are no or few direct studies 25 (37%)
Other/missing 23 (34%)
According to your experience, results from relevant meta-analyses are considered to (more than one choice allowed): Define the alternative effect size in power calculations 25 (58%)
Decide about the intervention in the comparator arm 19 (44%)
Define other parameters involved in sample size calculations 29 (67%)
Define health outcomes to be monitored 22 (51%)
Other/missing 7 (16%)
What do you think is the biggest barrier towards adopting the conditional trial design in designing trials? Lack of training 6 (14%)
Changing the paradigm of funders and researchers 16 (37%)
Lack of good-quality meta-analyses 4 (9%)
Other/missing 17 (40%)
Question Research proposals Median (25th to 75th percentile)
As a citizen supporting publicly funded research how would you rank (from 1 being the top priority to 5 being the least) the following proposals tackling the treatments for an important health condition? Consider also the cost for each research proposal (presented in parenthesis in arbitrary units). A well-powered three-arm randomized trial comparing the three most promising interventions (none of which is standard care) (100) 4.0 (3.0 to 5.0)
A well-powered three-arm randomized trial comparing the two most promising interventions and standard treatment (90) 2.0 (1.0 to 2.0)
A well-powered two-arm randomized trial comparing a newly launched treatment and standard treatment (70) 3.0 (2.0 to 4.0)
A large registry involving many countries (40) 5.0 (3.5 to 5.0)
A network meta-analysis comparing all available treatments using existing studies (10) 1.5 (1.0 to 3.0)
  1. The full text and questions are presented in Additional file 2
  2. CI confidence interval, GRADE Grading of Recommendations Assessment, Development and Evaluation, HTA health technology assessment, WHO World Health Organization