Table 1.
Activities comprising the consensus meeting and the time dedicated to each.
Table 2.
Data stimulation scenarios in which: μP (or μA) denotes the true long-run average change from baseline pain score at 26 weeks on pamidronate (or adalimumab); σ is the common standard deviation of change from baseline pain scores.
Table 3.
Expert responses to elicitation questions.
Fig 1.
Consensus prior densities for outcome model parameters.
(Fig 1B) shows consensus prior distributions for the average change in pain scores at 26 weeks on pamidronate or adalimumab. Prior modal values for the average change in pain score on pamidronate and adalimumab are -32.3 mm and -30 mm, respectively. (Fig 1C) displays the consensus prior for the treatment effect. Prior opinion is that there is a 90% chance that the true difference between average change scores on adalimumab versus pamidronate lies between -6.9 mm and 11.5 mm (with a negative difference indicating an advantage for adalimumab). Furthermore, there is a 68.4% prior probability that pamidronate is superior to adalimumab by any margin. However, it is clear that there is considerable uncertainty about the precise relative merits of the two treatments. (Fig 1A), plotted symbols represent the experts’ consensus answers to elicitation questions QA2-5 and QP2-5 minus 60mm, which was the pain score assumed to characterise a typical patient at baseline.
Fig 2.
Posterior densities for model parameters.
Table 4.
Simulated trials.