Central coordination as an alternative for local coordination in a multicenter randomized controlled trial: the FAITH trial experience

Abstract

Background: Surgeons in the Netherlands, Canada and the US participate in the FAITH trial (Fixation using Alternative Implants for the Treatment of Hip fractures). Dutch sites are managed and visited by a financed central trial coordinator, whereas most Canadian and US sites have local study coordinators and receive per patient payment. This study was aimed to assess how these different trial management strategies affected trial performance.

Methods: Details related to obtaining ethics approval, time to trial start-up, inclusion, and percentage completed follow-ups were collected for each trial site and compared. Pre-trial screening data were compared with actual inclusion rates.

Results: Median trial start-up ranged from 41 days (P25-P75 10-139) in the Netherlands to 232 days (P25-P75 98-423) in Canada (p = 0.027). The inclusion rate was highest in the Netherlands; median 1.03 patients (P25-P75 0.43-2.21) per site per month, representing 34.4% of the total eligible population. It was lowest in Canada; 0.14 inclusions (P25-P75 0.00-0.28), representing 3.9% of eligible patients (p < 0.001). The percentage completed follow-ups was 83% for Canadian and Dutch sites and 70% for US sites (p = 0.217).

Conclusions: In this trial, a central financed trial coordinator to manage all trial related tasks in participating sites resulted in better trial progression and a similar follow-up. It is therefore a suitable alternative for appointing these tasks to local research assistants. The central coordinator approach can enable smaller regional hospitals to participate in multicenter randomized controlled trials. Circumstances such as available budget, sample size, and geographical area should however be taken into account when choosing a management strategy.

Trial registration: ClinicalTrials.gov: NCT00761813.

Figure 1

Inclusion progression for the Netherlands (NL), Canada (CA) and the United States (US).

Figure 2

The percentage decline in total number of patients seen per month, number of inclusions per month and percentage inclusions of total number of patients, during the trial period in comparison with the pre-trial screening period. To calculate this percentage decline, the total number of patients seen in the trial period was divided by the total number of patients seen in the pre-screening trial period. Similar calculations were made for the number of inclusions per month and the inclusion percentage. These rates (a) were transformed to a percentage decline (b) using the following formula: b = (1-a) * 100%. This figure therefore shows that for all variables there were fewer patients during the trial period compared with the pre-trial screening period in all countries. NL, the Netherlands; CA, Canada; US, United States.