The implementation and utility of patient screening logs in a multicentre randomised controlled oncology trial

Abstract

Abstract Background The utility of patient screening logs and their impact on improving trial recruitment rates are unclear. We conducted a retrospective exploratory analysis of screening data collected within a multicentre randomised controlled trial investigating chemotherapy for upper tract urothelial carcinoma. Methods Participating centres maintained a record of patients meeting basic screening criteria stipulated in the trial protocol, submitting logs regularly to the clinical trial coordinating centre (CTC). Sites recorded the number of patients ineligible, not approached, declined and randomised. The CTC monitored proportions of eligible patients, approach rate (proportion of eligible patients approached) and acceptance rate (proportion recruited of those approached). Data were retrospectively analysed to identify patterns of screening activity and correlation with recruitment. Results Data were collected between May 2012 and August 2016, during which time 71 sites were activated—a recruitment period of 2768 centre months. A total of 1138 patients were reported on screening logs, with 2300 requests for logs sent by the CTC and 47% of expected logs received. A total of 758 patients were reported as ineligible, 36 eligible patients were not approached and 207 declined trial participation. The approach rate was 91% (344/380), and the acceptance rate was 40% (137/344); these rates remained consistent throughout the data collection. The main reason patients provided for declining (99/207, 48%) was not wanting to receive chemotherapy. There was a moderately strong correlation (r = 0.47) between the number reported on screening logs and the number recruited per site. Considerable variation in data between centres was observed, and 54/191 trial participants (28%) enrolled during this period were not reported on logs. Conclusions Central collection of screening logs can identify reasons for patients declining trial participation and help monitor trial activity at sites; however, obtaining complete data can be challenging. There was a correlation between the number of patients reported on logs and recruitment; however, this was likely confounded by sites’ available patient population. The use of screening logs may not be appropriate for all trials, and their use should be carefully considered in relation to the associated workload. No evidence was found that central collection of screening logs improved recruitment rates in this study, and their continued use warrants further investigation. Trial registration ISRCTN98387754. Registered on 31 January 2012