Impact of on-site initiation visits on patient recruitment and data quality in a randomized trial of adjuvant chemotherapy for breast cancer

International audiencePURPOSE: To provide empirical evidence on the impact of on-site initiation visits on the following outcomes: patient recruitment, quantity and quality of data submitted to the trial coordinating office, and patients' follow-up time. PATIENTS AND METHODS: This methodological study was performed as part of a randomized trial comparing two combination chemotherapies for adjuvant treatment of breast cancer. Centers participating to the trial were randomized to either receive systematic on-site visits (Visited group), or not (Non-visited group). RESULTS: The study was terminated after two years, while the main randomized trial continued. Of the 135 centers that had expressed an interest in the trial, only 69 randomized at least one patient (35/68 in the Visited group, 34/67 in the Non-visited group). Almost two-thirds of the patients were entered by 17 centers (10 in the Visited group, seven in the Non-visited group) that accrued more than 10 patients each. None of the prespecified outcomes favored the group of centers submitted to on-site initiation visits (ie, mean number of queries par patient: 6.1 +/- 9.7 versus 5.4 +/- 6.4, respectively for the Visited and Non-visited groups). Spontaneous transmittal of case report forms, although required by protocol, was low in both randomized groups (mean number of pages per patient: 1.5 +/- 2.0 versus 2.1 +/- 2.3, respectively), with investigators submitting about one-third of the expected forms on time (29% and 39%, respectively). LIMITATIONS: This study could not evaluate the impact of repeated on-site visits on clinical outcomes. CONCLUSION: Systematic on-site initiation visits did not contribute significantly to this clinical trial