Protocolization of multicenter clinical studies in the digital era. Is useful data centralization by a data-manager?

Area of Science:

• Oncology
• Clinical Research Methodology
• Data Management

Background:

• Multicenter studies require rigorous data protocolization to prevent biases.
• Data integrity is crucial for the validity of clinical trial outcomes.

Purpose of the Study:

• To compare data concordance and reliability between local and centralized protocolization in a multicenter clinical study.
• To assess the impact of data management on data quality in oncology research.

Main Methods:

• A national multicenter study on an infrequent carcinoma was conducted.
• Data underwent double protocolization: at the center of origin and by a central data manager.
• Interobserver variability was assessed using Cohen's kappa coefficient.

Main Results:

• High concordance was observed for tumor stage (Kappa=0.931), recurrence risk (Kappa=0.819), and sonography evaluation (TIRADS, Kappa=0.922).
• Disagreements were low, with tumor stage disagreements at 2.5% and recurrence risk at 7%.
• Transcription errors were detected in 4.6% of cases.

Conclusions:

• Centralized data protocolization by a data manager yields results comparable to direct protocolization at the center of origin.
• This approach supports reliable data collection in multicenter oncology studies.
• Centralized management appears effective in maintaining data quality and consistency.