Phase IV study of propofol: validation of the data set

Area of Science:

• Anesthesiology and Perioperative Medicine
• Clinical Epidemiology
• Pharmaceutical Research

Background:

• A large dataset from a 1989-1990 Phase IV study of propofol, involving over 26,000 patients, was made available for analysis.
• The study utilized data collected by 1819 physicians across 1761 hospitals, detailing anesthesia induction and maintenance methods.

Purpose of the Study:

• To describe the rigorous data verification process for a large Phase IV propofol study dataset.
• To establish a sound and reliable basis for subsequent epidemiological and clinical analyses of perioperative events and adverse reactions.

Main Methods:

• Data verification involved scrutinizing 46 recorded variables for accuracy, completeness, bias, and protocol adherence.
• Case reports with over 25% missing data were excluded, reducing the dataset from 26,841 to 25,981 patients.
• Statistical analysis confirmed that missing data occurred randomly and did not introduce significant bias.

Main Results:

• The final dataset comprised 25,981 patients, with minimal data entry inaccuracies and infrequent protocol violations.
• Adverse event profiles were consistent with findings from earlier Phase II and III clinical trials of propofol.
• The data verification process confirmed the dataset's validity for representing perioperative events in a broad patient population.

Conclusions:

• The verified Phase IV propofol dataset is robust and suitable for further analysis of perioperative events.
• Phase IV studies are valuable for their broad patient scope and ability to detect rare adverse events.
• Recommendations are made for improving future Phase IV study designs, particularly in data collection form efficiency and data entry processes.