Improvements in Data Quality Can Boost Efficiency and Reduce Development Costs: A Pharmacometric CRO's Perspective

Area of Science:

• Pharmacometrics and Drug Development
• Data Management in Clinical Research

Background:

• Drug development is a lengthy and expensive process heavily reliant on high-quality data.
• Contract Research Organizations (CROs) play a vital role in managing and analyzing data for drug sponsors.
• Model-Informed Drug Development (MIDD) requires robust and reliable datasets for accurate predictions and decision-making.

Purpose of the Study:

• To assess the impact of data management activities on Model-Informed Drug Development (MIDD) deliverables.
• To evaluate the baseline experience of CROs in assessing sponsor-provided data quality.
• To determine the time and cost associated with preparing analysis-ready datasets.

Main Methods:

• A survey was distributed to 44 pharmacometrics professionals across 32 companies offering data management services.
• The anonymous survey consisted of 11 questions (9 multiple-choice, 2 open-ended) focusing on data quality assessment and dataset preparation.
• Responses from 17 participating CROs were analyzed to identify common data quality challenges and time expenditures.

Main Results:

• 65% of CROs reported that sponsor-provided data was rarely usable (<10%) due to formatting issues, missing data, and inconsistencies.
• Over 50% identified a lack of clear data definitions and specifications as the primary cause of poor data quality.
• Data cleaning costs range from $750 to $6000 per dataset, requiring 3 to 24 hours of programming time.

Conclusions:

• Significant time and financial resources are invested by CROs in rectifying poor-quality sponsor data.
• While automated data quality checks enhance efficiency, they cannot fully address all data integrity issues.
• Improved communication, collaboration, and systematic approaches, including automation and AI, are essential for enhancing data quality in drug development.