A systematic approach for evaluating and scoring human data.

Area of Science:

• Toxicology and Chemical Safety
• Regulatory Science
• Data Management

Background:

• Existing frameworks for animal data quality assessment (Klimisch et al.) lack a direct human data counterpart.
• Need for systematic, transparent, and repeatable methods for evaluating human data quality in regulatory contexts.
• Challenges in harmonizing diverse human data sources for weight-of-evidence assessments.

Purpose of the Study:

• To describe a systematic approach for assessing and categorizing the quality of human data.
• To complement existing animal data quality frameworks and facilitate transparent weight-of-evidence comparisons.
• To propose definitions for data quality and adequacy and a categorization scheme for quality.

Main Methods:

• Development of a quality assessment scheme for human data, inspired by Klimisch et al.'s animal data criteria.
• Proposal of specific definitions for data quality and adequacy.
• Delineation of four distinct categories for data quality assessment.

Main Results:

• A systematic approach for human data quality assessment and categorization has been established.
• The scheme provides definitions for data quality and adequacy, with quality differentiated into four categories.
• The approach is demonstrated for evaluating data reliability, particularly for IUCLID database entries, and data relevance.

Conclusions:

• The proposed approach offers a standardized method for evaluating human data quality and reliability.
• It facilitates transparent and repeatable weight-of-evidence comparisons, crucial for regulatory submissions (e.g., IUCLID).
• The framework aims to harmonize human data evaluation processes globally, improving consistency and scientific rigor.