Time-Based Data in Occupational Studies: The Whys, the Hows, and Some Remaining Challenges in Compositional Data...

Area of Science:

• Occupational Health Sciences
• Biostatistics
• Epidemiology

Background:

• Occupational research frequently utilizes time-use data, often expressed in hours or percentages, which are inherently constrained or 'compositional'.
• Traditional statistical methods may not be suitable for analyzing compositional data due to their properties, where parts sum to a constant.
• Compositional data analysis (CoDA) is established in fields like geology and chemistry but is emerging in public and occupational health.

Purpose of the Study:

• To introduce Compositional Data Analysis (CoDA) to researchers in public and occupational health.
• To explain the necessity and application of CoDA for analyzing compositional time-use data in occupational settings.
• To provide a practical guide, including a worked example, for implementing CoDA in occupational health research.

Main Methods:

• Explanation of the principles and rationale behind using CoDA for compositional time-use data.
• Demonstration of CoDA methodology with a practical, worked example relevant to occupational exposures.
• Discussion of current challenges and limitations in applying CoDA within occupational research.

Main Results:

• Compositional data analysis (CoDA) provides a robust framework for handling time-use data in occupational research.
• The application of CoDA is essential for accurate interpretation of time-based occupational exposures.
• The study highlights the nascent stage of CoDA in occupational health and the need for further development.

Conclusions:

• CoDA is a vital statistical approach for analyzing constrained time-use data in occupational health.
• Further research and practical application are needed to advance CoDA in occupational exposure studies.
• Adoption of CoDA by occupational researchers can enhance the understanding of work exposures and associated health outcomes.