Within-subject biological variation estimated using real-world data strategies (RWD): a systematic review.

Area of Science:

• Clinical Chemistry
• Biomarker Analysis
• Laboratory Medicine

Background:

• Biological variation (BV) quantifies changes in measurand concentrations around a homeostatic set point.
• BV is essential for establishing analytical performance specifications, reference change values, and personalized reference intervals.
• Traditional BV studies are often laborious, expensive, and based on limited, ideal populations.

Purpose of the Study:

• To explore the application of Real-World Data (RWD) strategies for estimating biological variation.
• To compare RWD methods with traditional direct methods for BV studies.
• To evaluate proposed RWD algorithms for biomarker analysis.

Main Methods:

• Review of existing literature on biological variation estimation.
• Analysis of Real-World Data (RWD) strategies using Laboratory Information System (LIS) data.
• Examination of three proposed RWD algorithms (Loh et al., Jones et al., Marques-Garcia et al.) and the BiVaBiDa project algorithm.

Main Results:

• RWD methods offer a population-based approach to BV estimation, overcoming limitations of direct methods.
• The BiVaBiDa algorithm is identified as the most complete RWD approach, enabling subgroup analysis and confidence interval calculation.
• RWD methods allow for more detailed population subgroup analysis compared to direct methods.

Conclusions:

• RWD methods present a promising alternative for biological variation studies, enhancing biomarker analysis.
• Further development of RWD algorithms is needed to improve robustness and address data anonymization and standardization challenges.
• RWD strategies have the potential to significantly advance personalized medicine through more accurate biomarker interpretation.