Statistical criteria for separate reference intervals: race and gender groups in creatine kinase

Area of Science:

• Clinical Biochemistry
• Biostatistics
• Population Health

Background:

• Previous studies identified race and gender differences in creatine kinase (CK) levels.
• Establishing appropriate reference intervals for diverse populations is crucial for accurate clinical interpretation.
• The Los Angeles area served as the population sample for the initial data.

Purpose of the Study:

• To re-evaluate previously published creatine kinase (CK) data using new statistical criteria.
• To assess the suitability of proposed statistical methods for defining separate reference intervals.
• To determine if existing criteria adequately account for demographic variations in CK levels.

Main Methods:

• Re-analysis of existing creatine kinase (CK) data from Los Angeles area subgroups.
• Application of recently proposed statistical criteria for defining reference intervals.
• Comparative analysis of different statistical criteria's stringency.

Main Results:

• One statistical criterion was found to be potentially too lenient in suggesting separate intervals.
• Another statistical criterion was deemed too restrictive for defining necessary separate intervals.
• The study highlights variability in the effectiveness of different statistical approaches.

Conclusions:

• Current statistical criteria for establishing reference intervals require further refinement.
• Experience with diverse analytes and population sizes is needed to validate these criteria.
• Careful selection of statistical methods is essential for accurate clinical biochemistry interpretation.