Collaborative derivation of reference intervals for major clinical laboratory tests in Japan

Clinical laboratories rely on reference intervals to interpret test results, but these values can vary by population. Prior research has shown that reference intervals are often population-specific, leading to inconsistencies in diagnosis and treatment. In Japan, there was a lack of harmonized reference intervals for major clinical tests. Existing data were fragmented, with limited agreement on age and sex adjustments. This gap motivated the need for a unified approach to reference interval derivation. No prior work had resolved the issue of combining multiple datasets to create a national standard. The Japan Society of Clinical Chemistry recognized the need for standardized intervals to improve diagnostic accuracy. The challenge lay in ensuring statistical consistency across studies and accounting for demographic variables. This study aimed to address these limitations by pooling data from three multicentre trials.

The goal was to derive common reference intervals for 40 clinical tests using data from three multicentre studies in Japan. The researchers sought to harmonize data collection and statistical methods across the studies. They aimed to create intervals that would be applicable nationwide. The motivation stemmed from inconsistencies in current reference ranges and the need for age- and sex-specific adjustments. By combining datasets, they hoped to increase statistical power and reduce variability. The study focused on individuals with no medications and a BMI ≤28 to ensure healthy reference populations. The team wanted to assess whether sex and age adjustments were necessary for each test. Their objective was to provide a standardized framework for clinical interpretation.

The study merged data from three multicentre trials in Japan, each following comparable protocols. Participants were primarily hospital workers with BMI ≤28 and no medications. Age and sex distributions were balanced across datasets to reach a total of 6345 individuals. The SD ratio was calculated to assess between-subgroup variability. Statistical methods included parametric analysis and latent abnormal value exclusion. Reference intervals were derived using the parametric method with age and sex adjustments. The SD ratio ≥0.50 was used to determine the need for sex-specific intervals. Data were pooled after confirming between-study differences were within acceptable limits.

Merged datasets showed acceptable between-study variability, allowing for combined analysis. Sex-specific intervals were needed for 12 assays based on SD ratio ≥0.50. Five analytes required age-specific intervals for females aged 45 and above. Latent abnormal value exclusion narrowed reference ranges for 10 items linked to common disorders. Reference intervals showed unique sex- and age-related patterns for each analyte. The parametric method improved interval precision by excluding latent abnormalities. The derived intervals reflect healthy population profiles with no abnormal test results. These findings suggest that age and sex adjustments are critical for accurate interpretation.

The study successfully derived common reference intervals for 40 tests using data from three multicentre studies. Statistical methods ensured consistency across datasets and accounted for demographic factors. The findings suggest that sex-specific and age-specific intervals are necessary for 12 and five analytes, respectively. The parametric method improved interval accuracy by excluding latent abnormalities. The derived intervals provide a standardized framework for national use in Japan. These intervals are relevant for interpreting test results and applying clinical guidelines. The study confirms the importance of age and sex adjustments in reference interval derivation. The results support the use of harmonized intervals to improve diagnostic reliability.