Kiyoshi Ichihara1, Yoshikazu Yomamoto2, Taeko Hotta3, Shigemi Hosogaya4, Hayato Miyachi5, Yoshihisa Itoh6, Midori Ishibashi7, Dongchon Kang8. 1. Department of Laboratory Sciences, Faculty of Health Sciences, Yamaguchi University Graduate School of Medicine, Ube, Japan ichihara@yamaguchi-u.ac.jp. 2. Department of Clinical Laboratory Science, Tenri Health Care University, Tenri, Japan. 3. Clinical Laboratory, Kyushu University Hospital, Fukuoka, Japan. 4. Department of Medical Technology, Faculty of Health Sciences, Tokyo University of Technology, Tokyo, Japan. 5. Department of Laboratory Medicine, Tokai University School of Medicine, Isehara, Japan. 6. Clinical Laboratory, Eiju General Hospital, Tokyo, Japan. 7. Clinical Laboratory, New Tokyo Hospital, Chiba, Japan. 8. Department of Laboratory Medicine, Kyushu University School of Medicine, Fukuoka, Japan.
Abstract
OBJECTIVES: Three multicentre studies of reference intervals were conducted recently in Japan. The Committee on Common Reference Intervals of the Japan Society of Clinical Chemistry sought to establish common reference intervals for 40 laboratory tests which were measured in common in the three studies and regarded as well harmonized in Japan. METHODS: The study protocols were comparable with recruitment mostly from hospital workers with body mass index ≤28 and no medications. Age and sex distributions were made equal to obtain a final data size of 6345 individuals. Between-subgroup differences were expressed as the SD ratio (between-subgroup SD divided by SD representing the reference interval). Between-study differences were all within acceptable levels, and thus the three datasets were merged. RESULTS: By adopting SD ratio ≥0.50 as a guide, sex-specific reference intervals were necessary for 12 assays. Age-specific reference intervals for females partitioned at age 45 were required for five analytes. The reference intervals derived by the parametric method resulted in appreciable narrowing of the ranges by applying the latent abnormal values exclusion method in 10 items which were closely associated with prevalent disorders among healthy individuals. Sex- and age-related profiles of reference values, derived from individuals with no abnormal results in major tests, showed peculiar patterns specific to each analyte. CONCLUSION: Common reference intervals for nationwide use were developed for 40 major tests, based on three multicentre studies by advanced statistical methods. Sex- and age-related profiles of reference values are of great relevance not only for interpreting test results, but for applying clinical decision limits specified in various clinical guidelines.
OBJECTIVES: Three multicentre studies of reference intervals were conducted recently in Japan. The Committee on Common Reference Intervals of the Japan Society of Clinical Chemistry sought to establish common reference intervals for 40 laboratory tests which were measured in common in the three studies and regarded as well harmonized in Japan. METHODS: The study protocols were comparable with recruitment mostly from hospital workers with body mass index ≤28 and no medications. Age and sex distributions were made equal to obtain a final data size of 6345 individuals. Between-subgroup differences were expressed as the SD ratio (between-subgroup SD divided by SD representing the reference interval). Between-study differences were all within acceptable levels, and thus the three datasets were merged. RESULTS: By adopting SD ratio ≥0.50 as a guide, sex-specific reference intervals were necessary for 12 assays. Age-specific reference intervals for females partitioned at age 45 were required for five analytes. The reference intervals derived by the parametric method resulted in appreciable narrowing of the ranges by applying the latent abnormal values exclusion method in 10 items which were closely associated with prevalent disorders among healthy individuals. Sex- and age-related profiles of reference values, derived from individuals with no abnormal results in major tests, showed peculiar patterns specific to each analyte. CONCLUSION: Common reference intervals for nationwide use were developed for 40 major tests, based on three multicentre studies by advanced statistical methods. Sex- and age-related profiles of reference values are of great relevance not only for interpreting test results, but for applying clinical decision limits specified in various clinical guidelines.
Authors: Abigail S A Bawua; Kiyoshi Ichihara; Rosemary Keatley; John Arko-Mensah; Yvonne Dei-Adomakoh; Patrick F Ayeh-Kumi; Rajiv Erasmus; Julius Fobil Journal: Int J Lab Hematol Date: 2020-09-03 Impact factor: 2.877