Usefulness of Reference Change Values for Delta Check Limits in Clinical Laboratory Testing
10.15263/jlmqa.2020.42.3.121
- Author:
Seungok LEE
1
;
Jung O SON
;
Hyuk KWON
;
Kang Hoon PARK
;
Gyong Gi YU
;
Eunhee HAN
;
Dong Wook JEKARL
;
Yeongsic KIM
Author Information
1. Department of Laboratory Medicine, Incheon St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Incheon, Korea
- Publication Type:Original Article
- From:Journal of Laboratory Medicine and Quality Assurance
2020;42(3):121-129
- CountryRepublic of Korea
-
Abstract:
Background:In this study, the usefulness of within-subject biological coefficient of variation (CVI) and reference change values (RCVs) for delta check limits were investigated by comparing the population distributionbased delta check limits.
Methods:For six tests, including aspartate aminotransferase, alanine aminotransferase, γ-glutamyl transferase, glucose, creatinine, and hemoglobin, the RCV95%, RCV99%, and RCV99.9% delta limits were obtained. The nonparametric 95% and 99% delta limits were obtained from the population distribution of the delta percentage difference of the health examination group (January 2014 to December 2018) and the outpatient and inpatient groups (January to December 2018). Delta check alerts (%) in total and all three subgroups were examined according to the five different delta check limits. Additionally, we analyzed the correlation of the median CVIF estimates with population-delta check limits for the six tests.
Results:The delta percentage difference of the six tests showed a nonnormal distribution, and median value significantly differed among the health examination, outpatient, and inpatient groups (all, P <0.001). The overall delta check alerts of six tests decreased in the order of RCV95%, RCV99%, and RCV99.9%, population distribution -95%, and -99% delta limits; the proportion of the health examination group gradually decreased and that of inpatients increased. A good correlation was observed between median CVI (range, 2.7% to 10.1%) and population distribution delta limits (r =0.96 to 0.99).
Conclusions:The RCV delta check limits should be applied differently depending on the health and disease group. CVI can be useful for estimating the delta check limits of the population.
