Inter- Laboratory Variability of Haemoglobin Measurements Obtained From Selected Clinical Laboratories in Kenya

Abstract

Accurate determination of haemoglobin (Hb) concentration is a common element in assessing the extent of anaemia and an important variable in directing transfusion therapy in patients. Laboratory measurements should be reproducible and consistent from day to day as well as between laboratories so that comparable results will be obtained when a given specimen is tested in different laboratories. Variability in Hb measurements is caused by many factors such as laboratory error, numerous physiologic, temporal and methodologic factors. Therefore this study was aimed at determining the interlaboratory variability of Hb measurements obtained from selected clinical laboratories across Kenya. A total of 188 public and 105 private laboratories selected from a total of 21 out of 47 Kenyan Counties received three specimens with low (A), normal (B) and high (C) Hb concentrations for analysis, after which their results were compared and evaluated for accuracy. Laboratory performance was assessed using the Clinical Laboratory Improvement Amendments of 1988 (CLIA’88) criteria. Coefficient of Variation (CV) was calculated as a measure of interlaboratory variability while the accuracy of the analysers was evaluated using one-way analysis of variance (ANOVA). A total of 67.98%, 64% and 50.6% laboratories gave accurate results for samples A, B and C respectively. The results generated by the Celltac, Humalyzer Junior, Medonic, Mindray, Colourimeter, Hemocontrol and Sysmex analysers were not significantly different (P>0.05) from the reference values. However, the Diaspect and Sahli analysers underestimated the Hb readings, while Hemocue, Urit and Mission overestimated the Hb readings (P<0.05). Interlaboratory variation of 33.3%, 25.1% and 29.4% for samples A, B and C was recorded irrespective of the analyser a laboratory used. The CV for the automated, semi-automated and manual methods was 7.08%, 7.04% and 34.26% respectively. The interlaboratory variation in Hb measurements resulted from variation in methodologies and types of analysers. Regular participation in External Quality Assessment Schemes (EQAS) is essential in order to achieve interlaboratory comparability of Hb results. Laboratories should embrace automation and gradually replace manual with automated methods, which are more accurate and reliable.