Establishing laboratory reference ranges for adults and children in Kilifi, Kenya

Accurate laboratory reference ranges (RR) are essential for diagnosis and management of patients in routine clinical care and clinical trials. RRs vary between geographical location due to differences in population demographics and blood analysis equipment, so locally derived RRs are essential. Here we establish adult and paediatric RRs for a rural population in Kilifi, Kenya using clinical trial data from KEMRI-Wellcome Trust Research Programme (KWTRP). Data from healthy, non-pregnant participants from six clinical trials conducted between 2016 and 2020 were used. Coulter ACT 5 Diff and Ilab Aries were used for haematological and biochemical analysis respectively. Quality control was undertaken daily prior to sample analysis. Derived RRs were compared with RRs from other African countries and further afield. All analyses were performed using R version 3.6.1 (Reference Intervals package). 2338 adults and 2054 children were included, 52% of adults and 51% of children were male, median adult age was 32.5 years. Haemoglobin range was lower in women compared to men (9.5-14.2g/dL and 11.5-16.6g/dL respectively), platelet upper limit of normal (ULN) was higher in women compared to men (397 x 103/L vs 358 x 103/ L). Biochemistry values were higher in men (ALT ULN 57 U/L in men and 35 U/L in women, creatinine ULN 113umol/L in men and 91umol/L in women). Paediatric RRs showed differences in multiple parameters depending on the age of the child. In both adults and children, many parameters in 2023 Kilifi RRs differed from those in other countries. There was however little difference between 2023 and 2017 Kilifi paediatric RRs. This study provides RRs for adults and children in Kilifi, and the most extensive RRs available for much of East and Southern Africa. We show the need for locally derived reference ranges, highlighting differences between sex, age and geographical location.