Establishing Hemoglobin Variant Confidence Intervals to Improve Sickle Cell Anemia and Related Hemoglobinopathies Classification in Western Kenya

Abstract

Sickle cell disease (SCD) is one of the most prevalent inherited hemoglobinopathies in sub-Saharan Africa and remains a major cause of morbidity and mortality. In many resource-limited settings, diagnosis continues to rely on the sickling test, which cannot reliably distinguish SCD from sickle cell trait (SCT). Although hemoglobin phenotyping offers superior diagnostic accuracy, confidence intervals (CIs) for hemoglobin variants have not been established in Western Kenya, leading to challenges in accurate classification and patient management. This study aimed to establish 95% confidence intervals for hemoglobin variants to improve the classification of SCD, SCT, and related hemoglobinopathies in Western Kenya. A retrospective descriptive study was conducted using hematology records of 385 individuals tested between January 2015 and November 2021. Hemoglobin variant distributions (HbA1, HbS, HbA2, and HbF) were analyzed using Chi-square tests for categorical associations and One-way ANOVA to assess significant differences across diagnostic groups. The findings showed that SCD patients had markedly reduced HbA1 (<25%) and elevated HbS (>52%), while SCT cases presented with HbA1 >33% and HbS <33%. Pure beta-thalassemia (β-Thal) was characterized by HbA1 >70%, and heterozygous combinations such as HbSS/β-Thal, HbAS/β-Thal, and β-Thal minor consistently showed elevated HbA2 fractions. However, HbA2 and HbF were of limited value in distinguishing SCD from SCT. Establishing phenotype-specific confidence intervals for HbA1 and HbS enhances diagnostic accuracy, minimizes misclassification, and strengthens hemoglobinopathy diagnosis in resource-limited settings such as Western Kenya.