Muluken Walle ^1* Ermiyas Alemayehu ² Adisu Tesfaye ¹ Mesay Arkew ³ Haftu Asmerom ³ Melaku M. Agidew ⁴ Fasil Getu ¹

• ¹ Jigjiga University, Jijiga, Ethiopia
• ² Wollo University, Dessie, Amhara, Ethiopia
• ³ Haramaya University, Harar, Dire Dawa, Ethiopia
• ⁴ College of Health Science, Debre Tabor University, Debre Tabor, Amhara, Ethiopia

Erythrocyte sedimentation rate (ESR) is a widely used screening test in clinical practice as an indicator of inflammatory and degenerative malignant diseases. The Westergren method, renowned as the gold standard, is valued for its accuracy and cost-effectiveness but demands considerable time and blood volume. Emerging automated methods offer quicker and more convenient alternatives, aiming to replace manual techniques. Nonetheless, validating these automated methods against the reference Westergren method is essential to ensure reliability.Therefore, this study aimed to evaluate ESR measurement results obtained from both the reference Westergren method and the automated (SFRI ESR 3000) method.A Hospital-based comparative cross-sectional study was conducted at Jigjiga University Sheik Hassen Yabare Referral Hospital from July 15 to September 16, 2023. Following the acquisition of informed consent, blood samples were obtained from 158 participants, five milliliters of blood from each participant. These samples were then subjected to ESR estimation using both the Westergren (reference) method and the automated (SFRI ESR 3000) method. ethodes Subsequently, the collected data were analyzed using SPSS version 20 and MedCalc version 12.3.0.0 statistical Softwares. Statistical analyses such as Paired t-tests, Pearson correlation, linear regression, and the Bland and Altman plot were employed. The paired sample t-test analysis revealed no significant difference between the use of the reference Westergren method and the automated method for ESR determination, with a mean difference (MD) of 0.7 ± 9.2 mm/hr (P = 0.36). A significant correlation was observed between the two methods, with a remarkable correlation coefficient (r = 0.94, p < 0.001). The Bland-Altman data analysis indicated no evidence of systematic bias and demonstrated good agreement of ESR values between the two methods, with a limit of agreement of -17.3 to +18.7.Moreover, within-run imprecision analysis for the automated method across a range of ESR values showed coefficient of variation of 27.08%, 12.65%, and 10.32% for low, medium, and high ESR levels, respectively. The SFRI ESR 300 automated method demonstrates the potential for interchangeable use with the Westergren method for determining ESR, given the strong correlation and good agreement. The same reference range could be applied during interpretation.