Trimester-specific reference interval for glycated hemoglobin in non-diabetic pregnant women: A cross-sectional study.
Gestational Diabetes Mellitus (GDM) is a growing public health concern in Ethiopia. HbA1c is useful for assessing glycemia but is influenced by physiological changes in pregnancy. To our knowledge, no published trimester-specific HbA1c reference intervals exist for Ethiopian non-diabetic pregnant women. This limits population-specific clinical interpretation.
A facility-based cross-sectional study was conducted from February to July 2025 among 424 non-diabetic pregnant women across all trimesters. Normality of HbA1c distribution was assessed using the Shapiro-Wilk and Kolmogorov-Smirnov tests; outliers were identified using Tukey's fences method. Reference intervals were determined parametrically (mean ±2SD). Differences across trimesters were tested using one-way ANOVA. Because Levene's test indicated a violation of the homogeneity of variances (p < 0.001) and group sizes were unequal, pairwise comparisons were performed using the Games-Howell post hoc test. A p-value < 0.05 was considered statistically significant.
HbA1c showed a normal distribution across all trimesters. Reference intervals were 4.8-5.9% in the first, 4.2-5.8% in the second, and 4.5-5.9% in the third trimester. Mean HbA1c levels differed significantly between trimesters (p < 0.05). Lowest in the second trimester 5.00% (95% CI 4.93-5.07) and highest in the first trimester 5.35% (95% CI: 5.31-5.39). Games-Howell post hoc tests confirmed that all pairwise comparisons were significant (p < 0.05).
Trimester-specific HbA1c reference intervals were defined for non-diabetic pregnant women, demonstrating measurable physiological variation across gestation. The use of locally derived thresholds could improve the accuracy of GDM screening and support more appropriate clinical interpretation.
A facility-based cross-sectional study was conducted from February to July 2025 among 424 non-diabetic pregnant women across all trimesters. Normality of HbA1c distribution was assessed using the Shapiro-Wilk and Kolmogorov-Smirnov tests; outliers were identified using Tukey's fences method. Reference intervals were determined parametrically (mean ±2SD). Differences across trimesters were tested using one-way ANOVA. Because Levene's test indicated a violation of the homogeneity of variances (p < 0.001) and group sizes were unequal, pairwise comparisons were performed using the Games-Howell post hoc test. A p-value < 0.05 was considered statistically significant.
HbA1c showed a normal distribution across all trimesters. Reference intervals were 4.8-5.9% in the first, 4.2-5.8% in the second, and 4.5-5.9% in the third trimester. Mean HbA1c levels differed significantly between trimesters (p < 0.05). Lowest in the second trimester 5.00% (95% CI 4.93-5.07) and highest in the first trimester 5.35% (95% CI: 5.31-5.39). Games-Howell post hoc tests confirmed that all pairwise comparisons were significant (p < 0.05).
Trimester-specific HbA1c reference intervals were defined for non-diabetic pregnant women, demonstrating measurable physiological variation across gestation. The use of locally derived thresholds could improve the accuracy of GDM screening and support more appropriate clinical interpretation.
Authors
Takiso Takiso, Tsegaye Tsegaye, Dabulo Dabulo, Woldesenbet Woldesenbet, Leminie Leminie
View on Pubmed