Validation of formula-based glycosylated haemoglobin (hba1c) estimation and its potential utility for artificial intelligence model development: A prospective observational study.

Singh Satydeo; Prabhavathi; Samudyatha

doi:10.51168/8n2swj68

Authors

Dr. Satydeo Kumar Singh Final Year Postgraduate Student, Department of Biochemistry, Sri Devaraj Urs Medical College, Tamaka, Kolar, Karnataka, India
Dr. Prabhavathi K Professor and Head, Department of Biochemistry, Sri Devaraj Urs Medical College, Tamaka, Kolar, Karnataka, India
Dr. Samudyatha UC Associate Professor, Department of Community Medicine, Sri Devaraj Urs Medical College, Tamaka, Kolar, Karnataka, India

DOI:

https://doi.org/10.51168/8n2swj68

Keywords:

Diabetes mellitus, Glycosylated haemoglobin, HbA1c, Fasting blood glucose, Artificial intelligence

Abstract

Background

This study aimed to validate the formula hba1c (%) = 2.6 + 0.03 × fasting blood glucose (FBG) for estimating glycosylated haemoglobin (hba1c), compare calculated and laboratory-measured hba1c values, determine agreement between the two methods, and assess the suitability of formula-derived hba1c data for future Artificial Intelligence (AI) applications.

Methodology

A hospital-based prospective observational study was conducted among 1000 adults undergoing simultaneous FBG and hba1c testing at RL Jalappa Hospital, Kolar, Karnataka. FBG was measured using the glucose oxidase–peroxidase method, while hba1c was estimated by High Performance Liquid Chromatography (HPLC). Formula-derived hba1c values were compared with laboratory hba1c using Pearson correlation, paired t-test, linear regression, Bland–Altman analysis, intraclass correlation coefficient (ICC), mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE).

Results

The mean laboratory-measured hba1c was 7.66 ± 1.39%, while the mean formula-derived hba1c was 7.67 ± 1.64%. No significant difference was observed between the two methods (p = 0.284). A very strong positive correlation was found between measured and calculated hba1c values (r = 0.958, p < 0.001), with an adjusted R² of 0.918. Bland–Altman analysis demonstrated minimal bias (0.02%) and acceptable limits of agreement (−0.62% to +0.66%). The ICC was 0.962 (95% CI: 0.956–0.967). MAE, RMSE, and MAPE were 0.24%, 0.31%, and 3.18%, respectively. The validated dataset was successfully prepared and partitioned for future AI model development.

Conclusion

Formula-derived hba1c demonstrated excellent agreement with laboratory-measured hba1c and may serve as a reliable, cost-effective alternative when standardized hba1c testing is unavailable. The validated dataset may support future AI-based diabetes prediction models.

Recommendation

Future multicentre studies should validate formula-derived hba1c in diverse populations and evaluate its performance in actual AI and machine learning models for diabetes prediction and monitoring.

Author Biographies

Dr. Satydeo Kumar Singh, Final Year Postgraduate Student, Department of Biochemistry, Sri Devaraj Urs Medical College, Tamaka, Kolar, Karnataka, India

is a final year postgraduate student in the Department of Biochemistry at Sri Devaraj Urs Medical College, Tamaka, Kolar, Karnataka, India. His academic interests include clinical biochemistry, diabetes research, biomarker validation, and applications of Artificial Intelligence in laboratory medicine.
Dr. Prabhavathi K, Professor and Head, Department of Biochemistry, Sri Devaraj Urs Medical College, Tamaka, Kolar, Karnataka, India

is Professor and Head of the Department of Biochemistry at Sri Devaraj Urs Medical College. Her research interests include diabetes mellitus, metabolic disorders, clinical laboratory quality assurance, and translational medical research.
Dr. Samudyatha UC, Associate Professor, Department of Community Medicine, Sri Devaraj Urs Medical College, Tamaka, Kolar, Karnataka, India

is Associate Professor in the Department of Community Medicine, Sri Devaraj Urs Medical College. Her areas of interest include epidemiology, biostatistics, preventive medicine, public health research, and clinical study design.

References

1. 1. International Diabetes Federation. IDF Diabetes Atlas. 10th ed. Brussels: International Diabetes Federation; 2021.

2. 2. Deepa M, Elangovan N, Venkatesan U, Das HK, Jampa L, Adhikari P, et al.; ICMR-INDIAB Collaborative Study Group. Evaluation of Madras Diabetes Research Foundation-Indian Diabetes Risk Score in detecting undiagnosed diabetes in the Indian population: results from the Indian Council of Medical Research-india diabetes population-based study (INDIAB-15). Indian J Med Res. 2023;157(4):239-249. Doi:10.4103/ijmr.ijmr_2615_21.https://doi.org/10.4103/ijmr.ijmr_2615_21

3. 3. American Diabetes Association Professional Practice Committee. 2. Diagnosis and classification of diabetes: Standards of Care in Diabetes-2025. Diabetes Care. 2025;48(Suppl 1):S27-S49. Doi:10.2337/dc25-S002.https://doi.org/10.2337/dc25-S002

4. 4. World Health Organization. Global report on diabetes. Geneva: World Health Organization; 2016.

5. 5. Campbell L, Pepper T, Shipman K. Hba1c: a review of non-glycaemic variables. J Clin Pathol. 2019;72(1):12-19. Doi:10.1136/jclinpath-2017-204755.https://doi.org/10.1136/jclinpath-2017-204755

6. 6. Goldstein DE, Little RR, Lorenz RA, Malone JI, Nathan D, Peterson CM, et al. Tests of glycemia in diabetes. Diabetes Care. 2004;27(7):1761-1773. Doi:10.2337/diacare.27.7.1761.https://doi.org/10.2337/diacare.27.7.1761

7. 7. Little RR, Rohlfing CL, Hanson S, Connolly S, Higgins T, Weykamp C, et al. The National Glycohemoglobin Standardization Program: over 20 years of improving hba1c measurement. Clin Chem. 2019;65(7):839-848. Doi:10.1373/clinchem.2018.296962.

https://doi.org/10.1373/clinchem.2018.296962

8. 8. Lenters-Westra E, Slingerland RJ. Six of eight hemoglobin A1c point-of-care instruments do not meet the generally accepted analytical performance criteria. Clin Chem. 2010;56(1):44-52.https://doi.org/10.1373/clinchem.2009.130641

9. 9. Sherwani SI, Khan HA, Ekhzaimy A, Masood A, Sakharkar MK. Significance of hba1c test in diagnosis and prognosis of diabetic patients. Biomark Insights. 2016;11:95-104.https://doi.org/10.4137/BMI.S38440

10. 10. Bonora E, Tuomilehto J. The pros and cons of diagnosing diabetes with A1C. Diabetes Care. 2011;34(Suppl 2):S184-S190.

https://doi.org/10.2337/dc11-s216

11. 11. Higgins T. Hba1c for screening and diagnosis of diabetes mellitus. Endocrine. 2013;43(2):266-273.

https://doi.org/10.1007/s12020-012-9768-y

12. 12. Cohen RM, Lindsell CJ. When the blood glucose and hba1c do not match: turning uncertainty into opportunity. Diabetes Care. 2012;35(12):2421-2423.https://doi.org/10.2337/dc12-1479

13. 13. Jiang F, Jiang Y, Zhi H, Dong Y, Li H, Ma S, et al. Artificial intelligence in healthcare: past, present and future. Stroke Vasc Neurol. 2017;2(4):230-243.https://doi.org/10.1136/svn-2017-000101

14. 14. Ellahham S, Ellahham N, Simsekler MCE. Application of artificial intelligence in diabetes mellitus: current status and future prospects. World J Diabetes. 2020;11(4):163-176.

15. 15. Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care. 2004;27(5):1047-1053. Doi:10.2337/diacare.27.5.1047.https://doi.org/10.2337/diacare.27.5.1047

16. 16. Sun H, Saeedi P, Karuranga S, Pinkepank M, Ogurtsova K, Duncan BB, et al. IDF Diabetes Atlas: global, regional and country-level diabetes prevalence estimates for 2021 and projections for 2045. Diabetes Res Clin Pract. 2022;183:109119.

https://doi.org/10.1016/j.diabres.2021.109119

17. 17. Pradeepa R, Mohan V. Epidemiology of type 2 diabetes in India. Indian J Ophthalmol. 2021;69(11):2932-2938. Doi:10.4103/ijo.IJO_1627_21.https://doi.org/10.4103/ijo.IJO_1627_21

18. 18. Monnier L, Lapinski H, Colette C. Contributions of fasting and postprandial plasma glucose increments to the overall diurnal hyperglycemia of type 2 diabetic patients: variations with increasing levels of hba1c. Diabetes Care. 2003;26(3):881-885. Doi:10.2337/diacare.26.3.881.https://doi.org/10.2337/diacare.26.3.881

19. 19. Misquith A, Rangareddy H, Chikkanarayanappa V, Sultana A, Ashakiran S. Mathematical model for assessing glycemic control in type 2 diabetes mellitus. Bioinformation. 2024;20(2):116-120. Doi:10.6026/973206300200116.

https://doi.org/10.6026/973206300200116

20. 20. Rajan SS, Misquith A, Rangareddy H. Calculated glycosylated hemoglobin (hba1c) compared with estimated hba1c by nephelometry and its correlation to estimated average blood glucose (eag). Gal Int J Health Sci Res. 2020;5(4):65-71.

21. 21. Musa IR, Omar SM, Sharif ME, Ahmed ABA, Adam I. The calculated versus the measured glycosylated haemoglobin (hba1c) levels in patients with type 2 diabetes mellitus. J Clin Lab Anal. 2021;35(8):e23873. Doi:10.1002/jcla.23873.

https://doi.org/10.1002/jcla.23873

22. 22. Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ; a1c-Derived Average Glucose (ADAG) Study Group. Translating the A1C assay into estimated average glucose values. Diabetes Care. 2008;31(8):1473-1478.

https://doi.org/10.2337/dc08-0545

23. 23. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;1(8476):307-310.https://doi.org/10.1016/S0140-6736(86)90837-8

24. 24. Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med. 2016;15(2):155-163.https://doi.org/10.1016/j.jcm.2016.02.012

25. 25. Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019;25(1):44-56. Doi:10.1038/s41591-018-0300-7.https://doi.org/10.1038/s41591-018-0300-7

26. 26. Zhu T, Li K, Herrero P, Georgiou P. Deep learning for diabetes: a systematic review. IEEE J Biomed Health Inform. 2021;25(7):2744-2757. Doi:10.1109/JBHI.2020.3040225.https://doi.org/10.1109/JBHI.2020.3040225

27. 27. English E, Idris I, Smith G, Dhatariya K, Kilpatrick ES, John WG. The effect of anaemia and abnormalities of erythrocyte indices on hba1c analysis: a systematic review. Diabetologia. 2015;58(7):1409-1421. Doi:10.1007/s00125-015-3599-3.

https://doi.org/10.1007/s00125-015-3599-3