Articles Vol. 67 No. 8 28/08/2026

DIAGNOSTIC ACCURACY AND REFERRAL PERFORMANCE OF TELEMEDICINE FOR DIABETIC RETINOPATHY SCREENING AT PRIMARY HEALTH CARE FACILITIES IN NGHE AN

Du Hai Nam1,2, Tran Tat Thang1,2, Duong Dinh Chinh3,4, Mai Van Anh Khoa5,6, Tran Nhu Khoa7,8
1 Nghe An Eye Hospital
2 Bệnh viện Mắt Nghệ An
3 Nghe An Provincial Department of Health
4 Sở Y tế tỉnh Nghệ An
5 Hai Chau Regional Health Center
6 Trung tâm Y tế khu vực Hải Châu
7 Institute of Aerospace Medicine
8 Viện Y học Phòng không Không quân
DOI: 10.52163/yhc.v67i8.6254
15 Views
0 Downloads
Keywords
diabetic retinopathy; telemedicine; referral; screening; diagnostic accuracy; primary health care.
Abstract

Background: Diabetic retinopathy (DR) screening at the primary health care level remains challenging because of limited access to ophthalmic specialists. Although telemedicine has demonstrated high diagnostic accuracy, evidence regarding its role in supporting referral decisions within community-based screening programmes is still limited.

Objective: To evaluate the diagnostic accuracy and referral-support performance of a telemedicine-based diabetic retinopathy screening programme at primary health care facilities in Nghe An Province, Vietnam.

Methods: A community-based cross-sectional diagnostic accuracy study was conducted among 360 adults with type 2 diabetes. Digital fundus photographs were obtained at primary health care facilities and graded remotely using a telemedicine platform. Dilated fundus examination by retinal specialists using slit-lamp biomicroscopy with a Volk 90D lens served as the reference standard. Image gradability, sensitivity, specificity, predictive values, Cohen's kappa, and agreement in referral decisions were assessed.

Results: Of 360 participants, 330 image sets (91.7%) were gradable. For diabetic retinopathy detection, telemedicine achieved a sensitivity of 93.9%, specificity of 90.0%, positive predictive value of 80.2%, negative predictive value of 97.2%, and a Cohen's kappa of 0.78. High agreement was observed between telemedicine and direct examination regarding referral decisions. Disagreements were mainly related to ungradable images and early-stage retinal lesions.

Conclusions: Telemedicine demonstrated high diagnostic accuracy and substantial agreement in referral decision-making. These findings support the integration of telemedicine as a referral-support system, rather than merely an image-reading tool, within diabetic retinopathy screening programmes at primary health care facilities.

References
[1]
Teo Z.L, Tham Y.C, Yu M. và cộng sự. (2021). Global Prevalence of Diabetic Retinopathy and Projection of Burden through 2045: Systematic Review and Meta-analysis. Ophthalmology, 128(11), 1580–1591. DOI 10.1016/j.ophtha.2021.04.027 Google Scholar
[2]
Nguyễn Trọng Khải (2018), Nghiên cứu đặc điểm dịch tễ, lâm sàng bênh võng mạc đái tháo đường và hiệu quả can thiệp tại tỉnh Hà Nam, Luận án tiến sỹ y học, Đại học y Hà Nội. Google Scholar
[3]
Nguyễn Thị Thủy Tiên và Trần Đình Tùng (2024). Khảo sát bệnh lý võng mạc đái tháo đường trên bệnh nhân điều trị tại bệnh viện Thống Nhất. Tạp chí Y học cộng đồng Việt Nam, 65(10), 410–418. Google Scholar
[4]
American Diabetes Association Professional Practice Committee (2025). 12. Retinopathy, Neuropathy, and Foot Care: Standards of Care in Diabetes-2025. Diabetes Care, 48(1 Suppl 1), S252–S265. DOI 10.2337/dc25-S012 Google Scholar
[5]
Wong. T. Y, Jenifer S, Ryo K và cộng sự. (2018). Guidelines on Diabetic Eye Care: The International Council of Ophthalmology Recommendations for Screening, Follow-up, Referral, and Treatment Based on Resource Settings. Ophthalmology, 125(10). DOI 10.1016/j.ophtha.2018.04.007 Google Scholar
[6]
Lim J.I, Kim S.J, Bailey S.T và cộng sự. (2025). Diabetic Retinopathy Preferred Practice Pattern®. Ophthalmology, 132(4), P75–P162. DOI 10.1016/j.ophtha.2024.11.019 Google Scholar
[7]
Li H.K, Horton M, Bursell S.E và cộng sự. (2011). Telehealth practice recommendations for diabetic retinopathy, second edition. Telemed J E Health, 17(10), 814–837. DOI: 10.1089/tmj.2011.0075. Google Scholar
[8]
Shi L, Wu H, Dong J và cộng sự. (2015). Telemedicine for detecting diabetic retinopathy: a systematic review and meta-analysis. Br J Ophthalmol, 99(6), 823–831. doi: 10.1136/bjophthalmol-2014-305631 Google Scholar
[9]
Bossuyt P.M, Reitsma J.B, Bruns D.E và cộng sự. (2015). STARD 2015: An Updated List of Essential Items for Reporting Diagnostic Accuracy Studies. Clin Chem, 61(12), 1446–1452. Google Scholar
[10]
Buderer N.M. (1996). Statistical methodology: I. Incorporating the prevalence of disease into the sample size calculation for sensitivity and specificity. Acad Emerg Med, 3(9), 895–900. DOI: 10.1111/j.1553-2712.1996.tb03538.x Google Scholar
[11]
Yau J.W.Y, Rogers S.L, Kawasaki R. và cộng sự. (2012). Global Prevalence and Major Risk Factors of Diabetic Retinopathy. Diabetes Care, 35(3), 556–564. DOI 10.2337/dc11-1909 Google Scholar