June 2017
Volume 58, Issue 8
Open Access
ARVO Annual Meeting Abstract  |   June 2017
Accuracy of trained rural ophthalmologists versus non-medical image graders in the diagnosis of diabetic retinopathy in rural China
Author Affiliations & Notes
  • Nathan G Congdon
    Centre for Public Health, Queen's University Belfast, Belfast, Antrim, United Kingdom
    Preventive Ophthalmology, Zhongshan Ophthalmic Center, Guangzhou, China
  • Martha McKenna
    Centre for Public Health, Queen's University Belfast, Belfast, Antrim, United Kingdom
  • Tingting Chen
    Preventive Ophthalmology, Zhongshan Ophthalmic Center, Guangzhou, China
  • Helen McAneney
    Centre for Public Health, Queen's University Belfast, Belfast, Antrim, United Kingdom
  • Miguel Vazquez
    Preventive Ophthalmology, Zhongshan Ophthalmic Center, Guangzhou, China
  • Ling Jin
    Preventive Ophthalmology, Zhongshan Ophthalmic Center, Guangzhou, China
  • Wei Xiao
    Preventive Ophthalmology, Zhongshan Ophthalmic Center, Guangzhou, China
  • Footnotes
    Commercial Relationships   Nathan Congdon, None; Martha McKenna, None; Tingting Chen, None; Helen McAneney, None; Miguel Vazquez, None; Ling Jin, None; Wei Xiao, None
  • Footnotes
    Support  World Diabetes Foundation; Orbis International; Prof Congdon is supported by the Chinese government’s Thousand Man Plan and by the Ulverscroft Foundation (UK)
Investigative Ophthalmology & Visual Science June 2017, Vol.58, 2903. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Nathan G Congdon, Martha McKenna, Tingting Chen, Helen McAneney, Miguel Vazquez, Ling Jin, Wei Xiao; Accuracy of trained rural ophthalmologists versus non-medical image graders in the diagnosis of diabetic retinopathy in rural China. Invest. Ophthalmol. Vis. Sci. 2017;58(8):2903.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : To determine the diagnostic accuracy of trained rural ophthalmologists and non-medical image graders in the assessment of diabetic retinopathy (DR) in rural China.

Methods : Consecutive patients with diabetes were examined from January 2014 to December 2015 in an on-going program at ten county-level facilities in rural Southern China. Trained rural ophthalmologists performed a complete eye examination, including dilation of the pupil and evaluation of the fundus by slitlamp biomicroscopy, recording diagnoses using the United Kingdom National Health Service (NHS) system. Two field, cycloplegic, 45° digital photography was performed by ophthalmic nurses according to the NHS protocol, and graded using the NHS system on web-based software by trained graders with no medical background. A fellowship-trained retina specialist graded all images in masked fashion and served as reference standard. Accuracy was assessed for the diagnosis of any DR, DR requiring treatment and diabetic macular edema requiring referral. Logistic models were used to explore predictors of correct diagnosis.

Results : In total 375 participants (mean age 60 +/- 10 years, 48% men) were examined and 1277 images graded. Grader sensitivity and specificity ranged from 0.82-0.94 (median 0.88) and 0.91-0.99 (median 0.98) respectively, reaching or exceeding NHS standards for sensitivity (80%) and specificity (95%) in all domains except specificity in detecting any DR. Sensitivity and specificity for rural ophthalmologists ranged from 0.65-0.95 (median 0.66) and 0.59-0.95 (median 0.92) respectively. There was a strong agreement between graders and the reference standard in all domains (kappa=0.84-0.87, P<0.001) and weak-moderate agreement between rural doctors and the reference (kappa= 0.48-0.64, P<0.001).

Conclusions : This is the first study of diagnostic accuracy in DR grading among non-medical graders or ophthalmologists in low-resource settings. Non-medical graders can achieve high levels of accuracy, whereas accuracy of trained rural ophthalmologists is not optimal. Cost-effectiveness analyses of these approaches are needed.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×