Accuracy of trained rural ophthalmologists versus non-medical image graders in the diagnosis of diabetic retinopathy in rural China

Martha McKenna, Tingting Chen, Helen McAneney, Miguel Angel Vázquez Membrillo, Ling Jin, Wei Xiao, Tunde Peto, Mingguang He, Ruth Hogg, Nathan Congdon
2018 British Journal of Ophthalmology  
Telephone: +44 (0)28 9097 8929 Fax: +44 (0)28 9097 Word count-3,890 Precis Non-medical personnel in rural China graded images obtained by nurses on patients with high cataract prevalence, meeting standards of the United Kingdom's National Diabetic Eye Screening Programme, and surpassing the performance of trained local ophthalmologists. 4 Abstract Background/aims To determine the diagnostic accuracy of trained rural ophthalmologists and non-medical image graders in the assessment of diabetic
more » ... ment of diabetic retinopathy (DR) in rural China. Methods Consecutive patients with diabetes mellitus were examined from January 2014 to December 2015 at ten county-level facilities in rural Southern China. Trained rural ophthalmologists performed a complete eye examination, recording diagnoses using the United Kingdom National Diabetic Eye Screening Programme (NDESP) classification system. Two field, mydriatic, 45° digital photographs were made by nurses using NDESP protocols, and graded by trained graders with no medical background using the NDESP system. A fellowshiptrained retina specialist graded all images in masked fashion and served as reference standard. Results Altogether, 375 participants (mean age 60 +/-10 years, 48% men) were examined and 1277 images graded. Grader sensitivity (0ꞏ82-0ꞏ94 (median 0ꞏ88)) and specificity (0ꞏ91-0ꞏ99 (median 0ꞏ98)), reached or exceeded NDESP standards (sensitivity 80%, specificity 95%) in all domains except specificity detecting any DR. Rural ophthalmologists' sensitivity was 0ꞏ65-0ꞏ95 (median 0ꞏ66) and specificity 0ꞏ59-0ꞏ95 (median 0ꞏ91). There was strong agreement between graders and the reference standard (kappa=0ꞏ84-0ꞏ87, P<0ꞏ001) and weak-moderate agreement between rural doctors and the reference (kappa= 0ꞏ48-0ꞏ64, P<0ꞏ001). 5 Conclusion This is the first study of diagnostic accuracy in DR grading among non-medical graders or ophthalmologists in LMICs. Non-medical graders can achieve high levels of accuracy, whereas accuracy of trained rural ophthalmologists is not optimal. In rural China, 81% of patients with DM do not receive an annual eye examination,[8] as recommended by Chinese guidelines adopted from the American Academy of Ophthalmology,[9] and 69% have never received an eye examination, [8] . This is largely due to uneven distribution of providers, with 80% of Chinese ophthalmologists located in urban areas, [10] . Therefore, it is necessary to explore alternate DR screening approaches in order to reduce demands on the small cadre of rural Chinese ophthalmologists, so capacity is freed for seeing those requiring treatment. To achieve this, critical evaluation of the accuracy of different models of rural DR screening is required. The aim of the present study is to compare the accuracy of two methods of DR screening: trained rural doctors performing complete eye examinations with grading of DR severity, and trained non-medical graders remotely examining retinal photographs taken by ophthalmic nurses. METHODS Study Setting This study was undertaken in Guangdong Province, Southern China. Data were collected from the Comprehensive Rural Eye Service Training (CREST) programme, an ongoing, five year collaboration
doi:10.1136/bjophthalmol-2018-312440 pmid:29973366 fatcat:o7rsjdbyajac3idp4myy4kwobu