Wanjiku Mathenge

Background Evidence on the practical application of artificial intelligence (AI)-based diabetic retinopathy (DR) screening is needed. Methods Consented participants were screened for DR using retinal imaging with AI interpretation from March 2021 to June 2021 at four diabetes clinics in Rwanda. Additionally, images were graded by a UK National Health System-certified retinal image grader. DR grades based on the International Classification of Diabetic Retinopathy with a grade of 2.0 or higher were considered referable. The AI system was designed to detect optic nerve and macular anomalies outside of DR. A vertical cup to disc ratio of 0.7 and higher and/or macular anomalies recognised at a cut-off of 60% and higher were also considered referable by AI. Results Among 827 participants (59.6% women (n=493)) screened by AI, 33.2% (n=275) were referred for follow-up. Satisfaction with AI screening was high (99.5%, n=823), and 63.7% of participants (n=527) preferred AI over human grading. Compared with human grading, the sensitivity of the AI for referable DR was 92% (95% CI 0.863%, 0.968%), with a specificity of 85% (95% CI 0.751%, 0.882%). Of the participants referred by AI: 88 (32.0%) were for DR only, 109 (39.6%) for DR and an anomaly, 65 (23.6%) for an anomaly only and 13 (4.73%) for other reasons. Adherence to referrals was highest for those referred for DR at 53.4%. Conclusion DR screening using AI led to accurate referrals from diabetes clinics in Rwanda and high rates of participant satisfaction, suggesting AI screening for DR is practical and acceptable.

The aim of this study was to investigate barriers and enablers associated with the uptake of cataract surgery in Rwanda, where financial protection is almost universally available. This was a hospital-based cross-sectional study where potential participants were adults aged >18 years who accepted an appointment for cataract surgery during the study period (May–July 2019). Information was collected from hospital records and a semi-structured questionnaire was used for data collection. Of the 297 people with surgery appointments, 221 (74.4%) were recruited into the study, 126 (57.0%) of whom had attended their appointment. People more likely to attend their surgical appointment were literate, had fewer than 8 children, had poorer visual acuity, had access to a telephone in the family, received a specific date to attend their appointment, received a reminder, and reported no difficulties walking (95% significance level, p < 0.05). The most commonly reported barriers were insufficient information about the appointment (n = 40/68, 58.8%) and prohibitive indirect costs (n = 29/68, 42.6%). This study suggests that clear communication of appointment information and a subsequent reminder, together with additional support for people with limited mobility, are strategies that could improve uptake of cataract surgery in Rwanda.

The Royal College of Ophthalmologists (RCOphth) and the College of Ophthalmology of Eastern Central and Southern Africa (COECSA) are collaborating to cascade a Training the Trainers (TTT) Programme across the COECSA Region. Within the VISION 2020 Links Programme, it aims to develop a skilled motivated workforce who can deliver high quality eye care. It will train a lead, faculty member and facilitator in 8 countries, who can cascade the programme to local trainers. In phase 1 (2013/14) two 3-day courses were run for 16/17 selected delegates, by 3 UK Faculty. In phase 2 (2015/16) 1 UK Faculty Member ran 3 shorter courses, associated with COECSA events (Congress and Examination). A COECSA Lead was appointed after the first course, and selected delegates were promoted as Facilitators then Faculty Members on successive courses. They were given appropriate materials, preparation, training and mentoring. In 4 years the programme has trained 87 delegates, including 1 COECSA Lead, 4 Faculty Members and 7 Facilitators. Delegate feedback on the course was very good and Faculty were impressed with the progress made by delegates. A questionnaire completed by delegates after 6–42 months demonstrated how successfully they were implementing new skills in teaching and supervision. The impact was assessed using the number of eye-care workers that delegates had trained, and the number of patients seen by those workers each year. The figures suggested that approaching 1 million patients per year were treated by eye-care workers who had benefited from training delivered by those who had been on the courses. Development of the Programme in Africa initially followed the UK model, but the need to address more extensive challenges overseas, stimulated new ideas for the UK courses. The Programme has developed a pyramid of trainers capable of cascading knowledge, skills and teaching in training with RCOphth support. The third phase will extend the number of facilitators and faculty, develop on-line preparatory and teaching materials, and design training processes and tools for its assessment. The final phase will see local cascade of the TTT Programme in all 8 countries, and sustainability as UK support is withdrawn.

The epidemic rise of diabetes carries major negative public health and economic consequences particularly for low and middle-income countries. The highest predicted percentage growth in diabetes is in the sub-Saharan Africa (SSA) region where to date there has been no data on the incidence of diabetic retinopathy from population-based cohort studies and minimal data on incident diabetes. The primary aims of this study were to estimate the cumulative six-year incidence of Diabetes Mellitus (DM) and DR (Diabetic Retinopathy), respectively, among people aged ≥50 years in Kenya. Random cluster sampling with probability proportionate to size were used to select a representative cross-sectional sample of adults aged ≥50 years in 2007-8 in Nakuru District, Kenya. A six-year follow-up was undertaken in 2013–14. On both occasions a comprehensive ophthalmic examination was performed including LogMAR visual acuity, digital retinal photography and independent grading of images. Data were collected on general health and risk factors. The primary outcomes were the incidence of diabetes mellitus and the incidence of diabetic retinopathy, which were calculated by dividing the number of events identified at 6-year follow-up by the number of people at risk at the beginning of follow-up. Age-adjusted risk ratios of the outcomes (DM and DR respectively) were estimated for each covariate using a Poisson regression model with robust error variance to allow for the clustered design and including inverse-probability weighting. At baseline, 4414 participants aged ≥50 years underwent complete examination. Of the 4104 non-diabetic participants, 2059 were followed-up at six-years (50 · 2%). The cumulative incidence of DM was estimated at 61 · 0 per 1000 (95% CI: 50 · 3–73 · 7) in people aged ≥50 years. The cumulative incidence of DR in the sample population was estimated at 15 · 8 per 1000 (95% CI: 9 · 5–26 · 3) among those without DM at baseline, and 224 · 7 per 1000 (116.9–388.2) among participants with known DM at baseline. A multivariable risk factor analysis demonstrated increasing age and higher body mass index to be associated with incident DM. DR incidence was strongly associated with increasing age, and with higher BMI, urban dwelling and higher socioeconomic status. Diabetes Mellitus is a growing public health concern with a major complication of diabetic retinopathy. In a population of 1 · 6 million, of whom 150,000 are ≥50 years, we estimated that 1650 people aged ≥50 develop DM per year, and 450 develop DR. Strengthening of health systems is necessary to reduce incident diabetes and its complications in this and similar settings.