BACKGROUND: To determine the prevalence and severity of diabetic eye disease among adults aged ≥40 years with unrecognized diabetes in Fiji. DESIGN: Population-based cross-sectional survey using multistage cluster random sampling. PARTICIPANTS: 1381 (=73.0% participation). METHODS: Interview-based questionnaire; visual acuity measured; dilated ocular examination performed; glycosylated haemoglobin (HbA1c) concentration determined. MAIN OUTCOME MEASURES: Prevalence and grade of diabetic retinopathy/maculopathy. RESULTS: Sample prevalence of diabetes was 44.8% (95%CI 42.2-47.5%), with 63.4% (95%CI 59.5-67.1%) previously undiagnosed (384/606). Predictors of undiagnosed compared with previously diagnosed diabetes were female gender (P = 0.001), rural residence (P = 0.049) and not having a relative with known diabetes (P < 0.001). Twenty-two retinae of participants with previously undiagnosed diabetes were unexaminable (predominantly cataract). Of the remaining 746 eyes, 3.5% (95%CI 2.4-5.1%) had diabetic retinopathy/maculopathy, 1 (0.1%) had proliferative retinopathy and 4 (0.5%) had active significant maculopathy. Of eyes with diabetic disease, two (7.7%, 95%CI 1.0-25.3%) had diabetes-related vision impairment (3/60; 6/60). Sixteen previously undiagnosed participants (4.2%, 95%CI 2.5-6.7%) had diabetic disease evident in at least one eye: for four (all Melanesian women aged >50 years), this was vision-threatening (1.0%; 95%CI 0.3-2.8). Mean HbA1c (10.7 ± 2.6%) of participants undiagnosed and with diabetes eye disease was higher (P < 0.001) than that of those undiagnosed and without. CONCLUSIONS: The prevalence of diabetic eye disease was low among this cohort, but where present, severe vision-threatening retinopathy/maculopathy was relatively common. If diabetic eye disease is to be avoided or ameliorated in Fiji, then community awareness of and access to diabetes diagnostic services must improve, particularly for women and rural dwellers.
BACKGROUND: To determine the prevalence and severity of diabetic eye disease among adults aged ≥40 years with unrecognized diabetes in Fiji. DESIGN: Population-based cross-sectional survey using multistage cluster random sampling. PARTICIPANTS: 1381 (=73.0% participation). METHODS: Interview-based questionnaire; visual acuity measured; dilated ocular examination performed; glycosylated haemoglobin (HbA1c) concentration determined. MAIN OUTCOME MEASURES: Prevalence and grade of diabetic retinopathy/maculopathy. RESULTS: Sample prevalence of diabetes was 44.8% (95%CI 42.2-47.5%), with 63.4% (95%CI 59.5-67.1%) previously undiagnosed (384/606). Predictors of undiagnosed compared with previously diagnosed diabetes were female gender (P = 0.001), rural residence (P = 0.049) and not having a relative with known diabetes (P < 0.001). Twenty-two retinae of participants with previously undiagnosed diabetes were unexaminable (predominantly cataract). Of the remaining 746 eyes, 3.5% (95%CI 2.4-5.1%) had diabetic retinopathy/maculopathy, 1 (0.1%) had proliferative retinopathy and 4 (0.5%) had active significant maculopathy. Of eyes with diabetic disease, two (7.7%, 95%CI 1.0-25.3%) had diabetes-related vision impairment (3/60; 6/60). Sixteen previously undiagnosed participants (4.2%, 95%CI 2.5-6.7%) had diabetic disease evident in at least one eye: for four (all Melanesian women aged >50 years), this was vision-threatening (1.0%; 95%CI 0.3-2.8). Mean HbA1c (10.7 ± 2.6%) of participants undiagnosed and with diabetes eye disease was higher (P < 0.001) than that of those undiagnosed and without. CONCLUSIONS: The prevalence of diabetic eye disease was low among this cohort, but where present, severe vision-threatening retinopathy/maculopathy was relatively common. If diabetic eye disease is to be avoided or ameliorated in Fiji, then community awareness of and access to diabetes diagnostic services must improve, particularly for women and rural dwellers.