Objective: To determine the incidence rates and risk factors for development of diabetic retinopathy in Oklahoma Indians.
Research design and methods: Cohort follow-up study with baseline examination between 1972 and 1980 and follow-up examination between 1987 and 1991. Mean +/- SD follow-up time was 12.8 +/- 1.7 yr. Eleven Indian Health Service facilities (clinics and hospitals) in Oklahoma participated in the study. Study participants were a quasirandom sample of 1012 American Indians (379 men, 633 women) in Oklahoma with NIDDM, 927 of whom received a detailed eye examination at baseline. The mean age of participants was 52 yr with a duration of diabetes of 6.9 yr at baseline. The average quantum of Indian blood was 92% (77% full blood). At follow-up, 515 (55.6%) were alive, 408 (44.0%) were deceased, and 4 (0.4%) could not be traced. Of the living participants, 380 (73.8%) underwent an ophthalmoscopic examination.
Results: The incidence of retinopathy among the participants who were free of disease at baseline and who survived the follow-up interval was 72.3%. By multivariate analysis, significant independent predictors of retinopathy recorded at baseline were FPG level, therapeutic regimen, systolic blood pressure, and duration of diabetes. FPG levels > or = 11.1 mM (200 mg/dl) increased the risk of retinopathy 1.7 times that for levels < 7.8 mM (140 mg/dl). Insulin use was associated with a 20% greater incidence. Hypertension was a particularly significant risk factor for those with lower FPG levels.
Conclusions: Given that NIDDM is reaching epidemic proportions in Oklahoma Indians and that most may be afflicted with retinopathy, frequent ophthalmological examinations are clearly indicated for this high-risk population. The role of intervention, namely glycemic and hypertensive control, deserves further study.