Estimates of DR incidence and progression were lower than previously reported for American Indian and Alaska Native individuals served by the Indian Health Service, suggesting the potential for longer intervals between evaluations.
Alongside the expansion of diabetes medications, the incidence and progression of diabetic retinopathy (DR) among American Indian and Alaska Native individuals served by the Indian Health Service (IHS) appear to have substantially declined, according to the results of a new cohort study.1
In fact, the data suggested rates are substantially lower than those reported 30 or more years ago and are now comparable with estimates from non-American Indian and Alaska Native populations examined within the past 2 decades.
The investigative team, led by Stephanie J. Fonda, PhD, Estenda Solutions, Inc, suggests the low rates support the “viability of safely extending the follow-up interval for retinopathy assessment in IHS patients who have no evidence of DR or mild non-proliferative DR (NPDR).”
Previous data have indicated a decline in the prevalence of diabetic eye disease among American Indian and Alaska Native patients in the IHS primary care–based teleophthalmology program., However, the estimates are based on data from before 1992, leading the study to examine current incidence and progression rates among this population.
Between January 2015 and December 2019, the retrospective cohort study used deidentified medical record data obtained during routine clinical operations of the IHS program at 75 primary care clinics across 20 states. The study included patients with diabetes and no evidence of DR or had mild NPDR during the baseline period.
Patients were evaluated using nonmydriatic ultra-widefield imaging (UWFI) or nonmydriatic fundus photography (NMFP). The study measured the incidence and progression of DR as any increase in the level of DR, the occurrence of a 2 or more-step increase, and overall change in DR severity level.
The total number of patients evaluated by the program in the baseline year was 13,694. Of these patients, 8374 (mean age, 53.2 years; 4775 female patients [57.0%]) had at least 1 examination during the follow-up period. At baseline, in 2015, the mean hemoglobin A1c level (HbA1c) was 8.3% and the mean duration of diabetes was 8.6 years.
Upon analysis, of patients with no evidence of DR at baseline, 18.0% (1280 of 7097) had some level of DR at follow-up, with an incidence rate of 69.6 cases per 1000 person-years. Of the new DR found, 839 of 1280 cases (65.5%) were mild NPDR. Data showed the cumulative incidence of PDR was 0.1% (10 of 7097) for an incidence rate of 0.5 cases per 1000 person-years.
Meanwhile, of patients with no evidence of DR at baseline, 441 of 7097 (6.2%) had a 2 or more-step increase in DR over time (24.0 cases per 1000 PY). Additionally, a total of 347 of 1277 patients (27.2%) with mild NPDR at baseline developed a more severe DR level in the follow-up period, for an incidence rate of 111.7 cases per 1000 PY. Data showed a 2 or more-step increase in DR occurred for 2.3% of these patients (30 of 1277).
The characteristics associated with DR incidence as well as the occurrence of a 2 or more-step increase were expected, including longer diabetes duration, higher HbA1c level, and diabetes therapy, including insulin use alone or with oral medications. Notable characteristics associated with any progression from mild NPDR were additionally longer diabetes duration and higher HbA1c level, as well as the presence of peripheral neuropathy.
Investigators noted the current IHS teleophthalmology program recommends annual DR examinations, consistent with the recommendations of professional organizations, including the American Academy of Ophthalmology. However, they noted a biennial frequency might be an appropriate frequency, as recommended by the American Diabetes Association.
“Such a practice change, however, requires examination of adherence to the current recommendations,” investigators wrote. “If a practice change extending follow-up were implemented, further research would be needed to determine if the change affected vision outcomes and adherence rates.”