Introduction
Diabetic retinopathy (DR) represents the most common microvascular complication of diabetes mellitus and remains a leading cause of blindness among working-age adults worldwide. While the ocular manifestations of DR have been extensively studied, emerging evidence suggests that the presence and severity of retinopathy may serve as a window into broader systemic health trajectories. The concept of “oculomics”—the study of ocular biomarkers that reflect systemic disease—has gained increasing recognition as researchers seek to leverage eye examinations for early detection of systemic comorbidities.
A landmark retrospective cohort study published in Ophthalmology by Lee and colleagues, utilizing data from the Sight Outcomes Research Collaborative (SOURCE) consortium, has provided compelling evidence that baseline DR severity independently predicts mortality risk in patients with diabetes mellitus. This research, encompassing over half a million patients across multiple health systems, marks a significant advancement in our understanding of the systemic implications of ocular disease.
Study Design and Population
The SOURCE consortium retrospective cohort study analyzed clinical data from electronic health records of 524,687 patients with diabetes mellitus who received care at participating health systems between 2010 and 2023. Patient data were linked to mortality information obtained from the National Death Index, enabling robust outcome ascertainment.
At their initial ophthalmic evaluation, patients were stratified into six distinct categories based on the level of DR in the more affected eye: diabetes mellitus without DR; non-proliferative DR (NPDR) without diabetic macular edema (DME); NPDR with DME; proliferative DR (PDR) without DME; PDR with DME; and unspecified DR severity. This classification allowed for granular analysis of how disease severity and the presence of macular edema influence outcomes.
Cox regression modeling served as the primary analytical approach, with adjustment for multiple potential confounders including sociodemographic factors (age, sex, race/ethnicity, community affluence) and clinical parameters (glycosylated hemoglobin levels, among others). The study design accounted for the competing risk of death and employed robust statistical methods to minimize immortal time bias.
Demographic Characteristics
The study cohort demonstrated considerable demographic diversity, enhancing the generalizability of findings. The mean patient age was 59.1 ± 16 years. Sex distribution showed a slight female predominance, with 283,465 patients (54.0%) identifying as female. Regarding racial and ethnic composition, 95,160 patients (18.1%) were Black, 66,547 (12.7%) were Latinx, and 306,104 (58.3%) were White. This diverse population base allows for meaningful assessment of how DR-mortality associations may vary across demographic groups.
Key Findings: Mortality Risk Stratification
Over a mean follow-up period of 5.4 ± 4.1 years, 36,674 patients (7.0%) died. After comprehensive adjustment for potential confounders, the study demonstrated a clear dose-response relationship between DR severity and mortality hazard:
Non-Proliferative Diabetic Retinopathy:
Patients with NPDR without DME exhibited a 34% increased hazard of death compared to diabetics without retinopathy (HR: 1.34; 95% CI: 1.29-1.39). Interestingly, those with NPDR accompanied by DME showed a comparable risk elevation of 31% (HR: 1.31; 95% CI: 1.20-1.43). The similar magnitude of risk between these two subgroups suggests that the presence of NPDR itself, rather than concurrent macular edema, drives the mortality association.
Proliferative Diabetic Retinopathy:
The most striking findings emerged among patients with PDR. Those with PDR without DME demonstrated a remarkable 128% increased hazard of death (HR: 2.28; 95% CI: 2.17-2.39) relative to diabetics without DR. Even those with PDR and concomitant DME showed substantially elevated risk, with an 87% increase in mortality hazard (HR: 1.87; 95% CI: 1.72-2.05). These findings indicate that the transition to proliferative disease represents a critical threshold beyond which systemic risk escalates dramatically.
Additional Risk Factors
Beyond DR severity, several other factors independently predicted increased mortality risk. Advanced age demonstrated the strongest association with mortality, consistent with established epidemiological patterns. Male sex was associated with higher mortality hazard compared to female sex. Perhaps most notably, residence in a less affluent community emerged as an independent predictor of mortality (p < 0.01 for all associations), highlighting the profound influence of social determinants of health on diabetic outcomes.
Mechanistic Insights and Biological Plausibility
The association between DR severity and mortality likely reflects shared pathophysiological mechanisms underlying both microvascular disease and systemic health decline. Diabetic retinopathy development correlates with similar microvascular damage in the kidneys, heart, and peripheral vasculature. Endothelial dysfunction, chronic inflammation, and advanced glycation end-product accumulation—processes central to DR pathogenesis—likely contribute simultaneously to accelerated atherosclerosis, cardiovascular events, and overall mortality.
The dramatic risk escalation observed with proliferative disease suggests that the neovascularization characterizing PDR may reflect a systemic pro-angiogenic state associated with poorer prognosis. Alternatively, the management challenges inherent in advanced DR may correlate with broader difficulties in achieving glycemic control and managing systemic complications.
Clinical Implications
These findings carry significant implications for clinical practice. Ophthalmologists and optometrists performing diabetic eye examinations possess a unique opportunity to identify patients at elevated systemic risk. The study authors appropriately emphasize that eyecare professionals should be incentivized and empowered to communicate DR severity status to primary care providers and other medical specialists involved in diabetic management.
From a risk stratification standpoint, the presence of any DR—particularly NPDR—should prompt intensified surveillance for cardiovascular disease, renal dysfunction, and other diabetes-related complications. Patients with PDR warrant especially aggressive management of modifiable risk factors including glycemic control, blood pressure, and lipid profiles.
Limitations and Considerations
Several limitations warrant acknowledgment. The retrospective design introduces potential for residual confounding despite extensive statistical adjustment. Information on diabetes duration, specific antidiabetic medications, and comprehensive cardiovascular risk profiles was not uniformly available across the consortium. Additionally, the study population comprised patients who presented for eye care, potentially introducing selection bias toward those with greater healthcare engagement.
Generalizability to populations with limited access to ophthalmic care remains uncertain, particularly given the observed associations between community affluence and mortality. Future prospective studies incorporating detailed systemic assessments would help elucidate mechanisms underlying the DR-mortality relationship.
Conclusion
The SOURCE consortium study provides robust evidence that baseline DR severity independently predicts mortality risk in patients with diabetes mellitus. The gradient of increasing hazard—from 31-34% elevation in NPDR to 87-128% elevation in PDR—underscores the systemic significance of ocular microvascular disease. These findings support the emerging paradigm of oculomics as a tool for comprehensive health risk assessment and reinforce the importance of multidisciplinary coordination in diabetes care. Eye care professionals are uniquely positioned to contribute to early identification of high-risk patients, potentially improving outcomes through timely intervention and intensified medical management.
Reference
Lee SS, Marwah S, Gaddam S, French DD, Bryar PJ, Andrews CA, Evans CT, Rivera AS, Zhou X, Lavine JA, Stein JD, Sight Outcomes Research Collaborative (SOURCE) Consortium. Towards an Era of Oculomics: The Relationship Between Baseline Severity of Diabetic Retinopathy and Risk of Mortality. Ophthalmology. 2026-03-31. PMID: 41933805.