With greater awareness of DR risk factors, improved glycemic control, and broader access to community screening, DR prevalence and incidence have declined in developed nations like the U.S., Australia, and many parts of Europe. A systematic review and meta-analysis (1975–2008) revealed significantly lower DR prevalence and severe visual loss rates post-1985. The 10-year incidence of PDR and SVL dropped from 11.5% and 6.0%, respectively, to 6.6% and 2.6%. In WESDR, the annual incidence of PDR declined from 3.4% to 1.4%, and CSME incidence fell from 1.0% to 0.4% among T1DM patients 29.

Despite these improvements, more recent rural studies show higher DR prevalence compared to metropolitan areas, reflecting disparities in healthcare access 30. Although a clear decline in DR incidence among T1DM patients is documented, trends among T2DM populations remain less certain and require further study.

The Diabetes Control and Complications Trial (DCCT) and the UK Prospective Diabetes Study (UKPDS) were landmark studies demonstrating that tight glycemic control (HbA1c of 7% or lower) can reduce the risk of DR development and progression in both Type 1 (T1DM) and Type 2 Diabetes (T2DM) patients 31. In the DCCT for T1DM, intensive treatment (median HbA1c of 7.2%) led to a 76% reduction in DR incidence and a 54% reduction in DR progression compared to conventional treatment (median HbA1c of 9.1%) 31. In the UKPDS, intensive glycemic control resulted in a 25% reduction in DR risk, including the need for laser photocoagulation 32. For every 1% decrease in HbA1c, there is a 40% reduction in DR development, a 25% reduction in progression to vision-threatening DR (VTDR), a 25% reduction in the need for laser therapy, and a 15% reduction in blindness 33.

Intensive glycemic control also reduces the 4-year incidence of DME by 58% 34. The benefits of early intensive glycemic control are long-lasting, with a sustained impact over many years, a phenomenon known as "legacy effect" or metabolic memory. This suggests that early normalization of glycemia can stop the pathological processes caused by hyperglycemia, such as oxidative stress and glycation of proteins and lipids . However, tight control can lead to early worsening of DR and an increased risk of hypoglycemic episodes. The DCCT noted a higher incidence of DR worsening in the intensive treatment group compared to the conventional group (13.1% vs. 7.6%) but this was reversed by 18 months. Additionally, intensive treatment increases the risk of hypoglycemia three-fold 35.

The Action to Control Cardiovascular Risk in Diabetes (ACCORD) study showed that targeting an HbA1c level of less than 6% led to increased mortality and more hypoglycemic events compared to standard treatment 36. However, the ADVANCE trial did not show an increase in death with intensive treatment. It found that an HbA1c target between 6.5% and 7% reduces macrovascular and microvascular events and deaths 36.

While some epidemiological studies did not consistently identify blood pressure as a risk factor for DR, several randomized controlled trials (RCTs) have demonstrated the benefits of tight blood pressure control in reducing DR incidence and progression. The UKPDS, the first RCT to examine this, showed that patients with tight blood pressure control had a 34% reduction in DR progression and a 47% reduction in visual acuity deterioration 37. A 10 mmHg increase in systolic blood pressure corresponds to a 10% increased risk of early DR and a 15% risk of proliferative DR (PDR) or DME 38.

Additionally, medications targeting the renin–angiotensin system, such as angiotensin II receptor blockers (e.g., candesartan and losartan) and angiotensin-converting enzyme inhibitors (e.g., enalapril), have shown additional benefits in slowing DR progression, independent of their blood pressure-lowering effects 39. The Diabetic Retinopathy Candesartan Trials (DIRECT) found that candesartan reduced the incidence of DR by 18% in T1DM and 13% in T2DM, although the results did not reach statistical significance for the primary endpoint in T2DM 39.

The relationship between lipid levels and DR is still unclear. Some studies have shown that increased triglycerides are associated with DR severity, while higher LDL cholesterol and non-HDL cholesterol are associated with DME 39. The Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) study demonstrated that fenofibrate (a triglyceride-lowering agent) reduced the need for laser treatment in patients with DR, particularly those with pre-existing retinopathy 40.

Recent studies have shown that obesity, particularly a high waist-to-hip ratio and BMI above 31 kg/m² in men and 32 kg/m² in women, is associated with a higher risk of DR 40. However, evidence linking BMI directly with DR is still inconsistent. Some studies have found no significant association between obesity and DR progression, while others have linked a high BMI with more severe DR 40.

Puberty increases the risk of DR development, particularly in T1DM. The period post-menarche has been shown to carry a 30% higher risk of DR compared to the prepubertal period 40. Pregnancy, particularly in women with T1DM, increases the risk of DR progression by 2.3 times 33. Early stages of pregnancy with no DR or mild non-proliferative DR show low progression, but among those with NPDR, 47% progress to more severe forms, and 50% of these require laser treatment. Postpartum, 29% of women experience regression of DR 33.

Cataract surgery can lead to DR progression, particularly when preoperative glycemic control is poor 14. The shift to phacoemulsification surgeries has reduced the risk of DR progression post-surgery compared to older cataract extraction techniques. For those with severe DR (PDR), panretinal photocoagulation should ideally be performed before cataract surgery to reduce the risk of postoperative DME 15.

Chronic inflammation plays a role in the pathogenesis of DR. Elevated levels of inflammatory markers such as C-reactive protein (CRP), intercellular adhesion molecule 1 (ICAM-1), and vascular cell adhesion molecule 1 (VCAM-1) have been linked to DR progression 41. Elevated ICAM-1 levels, for example, have been associated with retinal hard exudates, a hallmark of DR 41.

Genetic factors contribute to DR development, although research is still in the early stages. Twin studies and family studies suggest that individuals with a family history of DR are at a significantly higher risk of developing DR themselves. Several genetic markers have been proposed, including genes related to aldose reductase (ALR2), vascular endothelial growth factor (VEGF), and the receptor for advanced glycation end products (RAGE) 41. However, many of these associations have been weak or inconsistent, and the field requires further study.

The risk factors for DR can be categorized into modifiable and non-modifiable factors (Table 1). Modifiable factors include hyperglycemia, hypertension, hyperlipidemia, and obesity (F 1), while non-modifiable factors include the duration of diabetes, puberty, and pregnancy, all of which contribute to the development and progression of DR.

A decrease of 1% in HbA1c can lead to a 40% reduction in retinopathy, a 25% reduction in the need for retinal laser treatment, and a 15% reduction in blindness risk for individuals with diabetes 41.

A 10-mmHg reduction in systolic blood pressure can decrease the risk of retinopathy by 35%, the need for retinal laser treatment by 35%, and the risk of blindness by 50% 42. However, two Asian clinic-based studies did not find a significant relationship between blood pressure and the incidence or progression of DR.

DR is associated with elevated triglyceride levels, while diabetic macular edema (DME) is linked to higher LDL cholesterol, non-HDL cholesterol, and an unfavorable HDL/LDL ratio 42.

Increased waist–hip ratio and BMI greater than 31 kg/m² for men and 32 kg/m² for women are associated with a higher risk of DR development. In contrast, BMI below 20 kg/m² also increases the risk of DR 42.

Post-pubertal individuals have a 30% higher risk of DR development. Furthermore, the onset of DR is typically faster (2 years earlier) compared to those in the prepubertal stage 42.

Pregnancy increases the risk of DR progression by 2.3 times . During the postpartum period, 29% of pregnant women experience DR regression. However, pregnant women with pre-existing DR are at a significantly higher risk of progression, with 47% experiencing progression and 50% requiring laser treatment 42.

Early detection and timely intervention can prevent up to 98% of diabetes-related visual impairment 6. Various screening methods are used for DR, including retinal photography, direct ophthalmoscopy, and tele-retinal screening. The International Council of Ophthalmology recommends that eye care providers evaluate visual acuity, HbA1c levels, blood pressure, and other diabetes-related complications to determine the urgency of referral to ophthalmologists 7.

Different methods for screening diabetic retinopathy have been developed to cater to various settings, including primary care, specialized clinics, and telemedicine platforms. These methods differ in terms of the technology used, the need for pupil dilation, and the level of training required for professionals interpreting the results.

These methods aim to identify different stages of diabetic retinopathy, from early non-proliferative diabetic retinopathy (NPDR) to advanced proliferative diabetic retinopathy (PDR). Table 2 highlights the sensitivity (the ability to correctly identify individuals with DR) and specificity (the ability to correctly identify individuals without DR) of various screening methods. Higher sensitivity and specificity are ideal as they reduce both false-negative and false-positive results, ensuring effective detection and management of DR.

In many countries, including the United Kingdom, the United States, and Australia, national screening programs have been developed to detect DR early. For example, the National Health Service (NHS) in the UK recommends using two-field mydriatic retinal photography for screening, aiming for a sensitivity of at least 80% and specificity above 95%.

In contrast, telemedicine-based screening programs have proven effective, especially in rural and underserved areas. Tele-retinal screening involves digital retinal imaging and remote interpretation, increasing access to screening while minimizing costs. For example, in the United States, telemedicine programs have expanded DR screening to veterans and military personnel, reducing the need for in-person visits.

Non-mydriatic retinal photography, which doesn't require pupil dilation, is becoming more common in primary care settings. However, while it provides a convenient and accessible option, its higher failure rate and inability to provide stereoscopic views may limit its effectiveness in detecting advanced DR stages.

Classifying the severity of DR is critical for determining the appropriate management and treatment. Several classification systems have been developed to help practitioners identify the extent of the disease, ranging from simple grading systems for primary care to more detailed systems for research settings.

One widely used classification system is the Airlie House seven standard 30° stereoscopic fields with the ETDRS (Early Treatment Diabetic Retinopathy Study) grading scale, which assigns severity levels for DR based on clinical findings. For primary care settings, simpler systems are often preferred, such as the World Health Organization (WHO) classification, which categorizes DR into three levels based on clinical findings that indicate the need for referral:

Another commonly used classification system is the International Clinical Diabetic Retinopathy and Diabetic Macular Edema Disease Severity Scale, which helps in diagnosing the severity of DR based on fundus findings such as microaneurysms, hemorrhages, and retinal vascular changes (Table 3).

The UK National Institute for Clinical Excellence (NICE) guidelines recommend DR screening tests to have a sensitivity of at least 80% and specificity of 95%, with a technical failure rate under 5%. The Public Health England has introduced guidelines for the National Health Service (NHS) diabetic eye screening programs, emphasizing the importance of two-field mydriatic retinal photography.

Mydriatic retinal photography, when combined with ophthalmoscopy for ungradable cases, is considered the most effective DR screening strategy. It ensures better-quality retinal images with a minimum sensitivity of 80% in detecting any grade of DR.

However, pupil dilation remains a concern for some healthcare providers, due to the risk of acute angle-closure attacks in certain populations, particularly those of Asian descent.

Non-mydriatic retinal photography is popular in primary care since it doesn't require pupil dilation. While it offers a cost-effective option, it comes with challenges such as higher technical failure rates and difficulties in obtaining stereoscopic views.

Tele-retinal and mobile eye screening have proven to be cost-effective methods for DR screening in many countries, including Australia, the United States, the United Kingdom, and India. Tele-retinal screening uses digital retinal imaging combined with remote interpretation, which significantly enhances access to DR screening, especially in rural and remote areas.

For instance, in the United States, a national tele-retinal imaging program has been implemented, providing services to veterans and military personnel. Similarly, the National Health Service in the United Kingdom has developed a National Diabetic Retinopathy Screening Program, aiming for a 100% screening rate for patients with diabetes.

A major strength of this review lies in its comprehensive and global perspective, covering diverse geographic regions and populations. By integrating data from both high- and low-income countries, the review offers a nuanced understanding of diabetic retinopathy (DR) prevalence, risk factors, progression, and screening strategies across different healthcare systems. Another strength is the inclusion of both type 1 and type 2 diabetes, highlighting key differences in epidemiology and disease progression, which is crucial for tailoring prevention and management strategies. The review also synthesizes findings from large, landmark studies such as the DCCT, UKPDS, WESDR, and newer telemedicine-based screening programs, providing robust evidence for clinical decision-making and policy formulation.

However, several limitations must be acknowledged. First, heterogeneity in study methodologies, including differences in DR classification systems, imaging modalities, and diagnostic criteria, makes direct comparison challenging. Some studies rely on self-reported diabetes duration or single-timepoint HbA1c measures, which can reduce data accuracy. Geographic and population coverage is uneven; while data from North America and Western Europe are well represented, fewer high-quality studies are available from sub-Saharan Africa, parts of Latin America, and rural Asia. Moreover, the majority of incidence and progression data are derived from historical cohorts, which may not reflect recent advances in diabetes care, particularly in low- and middle-income countries. Lastly, language and publication bias may limit inclusion of non-English or unpublished local studies, particularly from underrepresented regions.

This review reinforces the importance of early, individualized, and sustained management of diabetes to prevent or delay the onset and progression of DR. Clinicians must prioritize tight glycemic control, particularly early in the disease course, to leverage the "legacy effect" and reduce long-term retinal damage. Equally important is the aggressive management of hypertension and dyslipidemia, which, when controlled, can significantly reduce the risk of DR progression and associated visual impairment. The evidence also supports the need for regular, structured DR screening using validated imaging modalities. Implementing national screening programs and telemedicine services can dramatically improve detection and timely referral, especially in resource-constrained settings. Furthermore, clinicians should be aware of life stages such as puberty and pregnancy that heighten DR risk, necessitating closer monitoring. Ultimately, integrating DR screening into routine diabetes care and empowering primary care providers to initiate screening and risk management will be key to reducing the global burden of diabetic vision loss.

Future research should prioritize high-quality, longitudinal studies from underrepresented regions such as sub-Saharan Africa, rural Latin America, and Southeast Asia to address data gaps and improve global generalizability. There is also a need to evaluate the cost-effectiveness, scalability, and long-term outcomes of emerging screening models, including artificial intelligence–assisted retinal image analysis and smartphone-based fundus photography. Research should focus on optimizing risk stratification tools that incorporate clinical, genetic, and socio-demographic variables to guide personalized screening intervals and management plans. Moreover, studies examining barriers to screening uptake, particularly among underserved populations, and evaluating the effectiveness of community-based interventions and health education programs would provide actionable insights. Finally, interventional trials comparing the impact of newer antihyperglycemic agents, blood pressure drugs, and lipid-lowering therapies specifically on DR progression will be crucial to refine therapeutic strategies.