A cross-sectional study on the prevalence and risk factors of age-related macular degeneration

Objectives

1. To determine the prevalence of AMD among individuals aged 50 years and above attending Jagjivanram Hospital, a tertiary care hospital in Mumbai

2. To identify demographic, systemic and lifestyle risk factors – including age, smoking, hypertension, diabetes and sunlight exposure – associated with AMD

3. To classify AMD cases into non-neovascular and neovascular and further grade into early, intermediate and late stages based on fundus examination findings using AREDS criteria.

Study design and setting

This was a hospital-based, cross-sectional observational study conducted at the Department of Ophthalmology, Jagjivanram Railway Hospital, Mumbai, Maharashtra, India, from January to June 2023.

Study population

A minimum sample size of 138 was calculated using an expected AMD prevalence of 10%, a precision of 5% and 95% confidence level. To account for incomplete data, 150 participants, aged 50 years and above, attending the outpatient clinic during the study, were enrolled. Patients with media opacities precluding fundus examination or a history of retinal surgery were excluded.

Data collection

Demographic data (age, gender), systemic comorbidities (diabetes, hypertension) and lifestyle exposures (tobacco use, daily sunlight exposure, use of protective eyewear) were recorded using a structured interviewer-administered questionnaire. Sunlight exposure was self-reported based on average daily outdoor duration (<2 hours or ≥2 hours), along with occupational history and use of caps or sunglasses.

Ophthalmic examination

All participants underwent best-corrected VA testing (Snellen chart), slit-lamp biomicroscopy, intraocular pressure measurement and dilated fundus examination using a 90D lens. AMD was diagnosed and staged according to AREDS classification into non-neovascular (dry) and neovascular (wet) forms and into early, intermediate or late stages.

Statistical analysis

Data were analysed using Statistical package for the social sciences (version 25). Categorical variables were expressed as frequencies and percentages. Chi-square or Fisher’s exact test assessed associations between AMD and risk factors; P < 0.05 was considered statistically significant.

Visual acuity

Moderate visual impairment (VA <6/18 in better eye) was present in 23.3% of participants, while the majority retained VA between 6/6 and 6/18. The most common VA group was 6/9 (24%) [Table 2]. The VA distribution represents the overall study population. Visual loss directly attributable to AMD is recorded in the third column.

Risk factor associations

AMD was documented in 16.7% (n = 25) of participants. The prevalence was highest in the 80–89 years of age group (22.6%). However, no statistically significant association was found between AMD and age (P = 0.547). Similarly, no significant association was observed for gender (P = 0.531) or hypertension (P = 0.414).

A higher prevalence was observed in diabetics (22.5%) compared to non-diabetics (11.4%), with borderline significance (P = 0.054).

Smoking demonstrated a strong and statistically significant association with AMD (24.0% in smokers vs. 9.3% in non-smokers, P = 0.014).

Daily sunlight exposure of <2 h was associated with a higher prevalence of AMD (22.2%) compared to those with greater exposure (10.1%), P = 0.038. The associations between AMD and various sociodemographic and systemic risk factors are summarized in Table 3.

AMD types and stages

Among 25 AMD cases, dry AMD was present in 13 (8.7%) and wet AMD in 12 (8.0%). By severity, 6 (24%) were early stage, 8 (32%) intermediate, and 11 (44%) late stage. The distribution of AMD presence, type, and grading is shown in Table 4.

Smoking as a major modifiable risk factor

Smoking emerged as a statistically significant determinant of AMD in this study (P = 0.014), with smokers showing a markedly higher prevalence (24.0%) compared to non-smokers (9.3%), corroborating findings from the Blue Mountains Eye Study and the Beaver Dam Eye Study, both of which established cigarette smoking as the strongest modifiable risk factor for AMD.^[7,8] Population-based epidemiological studies have further demonstrated that smoking acts as an important modifier of AMD risk across age and gender strata, contributing to both disease onset and severity.^[11] In addition, pooled analyses from multiple continents have confirmed advancing age, smoking, and systemic vascular factors as significant determinants of incident AMD, reinforcing the global relevance of smoking as a modifiable exposure.^[14] Longitudinal cohort studies have also shown that smoking increases both the incidence and long-term progression of AMD, with a dose-dependent relationship observed in several populations.^[15]

The biological mechanisms linking smoking to AMD are multifactorial and include increased oxidative stress, direct toxicity to the retinal pigment epithelium, impaired choroidal blood flow, and upregulation of inflammatory and angiogenic pathways. Chronic exposure to free radicals and carbon monoxide is thought to accelerate photoreceptor damage and compromise Bruch’s membrane integrity, thereby predisposing individuals to both non-neovascular and neovascular forms of AMD.

In addition to active smoking, there is growing evidence that exposure to secondhand smoke may also increase the risk of AMD. Individuals with significant passive smoke exposure have been reported to exhibit a higher likelihood of developing AMD compared with unexposed non-smokers.^[11] Although passive smoking was not specifically assessed in the present study, this represents an important limitation and highlights an area for future research, particularly in densely populated urban settings.

Sunlight exposure: Contradictory findings

An intriguing observation in this study was the inverse association between sunlight exposure and AMD, with participants exposed to more than 2 h of daily sunlight showing a lower prevalence of disease (10.1% vs. 22.2%, P = 0.038). This finding appears contradictory to the established consensus that prolonged and unprotected exposure to UV radiation increases oxidative stress within the retina and accelerates AMD pathogenesis.^[12] Long-term population-based cohort studies have demonstrated a progressive increase in the incidence and severity of AMD with advancing age and cumulative lifetime exposure, underscoring the importance of longitudinal environmental influences on disease pathogenesis.^[16] Only after this established evidence, long-term follow-up data from the Blue Mountains Eye Study have further demonstrated that smoking significantly increases the long-term incidence of AMD.^[17] The apparent inverse association observed in the present study is therefore unlikely to represent a true protective effect of sunlight exposure.

Several methodological factors may account for this discrepancy. Sunlight exposure in this study was assessed using self-reported average daily duration, without objective quantification of ultraviolet exposure, cumulative lifetime exposure, or differentiation between occupational and recreational outdoor activity. This introduces the potential for recall bias and exposure misclassification. Additionally, the use of protective measures such as caps, hats, or sunglasses was not quantitatively analysed, which may have further confounded the observed association.

Importantly, these findings should not be interpreted as evidence supporting increased sunlight exposure as a protective factor against AMD. Rather, the results highlight the limitations inherent in cross-sectional, questionnaire-based exposure assessment and reinforce the need for cautious interpretation. Future studies incorporating objective UV exposure measurements, detailed occupational histories, and longitudinal follow-up are required to clarify the true relationship between sunlight exposure and AMD risk.

Age, gender and hypertension: null associations

Contrary to global evidence, this study did not identify statistically significant associations between AMD and age (P = 0.547), gender (P = 0.531) or hypertension (P = 0.414). This result diverges from landmark cohort studies such as the Beaver Dam Eye Study and the Blue Mountains Eye Study, which consistently demonstrated age as the single most important risk determinant.^[7,8] Several possible explanations exist:

1. Sample size and power – With 150 participants, the study may have been underpowered to detect subtle associations, especially within older age strata

2. Survivor bias – Elderly individuals with severe systemic or ocular comorbidities may have been less likely to attend the hospital, thereby skewing the sample

3. Hospital-based sampling – Unlike population-based studies, tertiary care cohorts may reflect selective health-seeking patterns rather than true community prevalence.

Nevertheless, the absence of significant associations does not negate the well-established roles of age and hypertension as risk factors. Instead, these null findings should be interpreted as limitations of study design and sample size, highlighting the need for larger, community-based studies in India.

In addition, AMD is strongly age-linked, but the present cohort contained a relatively younger elderly population, with two-thirds of participants below 70 years.^[18,19] This restricted age variability likely reduced the ability to detect age-related differences. Further, lifetime exposure variables such as cumulative smoking burden, long-term sunlight exposure, systemic vascular status and genetic susceptibility were not captured in detail, limiting the explanatory strength of null findings.

Diabetes mellitus

A higher prevalence of AMD among diabetics (22.5%) compared with non-diabetics (11.4%) approached statistical significance (P = 0.054). Although diabetes is not universally recognised as a primary AMD determinant, chronic hyperglycaemia, microvascular ischaemia and inflammatory cytokine release may accelerate oxidative retinal injury. Several longitudinal studies, including those from Japan and western cohorts, have documented higher AMD incidence among individuals with poor glycaemic control.^[13] These findings emphasise the need for comprehensive ocular screening in patients with long-standing diabetes, even in the absence of diabetic retinopathy.

Genetic susceptibility

Beyond environmental exposures, genetic predisposition plays a critical role in AMD pathogenesis. Variants in the CFH gene, ARMS2/HTRA1 locus and genes involved in oxidative stress and angiogenesis (e.g. VEGF polymorphisms) have been strongly implicated.^[9,15] While genetic polymorphisms were not analysed here due to resource limitations, future multi-centric Indian studies should integrate CFH and ARMS2/HTRA1 genotyping to better elucidate gene– environment interactions in AMD.

Public health implications

The predominance of late-stage AMD in this cohort underscores the urgent need for:

• Routine retinal screening for individuals ≥50 years, especially smokers and diabetics

• Community education on modifiable risk factors, focusing on smoking cessation and diabetes control and safe sunlight practices

• Promotion of safe protective eyewear (caps, hats, sunglasses) to mitigate UV-related risks

• Incorporation of AMD detection into the National programme for control of blindness and visual impairment.

Limitations

This study is limited by a relatively small sample size of 150 participants from a single tertiary hospital, which restricts generalizability to the broader population. The cross-sectional design limits the ability to infer causality between exposures and AMD. Associations identified, including smoking and sunlight exposure, represent correlations rather than causal relationships. Future larger population-based and longitudinal studies with objective UV assessment and genetic profiling are essential to clarify these associations.