The advent of intravitreal anti-vascular endothelial growth factor (VEGF) therapy has transformed the prognosis of neovascular age-related macular degeneration (nAMD), yet outcomes in routine clinical practice remain uneven [1]. The analysis reported in this issue of Eye provides important insights into how social deprivation influences both presentation and visual outcomes in nAMD within the National Health Service (NHS) in England [2].

The Royal College of Ophthalmologists should be commended for establishing the National Ophthalmology Database (NOD) AMD Audit, now encompassing more than 48,000 treated eyes, initially led by Martin McKibbin and now under the leadership of Romi Chhabra. In this first publication from the initiative, the investigators identified that patients from the most deprived communities tended to present with worse baseline visual acuity and to experience poorer visual outcomes at 12 months compared with those from the least deprived areas [2].

What makes these findings particularly notable is what does not appear to explain them. Completion of the loading phase and the median number of injections during the first year varied little between socioeconomic groups. In other words, once patients entered the NHS treatment pathway, care appeared broadly equitable.

Yet outcomes remained unequal.

These findings highlight a fundamental reality: many determinants of visual outcome are established long before the first injection is given. Baseline visual acuity is one of the strongest predictors of long-term outcome in nAMD, and real-world studies consistently demonstrate that outcomes depend strongly on both baseline vision and treatment intensity [1, 3,4,5,6,7].

Access to primary eye care is likely to play an important role. Less access to optometrists in more deprived communities may delay detection of early nAMD until vision has already deteriorated significantly. Broader social determinants of health may also influence the timing of presentation [8, 9]. Competing medical or socioeconomic priorities may also delay presentation of new onset distortion. In a condition such as nAMD, where timely treatment is critical, even relatively short delays in presentation may result in irreversible visual loss.

Engagement with long-term treatment may also differ across socioeconomic groups, as intravitreal therapy requires repeated visits over many years and often depends on transportation and time away from work for carers. The rates of loss to follow-up were, in general, higher in more deprived communities. Treatment burden and healthcare utilisation associated with intravitreal therapy are increasingly recognised challenges in nAMD care [10]. Loss to follow-up remains a significant challenge globally, with multiple studies identifying socioeconomic factors, logistical barriers and treatment burden as important determinants of adherence [11, 12]. This study was limited to 12-month outcomes, and longer-term follow-up may reveal even more pronounced associations between social deprivation and visual outcomes.

An intriguing aspect of the present analysis is the distinction between first and second treated eyes. The socioeconomic gradient in baseline visual acuity was observed primarily in first treated eyes but not in second eyes, which are often diagnosed through monitoring within secondary care [13]. This observation suggests that once patients are engaged within retinal services, disparities in disease detection may be partially mitigated. However, worse visual outcomes were still observed across deprivation groups, even for second eyes, implying that factors beyond delayed presentation continue to influence long-term outcomes.

Incomplete referral pathway data in the audit also highlights a broader structural challenge for large real-world ophthalmology datasets. As ophthalmology increasingly relies on large-scale real-world evidence to inform service planning and policy, the robustness of these datasets becomes critical. Ophthalmology now produces some of the largest real-world datasets in medicine; the greatest barrier to insight is no longer data volume, but data completeness. In addition to missing referral pathway data, there was also substantial missing visual acuity data.

Furthermore, not all NHS services providing nAMD treatment contributed data to the audit. Integrated Care Boards could strengthen the completeness and utility of national datasets by making participation in the National Ophthalmology Database audit a commissioning requirement for services delivering AMD care.

Addressing this issue will require system-level solutions. Contracts with electronic medical record (EMR) vendors should include provisions for ongoing audit support and standardised data extraction to remove cost barriers to service providers to participate in national audits. Given that only three EMR systems—EPIC, mediSIGHT and OpenEyes—were used by the vast majority of service providers contributing to this audit, significant opportunities exist to improve data completeness with vendor cooperation and better data harmonisation by adoption of Fast Healthcare Interoperability Resources [14]. In parallel, the adoption of minimum mandatory datasets determined by international consensus would improve consistency and data quality across services [15].

As retinal therapies continue to evolve—with longer-acting drugs, sustained-delivery systems and gene therapies on the horizon—treatment burden should decrease and we hope long-term adherence will improve [16, 17]. Therefore, earlier detection of neovascular conversion in high-risk patients will become increasingly important. Emerging technologies such as home or community-based monitoring systems, AI-assisted triage, and tele-ophthalmology-enabled referral pathways could support earlier detection and more efficient referral of nAMD [18, 19].

The work reported in this issue, therefore, highlights an uncomfortable but necessary reality. The success of modern AMD therapy is not determined solely by pharmacotherapy or clinic efficiency, but also by where the patient lives and how easily they can access care. As the therapeutic landscape of AMD continues to advance, ensuring that earlier diagnosis and sustained engagement with eyecare are achievable for all patients—irrespective of socioeconomic background—will be as important as the development of new treatments themselves. Addressing these disparities will require coordinated action across retinal services, primary eye care, technology providers, and healthcare commissioners to ensure that early diagnosis, consistent monitoring and equitable access to treatment are a reality for all communities.