Age-Specific Increases in the Prevalence and Treatment Burden of Neovascular Age-Related Macular Degeneration in a Super-Aged Society

Abstract

Purpose

To characterize age-specific changes in the prevalence and treatment burden of neovascular AMD (nAMD) from 2014 to 2019 in a super-aged society and to assess whether comorbidities contribute to this increase.

Methods

We performed a nationwide repeated cross-sectional study using Japan's national claims database (>95% population coverage). Adults aged ≥40 years in 2014 and 2019 were included to estimate age-stratified nAMD prevalence. Treatment burden was evaluated as the total number of nAMD-related treatment claims and mean administrations among users of each agent. Diabetes, hypertension, and dyslipidemia were identified from claims. Multivariable logistic regression compared nAMD cases with controls and assessed whether comorbidity adjustment attenuated the odds of nAMD in 2019 versus 2014.

Results

We identified 76,125 cases in 2014 and 125,190 in 2019; age-standardized prevalence increased from 101.3 to 152.4 per 100,000. Prevalence peaked at ages 80–84 in both years but increased most steeply among those ≥75 years (62.9%). Total treatment claims increased markedly, driven primarily by aflibercept. Among users of each agent, mean administrations increased modestly. Comorbidities were associated with higher odds of nAMD, but adjusting for them did not change the higher odds of nAMD in 2019 compared with in 2014.

Conclusions

Between 2014 and 2019, the prevalence and treatment burden of nAMD increased substantially in Japan, with a disproportionate rise among older age groups in a super-aged population. Comorbidities did not account for the temporal increase, suggesting that factors beyond aging and comorbidity are contributing to the growing burden.

AMD is a prototypical ocular aging-related disorder that is the leading cause of vision loss in older adults worldwide.¹ Several studies have suggested that some age-related diseases, such as cancer and diabetes, become less prevalent among the oldest old, indicating a non-linear association between aging and disease risk.^2,3 Emerging evidence suggests that AMD may follow a similar pattern, with its prevalence declining in the most advanced age groups.^4,5 Of its subtypes, neovascular AMD (nAMD) carries a high risk of visual deterioration and requires frequent anti-VEGF injections, creating a substantial healthcare burden.

Japan is a super-aged society in which adults aged 65 years or older comprise more than 21% of the total population, and its rapidly advancing population aging has important public health implications for nAMD.^6,7 Despite the recognition of nAMD as a major public health issue, only limited epidemiological data are available on the trends in nAMD prevalence within Japanese older adults over time.^8–12 Moreover, it remains unclear how diseases like nAMD—similar to cancer and diabetes, which have been shown to decline in prevalence among the oldest old—behave in the context of further demographic aging.

Using the National Database of Health Insurance Claims and Specific Health Check-ups of Japan (NDB), which covers more than 95% of the population,^13–15 a prior study reported the incidence of active exudative AMD and described real-world initial treatment patterns and injection frequency among newly diagnosed patients from 2011 to 2018.⁵ However, temporal changes in treated nAMD prevalence and national-level treatment burden across age strata have not been directly quantified at multiple time points. The lack of comprehensive, population-based data on temporal changes in AMD prevalence imposes challenges to the assessment of the national-level burden of nAMD and development of targeted public health interventions.

To fill this evidence gap, we analyzed the NDB in two pre-pandemic years, 2014 and 2019, providing a population-based, repeated cross-sectional assessment of evolving treated nAMD epidemiology and treatment burden in Japan's rapidly aging society. Because cardiometabolic diseases have been implicated as risk factors for AMD, we also examined whether the comorbidities might partly account for the rising burden of nAMD in this super‑aged society.

Methods

Study Design and Ethical Approval

This retrospective nationwide population-based study using anonymized data from the NDB was approved by the Institutional Review Board and the Ethics Committee of Kyoto University Hospital and Kyoto University Graduate School of Medicine (No. R3987). All procedures adhered to the tenets of the Declaration of Helsinki and its subsequent amendments. The requirement for individual informed consent was waived for the NDB portion because of its anonymized nature.

Data Sources

We used data from the NDB. This extensive database, managed by Japan's Ministry of Health, Labor, and Welfare, covers more than 95% of all medical claims issued in Japan, encompassing data from over 126 million individuals, and generates around 1.6 billion electronic claims annually.^13–15 The NDB contains a comprehensive range of information, including unique identifiers for patients, demographic information including age and sex, and diagnostic details aligned with the International Classification of Diseases, Tenth Revision (ICD-10). This information is coded as per the local claim codes of Japan which include: NDB diagnostic codes, NDB drug codes, and NDB procedure codes. Access to the NDB was granted by the Ministry of Health, Labour, and Welfare, and data were accessed through the NDB Onsite Research Center in Kyoto, which is one of only two centers that are authorized to provide remote access to the complete NDB dataset. This study was conducted during the authorized research period between August 28, 2023, and August 28, 2024. The detailed information of this database has also been presented in previous research.^5,16–22

Study Population and Definitions

First, we used two types of identifiers to link the insurance claims of individual patients and to further enhance traceability by using methods that were previously reported.²³ Next, we identified cases of AMD using diagnostic codes specific to each AMD subtype. We focused on treatment-requiring nAMD because it is a vision-threatening subtype that necessitates high-cost, repeated treatments with anti-VEGF therapy or photodynamic therapy (PDT), making its national burden directly relevant to healthcare planning and resource allocation. We included adults aged ≥40 years to comprehensively capture the national burden of treatment-requiring nAMD while minimizing arbitrary exclusions. Although AMD is uncommon in individuals younger than 50 years, the Japanese clinical practice guideline for nAMD notes that nAMD can occur in a small number of younger patients and does not recommend a strict age cut-off when considering the diagnosis.²⁴ To enhance the accuracy of patient classification, we defined cases based on both diagnostic criteria and specific treatment interventions. Treated nAMD cases were identified as individuals with an AMD diagnosis recorded in the NDB and evidence of nAMD-related treatment in claims (intravitreal anti-VEGF therapy and/or verteporfin PDT) within the study year, as established in previous studies.^5,25 This claims-based definition leverages the strengths of administrative claims data to more accurately capture treatment-requiring nAMD and is not intended to estimate the prevalence of AMD across all stages. For the precise identification of nAMD cases, we used NDB diagnostic codes, which offer greater specificity than the standard ICD-10 codes. A correspondence table aligning NDB diagnostic codes with ICD-10 codes is included in the Supplementary Table S1.

Factors Evaluated

We summarized the number of patients with nAMD according to age and sex categories for 2014 and 2019. We determined the prevalence rates stratified by age and sex by dividing the number of nAMD within each group by the population at risk within the corresponding group that was published online by the Japanese government.²⁶ Age-specific prevalence was estimated using five-year age bands. For descriptive purposes, we additionally summarized prevalence using broader age-threshold categories to provide a concise comparison between non-elderly adults and older age groups. Age-standardized prevalence rates were calculated using the direct method with the 2015 Japanese standard population as the reference. We selected the 2015 Japanese standard population because it is widely used in Japanese epidemiologic studies and provides a natural midpoint reference between 2014 and 2019. We summarized changes in Japan's age structure between 2014 and 2019 using official population statistics (Supplementary Table S2). To estimate the mean age at new-onset treated nAMD, we identified incident treated cases using a one-year washout period and defined onset age as the age at the first record meeting the treated nAMD definition in each calendar year. The covariates evaluated in this study included comorbid conditions, such as diabetes, dyslipidemia, and hypertension. Each comorbid condition was defined using diagnostic codes from medical records, with disease definitions validated by prior studies.^27,28

In addition, to quantify the national treatment burden and its temporal change using a consistent claims-based approach across years, we measured treatment burden as the number of claims for intravitreal anti-VEGF administrations (aflibercept, ranibizumab, pegaptanib) and verteporfin PDT. For each patient, we counted the number of treatment administrations recorded in claims during a standardized 6-month window (July–December) in 2014 and 2019. This window was chosen because of practical constraints in data extraction and to provide a standardized assessment period using identical calendar months across years for comparability. Because eye laterality is not consistently available in the NDB, administrations were summarized at the patient level and were interpreted as claims-based administrations rather than eye-specific injection counts. For each anti-VEGF agent and PDT, we also assessed the number of treated patients who received the agent at least once during the six-month period and the mean number of administrations per patient among those recipients. For age-stratified analyses, we examined the mean total number of intravitreal anti-VEGF administrations per treated patient during the period, combining all anti-VEGF agents.

Association Between nAMD and Systemic Comorbidities

To examine associations between treated nAMD and systemic comorbidities and to assess whether these comorbidities could explain temporal changes, we constructed year-specific case–control datasets for 2014 and 2019. For each study year, controls were defined as individuals aged ≥40 years in the NDB with no recorded AMD diagnosis codes in the corresponding year. To construct the control cohort, we randomly sampled claims from a 6-month period among individuals aged ≥40 years, identified individuals from those claims, excluded individuals meeting the treated nAMD definition for that year, and then randomly selected 10% of the remaining eligible individuals as controls. This procedure was performed independently for 2014 and 2019.

Statistical Analysis

All statistical analyses were performed using PostgreSQL EnterpriseDB version 14.1 and R version 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria). Continuous variables were compared between groups by using the Student t-test or Welch t-test depending on variance homogeneity. Prevalence differences within each five-year age band between 2014 and 2019 were evaluated using Wald tests with two-sided 95% confidence intervals (CIs). Age-standardized prevalences across all ages were calculated using the direct method with the 2015 Japanese population as the standard. Standard errors for standardized estimates were computed via the delta method and used to derive 95% CIs and P values.

We used logistic regression with treated nAMD status as the dependent variable (case vs. control). We first fitted a univariable model including calendar year (2019 vs. 2014) to obtain the crude odds ratio (OR). We then fitted Model 1 adjusted for age and sex and Model 2 further adjusted for diabetes mellitus, hypertension, and dyslipidemia (binary indicators). ORs and 95% CIs were estimated for all covariates, and change in the calendar-year odds ratio from Model 1 to Model 2 was used to assess attenuation by comorbidities. A two-sided P value ≤ 0.05 was considered statistically significant.

Results

Changes in nAMD Prevalence and Patient Characteristics

The age-specific prevalence of nAMD in both 2014 and 2019 rose with age, peaking among individuals aged 80–84 years, and then declined in the oldest old (Fig. 1). A total of 76,125 nAMD cases were identified in 2014 among individuals aged ≥40 years (crude prevalence: 100.40 per 100,000), which increased to 125,190 cases in 2019 (160.08 per 100,000) (Tables 1, 2). Among treated nAMD cases identified in 2014, 21,326 had at least one nAMD-related treatment claim in 2019, representing 28.0% of the 2014 cohort and 17.0% of treated nAMD patients in 2019. After age adjustment using the 2015 Japanese standard population, the nAMD prevalence increased from 101.30 to 152.38 per 100,000, which indicated that the increase was not solely attributable to population aging. The difference between crude and age-standardized estimates was modest, consistent with the relatively modest change in Japan's age structure between 2014 and 2019. Mean age at first recorded treated nAMD increased from 72.52 ± 10.43 years in 2014 to 73.97 ± 10.96 years in 2019 (P < 0.001), and mean age of all treated patients increased from 74.75 ± 9.45 to 76.42 ± 9.46 years (P < 0.001). Men accounted for 64% to 65% of cases in both years. Age-distribution analysis showed that the number of cases increased across all age groups, with marked increase noted in patients aged ≥75 years (Fig. 2; Table 2). In 2019, the prevalence per 100,000 population was 29.77, 192.55, 435.44, and 411.52 for those aged 40–64, 65–74, 75–84, and ≥85 years, respectively (Supplementary Table S3). Sex-specific analyses demonstrated higher prevalence among older men (Figs. 1B, 1C; Supplementary Tables S4–S7). In men aged ≥40 years, nAMD prevalence increased from 138.37 to 217.12 per 100,000 whereas, in women, it increased from 66.74 to 109.28. The highest male prevalence was observed in men aged 85–89 years (797.88 per 100,000 in 2019). The highest female prevalence was seen in those aged 80–84 years (299.55 per 100,000 in 2019) and 85–89 years (298.42 per 100,000 in 2019) (Supplementary Tables S4–S7). Prevalence of nAMD in 2014 and 2019 across age-threshold categories is also summarized in Supplementary Tables S8–S10.

Figure 1.

Comparison of age- and sex-stratified prevalence rates of nAMD per 100,000 population between 2014 and 2019. (A) Overall age distribution shows a marked left-skewed pattern in both years, with a pronounced peak in individuals aged 80–84. (B) Among males, the prevalence increased notably in all age groups ≥70 years, with higher rates than females across all strata. (C) Female prevalence also rose, particularly among those ≥75 years, but remained consistently lower than male rates.

Table 1.

Baseline Characteristics of Patients With nAMD in 2014 and 2019

	2014 nAMD	2019 nAMD
Total Cases (≥40 years)	76,125	125,190
Age, years, mean (SD)	74.75 (9.45)	76.42 (9.46)
Age, years, median (IQR)	76 (69–81)	77 (71–83)
Age at first recorded for New Cases	72.52 (10.43)	73.97 (10.96)
Sex
Male	49,291 (64.7%)	79,994 (63.9%)
Female	26,834 (35.3%)	45,196 (36.1%)
DM Present	30,479 (40.0%)	53,259 (42.5%)
HT Present	49,884 (65.5%)	86,839 (69.4%)
DL Present	28,144 (37.0%)	48,289 (38.6%)

DL, dyslipidemia; DM, diabetes mellitus; HT, hypertension

All percentages refer to the proportion among patients with nAMD in that year.

Table 2.

Age Distribution of Overall Patients With nAMD (2014 Vs. 2019, Age ≥40 Years)

Age Group (Years)	2014 (n = 76,125)	Prevalence Rate Per 100,000 (95% CI)	2019 (n = 125,190)	Prevalence Rate Per 100,000 (95% CI)	Increase in Prevalence (Per 100,000)	Relative Increase in Prevalence
40–44	371	3.79 (3.41–4.19)	518	5.94 (5.44–6.48)	2.15	56.9%
45–49	728	8.46 (7.85–9.09)	1117	11.40 (10.74–12.08)	2.94	34.8%
50–54	1339	17.19 (16.28–18.13)	2051	23.94 (22.92–25.00)	6.75	39.3%
55–59	2443	31.92 (30.67–33.21)	3211	41.64 (40.21–43.11)	9.72	30.4%
60–64	5484	61.08 (59.47–62.71)	5703	75.80 (73.84–77.79)	14.72	24.1%
65–69	9094	99.33 (97.30–101.40)	12,448	142.95 (140.45–145.48)	43.62	43.9%
70–74	14,865	187.50 (184.50–190.54)	21,045	242.29 (239.02–245.58)	54.79	29.2%
75–79	16,448	262.37 (258.38–266.41)	29,017	400.68 (396.08–405.31)	138.31	52.7%
80–84	14,396	295.73 (290.92–300.60)	25,718	482.70 (476.81–488.63)	186.97	63.2%
85–89	8425	275.06 (269.21–280.99)	17,137	474.58 (467.50–481.74)	199.52	72.5%
90–94	2213	169.58 (162.59–176.79)	6235	354.06 (345.33–362.96)	184.48	108.8%
95–99	301	85.75 (76.34–96.01)	929	193.95 (181.67–206.83)	108.19	126.2%
100+	18	30.00 (17.78–47.41)	61	88.41 (67.62–113.56)	58.41	194.7%
Total	76,125	100.40 (99.69–101.11)	125,190	160.08 (159.19–160.97)	59.68	59.4%

Figure 2.

Age-specific increase in the prevalence of neovascular AMD between 2014 and 2019. (A) Age-specific absolute increase. Absolute increases were minimal in younger adults but rose sharply in older age groups. Error bars indicate 95% CIs. (B) Age-specific relative increase. The relative increase was modest at younger ages but rose substantially among older adults. Error bars indicate 95% CIs. Estimates for individuals aged ≥100 years are not displayed due to sparse counts and wide uncertainty; results for this age group are shown in Table 2.

Regional Trends and Associations With Comorbidities

Increases were observed nationwide and across sexes (Supplementary Figs. S1, S2). However, increases in AMD prevalence were not significantly associated with the proportion of adults aged ≥65 years or with changes in that proportion by prefecture (correlation with proportion: r = −0.06, P = 0.705; correlation with increase: r = −0.25, P = 0.0887; Supplementary Fig. S3).

Baseline characteristics of patients with and without nAMD in the 2014 and 2019 case-control datasets are summarized in Supplementary Table S11. In a univariable model including calendar year only, the odds of nAMD were higher in 2019 than in 2014 (OR = 1.55; 95% CI, 1.53–1.56). After adjustment for age and sex (Model 1, Table 3), the calendar-year OR decreased modestly to 1.45 (95% CI, 1.44–1.47). Further adjustment for diabetes, hypertension, and dyslipidemia (Model 2) did not materially change the calendar-year OR (OR = 1.45; 95% CI, 1.43–1.46), indicating negligible attenuation by these comorbidities coded in claims. In Model 2, diabetes, dyslipidemia, and hypertension were each independently associated with higher odds of nAMD.

Table 3.

Multivariable Logistic Regression for Factors Associated With Neovascular AMD

	Model 1 (Without Comorbidities)		Model 2 (With Comorbidities Added)
Variable	OR (95% CI)	P Value	OR (95% CI)	P Value
Calendar year (2019 vs 2014, Reference: 2014)	1.45 (1.44–1.47)	<0.001	1.45 (1.43–1.46)	<0.001
Age (yr)	1.07 (1.07–1.07)	<0.001	1.07 (1.07–1.07)	<0.001
Sex (Reference: Male)	0.37 (0.37–0.38)	<0.001	0.38 (0.38–0.39)	<0.001
Presence of diabetes mellitus	—	—	1.24 (1.22–1.25)	<0.001
Presence of hypertension	—	—	1.07 (1.06–1.08)	<0.001
Presence of dyslipidemia	—	—	1.10 (1.09–1.12)	<0.001

Model 1: Logistic regression model adjusted for age and gender only; Model 2: Logistic regression model fully adjusted for all relevant covariates, including age, sex, and comorbidities.

Changes in Injection Patterns for nAMD

The total number of treatment claims for nAMD increased markedly from July–December 2014 to the same period in 2019 (Fig. 3A). Aflibercept use more than doubled, Ranibizumab remained stable, Pegaptanib declined, and Visudyne (PDT) increased slightly. Pegaptanib use was rare in both years. In Japan, pegaptanib remained commercially available and in clinical use until 2020,²⁹ which may explain the small number of claims observed in 2019.

Figure 3.

Trends in the use of intravitreal injections for nAMD in Japan (Jul–Dec 2014 vs. Jul–Dec 2019). (A) Total number of treatment claims during the six-month period (July–December) in 2014 and 2019, stratified by agent: intravitreal anti-VEGF injections (aflibercept, ranibizumab, pegaptanib) and verteporfin photodynamic therapy (Visudyne; PDT). The inset enlarges the scale for pegaptanib and PDT. (B) Number of treated patients receiving each agent at least once during the same six-month period (2014 vs. 2019). The inset enlarges the scale for pegaptanib and PDT. (C) Mean number of treatment claims per patient among users of each agent (patients with ≥1 claim for that agent) during the six-month period. (D) Age-specific mean number of intravitreal anti-VEGF injection claims per treated patient during the six-month period, comparing 2014 and 2019 across five-year age bands.

The number of treated patients receiving aflibercept at least once also more than doubled, whereas the number of ranibizumab users decreased slightly between the years (Fig. 3B). Pegaptanib users were rare and further declined, and PDT use increased but remained confined to a small subset of treated patients.

When anti-VEGF treatment frequency was evaluated among users of each agent, mean (SD) administrations increased slightly (Fig. 3C): aflibercept, from 2.04 (1.09) in 2014 to 2.10 (1.15) in 2019; ranibizumab, from 1.81 (1.02) to 1.90 (1.08); and pegaptanib, from 2.03 (1.15) to 2.18 (1.36). PDT showed no meaningful change in treatment frequency. In age-stratified analyses using total anti-VEGF injections per treated patient (all agents combined), mean injection frequency during the six-month window was broadly similar across age groups and slightly higher in older age strata (Fig. 3D). These estimates are based on a six-month cross-sectional assessment period that includes both incident and prevalent treated cases and thus reflects a mixture of induction- and maintenance-phase treatment.

Discussion

In this study, we used data from NDB, which covers more than 95% of the population, to provide a comprehensive overview of nAMD epidemiology. The nationwide study revealed a marked increase in the prevalence of nAMD between 2014 and 2019, a period during which population aging further progressed.²⁶ Our estimates extend prior NDB-based work on treatment-requiring nAMD.⁵ Our treated prevalence in 2014 was approximately 0.10%, and we further demonstrate that treated nAMD prevalence increased substantially by 2019 under the same treated case definition. Moreover, the marked increase in total treatment volume and the dominance of aflibercept in 2019 are in line with previously reported real-world shifts in treatment practice in Japan. Our findings also extend prior reports of increasing nAMD prevalence in rapidly aging societies and underscores that this trend is apparent in both sexes and across all age groups, particularly among older individuals. The results indicate that the overall upward trend in nAMD prevalence is not solely attributable to population aging. The lack of a positive correlation between the increase in the aging rate and the increase in nAMD prevalence across prefectures further supports the notion that population aging was not the primary driver of the increase in nAMD prevalence in Japan from 2014 to 2019.

In a previous report, we estimated the prevalence of active nAMD—defined as a diagnosis of nAMD followed by anti-VEGF therapy—as approximately 100 per 100,000 population in 2015.⁵ Although prior studies have examined the incidence or prevalence of nAMD,^{5,8–12,30–41} the present analysis is the first to compare its prevalence over time using the same definition in a nationwide population.⁵ From a health economics and policymaking perspective, precisely determining the number of patients who require these high-cost anti-VEGF injections is of paramount importance. Our current results show that, before the impact of the COVID-19 pandemic, the prevalence had further increased by 2019.

Prefecture-level differences in population aging were unlikely to explain the heterogeneity in treated nAMD trends: the change in treated nAMD prevalence was not significantly correlated with either the level or the change in the proportion of adults aged ≥65 years across prefectures. These findings support the interpretation that factors beyond demographics are likely contributing to the observed temporal and regional variation. In addition, in our multivariable analyses, diabetes, hypertension, and dyslipidemia were each independently associated with higher odds of nAMD, supporting the concept that nAMD is associated with cardiometabolic and cardiovascular diseases. These findings are consistent with earlier reports suggesting that systemic disorders such as dyslipidemia and diabetes may exacerbate the development or progression of nAMD.^42–45 However, adding these comorbidities to the regression model did not materially attenuate the higher odds of nAMD in 2019 versus 2014, suggesting that the secular rise in nAMD prevalence is unlikely to be explained by changes in comorbidity burden alone. Given the claims-based binary measurement of comorbidities and the absence of several important risk factors, residual confounding remains possible. Other factors may also have contributed to the secular increase in treated nAMD. Wider optical coherence tomography adoption may have improved detection and earlier identification of treatment-requiring disease, and changes in referral patterns and treatment thresholds during the dissemination of optical coherence tomography and anti-VEGF therapy may have increased treated case capture.^5,46 In addition, improved survival related to declining cardiovascular mortality may have expanded the pool of older adults at risk.⁴⁷ These mechanisms are plausible but cannot be disentangled directly within the current claims-based design.

The rising treatment burden indicates increasing demand for resource-intensive nAMD care. The shift toward greater aflibercept use, together with modest increases in administrations among users, suggests that changes in real-world management may be amplifying the health-care system burden in addition to changes in the underlying disease burden. The slightly higher injection frequency observed particularly in older age strata may reflect a greater burden of treated nAMD in older adults.

Our study had some limitations. First, our outcome represents claims-based treated nAMD and may under-ascertain individuals with AMD who are undiagnosed, untreated, or not captured by our treatment-based algorithm. Undiagnosed or untreated AMD may be non-trivial and could vary by age, region, healthcare access, and socioeconomic factors. Second, temporal changes in diagnostic sensitivity, referral patterns, and treatment thresholds or strategies may influence treated prevalence over time. Therefore observed increases may reflect a combination of changes in underlying disease burden and changes in detection and treatment practices. Third, because imaging and clinical chart data are unavailable in the NDB, residual misclassification with other causes of macular neovascularization cannot be completely excluded, particularly among younger individuals. To mitigate this, we used NDB diagnostic codes and additionally required evidence of nAMD-related treatment. In our prior validation study of this operational definition, the specificity was 98.3%,²⁵ suggesting that inclusion of other diseases is unlikely to be substantial. Fourth, although we used NDB diagnostic subtype codes to classify AMD (Supplementary Table S1), the validity of claims-based AMD subtyping has not been established. Therefore we cannot provide reliable subtype-specific estimates of which AMD diagnostic subtypes received PDT.

Conclusions

This study provides robust evidence that the burden of nAMD has significantly increased in Japan between 2014 and 2019. Future research should focus on longitudinal follow-up of patients who are newly diagnosed with AMD and investigate the interplay between systemic risk factors, ocular treatment modalities, and disease progression.