Abstract
Purpose
To investigate whether primary open‐angle glaucoma (POAG) is associated with an increased long‐term risk of developing Alzheimer's disease, given its shared neurodegenerative features.
Methods
A 20‐year longitudinal, registry‐based matched cohort study was conducted using Danish national health registries from 1998 to 2018. Individuals aged 65 years or older with POAG identified by registration with the diagnostic code (ICD‐10 = H40.1*) or at least four redeemed glaucoma prescriptions (ATC = S01E*) within 1 year entered the cohort. Each individual with POAG was randomly matched with five controls of the same sex and birth year. The outcome of Alzheimer's disease was defined by registration with the diagnostic code (ICD‐10 = G30*, F00*). A Cox regression model adjusting for age, sex and systemic comorbidities estimated the hazard ratio (HR) for Alzheimer's disease and a Fine–Grey competing‐risk model accounted for death as a competing event.
Results
The study included 61 829 individuals with POAG and 306 794 matched controls (42.63% males, 57.37% females; median age 75.22 years, IQR: 70.35–80.63). Alzheimer's disease developed in 1.61% of individuals with POAG and 1.59% of controls. POAG did not appear to increase the risk of Alzheimer's disease (adjusted HR 0.98, 95% CI: 0.93–1.03). Competing risk analyses found a slightly increased risk of Alzheimer's disease observed among men with POAG (sHR 1.12, 95% CI: 1.03–1.21), whereas no association was found among women (sHR 1.00, 95% CI: 0.94–1.07).
Conclusion
In this nationwide matched cohort study, POAG was not clearly associated with an overall increased risk of developing Alzheimer's disease over 20 years.
Keywords: Alzheimer's disease, epidemiology, glaucoma, matched cohort study, primary open‐angle glaucoma and registry‐based study
1. INTRODUCTION
Glaucoma is a group of chronic and progressive optic nerve disorders characterized by degeneration of retinal ganglion cells (Jayaram et al., 2023). It is the leading cause of irreversible blindness worldwide, (Kang & Tanna, 2021) affecting approximately 3.5% of individuals aged 40–80 years (Tham et al., 2014), a figure that gains significance in the context of a growing elderly population. Among the various forms of glaucoma, primary open‐angle glaucoma (POAG) is the most prevalent (Quigley & Broman, 2006), characterized by open iridocorneal angles and slow, insidious progression of optic nerve damage and visual field loss (Weinreb & Khaw, 2004).
Although glaucoma has traditionally been considered a primary disease of the eye, recent research has unveiled a potential association between glaucoma and the development of cerebral neurodegenerative conditions, such as Alzheimer's disease.
Alzheimer's disease is a severe neurological disorder and a major cause of dementia, contributing to 60–80% of all dementia cases worldwide (Alzheimer's Association, 2024). It is characterized by extracellular deposits of beta‐amyloid (Aβ) and intraneuronal accumulations of hyperphosphorylated tau protein (pTau) (Scheltens et al., 2021). Similar extracellular deposits of Aβ have been observed in retinal layers of transgenic mouse models of Alzheimer's disease, suggesting a potential neurodegenerative overlap between the retina and the brain (Chiu et al., 2012). Although these findings remain inconclusive in humans, similar mechanisms are hypothesized to occur.
A systematic meta‐analysis (Xu et al., 2019) reported an overall increased risk of Alzheimer's disease among individuals with glaucoma. However, the analysis was based on a limited number of heterogeneous observational studies, and the positive association was mainly driven by case–control data rather than cohort studies, which found no significant relationship.
In a more recent meta‐analysis consisting of 11 cohort studies and a total of 198 298 individuals with glaucoma, Zhao et al. (2021) failed to find evidence supporting glaucoma as an independent risk factor for Alzheimer's disease.
The current study aims to clarify whether POAG increases the long‐term risk of developing Alzheimer's disease in a Danish national cohort. By addressing existing gaps in the literature and leveraging robust population‐based data, this research seeks to contribute to a deeper understanding of the potential links between these two conditions, with implications for both clinical practice and future research.
2. MATERIALS AND METHODS
2.1. Study design
This is a nationwide, registry‐based matched cohort study with up to 20 years of follow‐up using Danish health registries.
2.2. Data sources
The Danish National Patient Registry (DNPR) has systematically recorded detailed administrative and clinical data on all patients discharged from Danish non‐psychiatric hospitals since 1977. From 1995 and onwards, the DNPR expanded to include data on psychiatric inpatients, emergency department visits and outpatient specialty clinic encounters. Each DNPR patient contact is documented with International Classification of Diseases (ICD)‐coded diagnoses. The DNPR used the ICD‐8 classification system until it was replaced by the ICD‐10 system in 1994 (Schmidt et al., 2015).
The Danish National Prescription Registry (DNPrR) has recorded comprehensive information on all prescription drugs dispensed at Danish community pharmacies since 1995. Medications are classified according to the anatomical therapeutic chemical (ATC) classification system (Pottegård et al., 2017).
The Danish Civil Registration System (CRS) has maintained administrative records of all individuals residing in Denmark since 1968. This registry contains unique personal identification numbers, sex, dates of birth and continually updated information on vital status (Mainz et al., 2019).
The unique personal identification number assigned to all Danish residents enables linkage between the registries and other data sources, facilitating comprehensive and reliable data analysis.
2.3. Study population
Individuals aged 65 years or older on the index date with POAG identified from Danish national registries between 1998 and 2018 were included in a cohort. This age group was chosen as late‐onset Alzheimer's disease, defined as Alzheimer's disease with onset at age 65 or older, represents the majority of cases, unlike early‐onset Alzheimer's disease, which is rare and often hereditary (Bertram & Tanzi, 2005).
Cases with POAG were defined by one of two criteria: (1) Diagnostic code criterion is defined by registration in the DNPR with the POAG diagnosis (ICD‐10 = H40.1, including subcodes = *) as the most recent glaucoma or ocular hypertension diagnosis (ICD‐10 = H40*) from an ophthalmology department; (2) Medical treatment criterion defined by registration in DNPrR with medical treatment of glaucoma (ATC = S01E*), with at least four unique dates of redeemed prescriptions within a one‐year period. Furthermore, patients within the medical treatment criterion are excluded with registration in the DNPR of primary angle‐closure glaucoma (ICD‐10 = H40.2*), secondary forms of glaucoma (ICD‐10 = H40.3, H40.4, H40.5*, H40.6) or unspecified glaucoma (ICD‐10 = H40.8) recorded as the most recent glaucoma or ocular hypertension diagnosis by an ophthalmology department.
The index date was defined as the occurrence of the first qualifying event. The event of diagnostic code was defined as the date of the first contact at an ophthalmology department which led to registration in the DNPR with the POAG diagnosis. The event of medical treatment was defined as the first registration in DNPrR with medical treatment of glaucoma among the four required prescriptions within a year.
Individuals registered in the DNPR with any glaucoma or ocular hypertension diagnosis (ICD‐8 = 36 402, 37 500–37 599, 74 420; ICD‐10 = H40*) from an ophthalmology department or registered in the DNPrR with any medical treatment of glaucoma before the year of 1998 was excluded.
Furthermore, individuals registered in the DNPR with an Alzheimer's disease diagnosis (ICD‐8 = 29 010; ICD‐10 = G30*, F00*), Parkinson's disease (PD) diagnosis (ICD‐8 = 34 299; ICD‐10 = G20*), dementia (ICD‐8 = 29 009, 29 018, 29 019; ICD‐10 = F01*, F02*) or other cerebral neurodegenerative diagnosis (ICD‐8 = 33 109; ICD‐10 = G318E) from a neurological, geriatric or psychiatric department before or at the index date were excluded.
2.4. Matched controls
Controls were randomly selected from the CRS and matched to cases based on sex and birth year, with five controls assigned per case. Controls formed a matched comparison cohort and were followed from the same index date as their matched POAG individual.
Controls were not eligible if they were registered in DNPR with any glaucoma or ocular hypertension diagnosis from an ophthalmology department, registered in the DNPrR with any medical treatment of glaucoma or registered in the DNPR with an Alzheimer's disease diagnosis, PD diagnosis or other cerebral neurodegenerative diagnosis from a neurological, geriatric or psychiatric department before or at the index date of their matched case.
A control could be matched to more than one POAG individual, provided they had not been diagnosed with Alzheimer's disease before the index date of any matched POAG individual.
2.5. Outcome
The outcome of Alzheimer's disease was defined by registration in the DNPR with an Alzheimer's disease diagnosis (ICD‐10 = G30*, F00*) from a neurological, geriatric or psychiatric department.
2.6. Covariates
Covariates included sex, age as a continuous variable and time‐varying comorbidity.
Comorbidity was assessed using the Charlson comorbidity index (CCI) score and calculated from ICD codes (Quan et al., 2011). Since Alzheimer's disease is characterized by dementia included in the CCI, the full index score could not be incorporated into our analysis. Instead, the multivariable model was adjusted for the remaining comorbid conditions listed in the C CI to ensure a comprehensive assessment of potential confounders.
2.7. Statistical analysis
Baseline characteristics were summarized using descriptive statistics.
The incidence rate (IR) of Alzheimer's disease was calculated for both the individuals with POAG and their matched controls. IRs were calculated by dividing the number of individuals diagnosed with Alzheimer's disease by the total person‐time at risk and expressed per 1000 person‐years. Person‐time was accrued for each individual starting from the index date and continued until the first occurrence of one of the following censoring events: (i) a diagnosis of Alzheimer's disease, (ii) emigration, disappearance or death as recorded in the Danish Civil Registration System or (iii) end of follow‐up on 31 December 2018.
Hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) were estimated using Cox proportional hazards regression to assess the relative risk of developing Alzheimer's disease among individuals with POAG compared to their matched control cohort. Cox regression was selected for its ability to model time‐to‐event data while accounting for right‐censored observations.
Two models were fitted: (1) a crude model including only POAG status and (2) an adjusted model controlling for sex, age and comorbidities from CCI.
Potential interactions were tested to determine whether they significantly influenced the outcome. Specifically, we evaluated interactions between cohort and sex, and between cohort and age, where cohort refers to individuals with POAG and their matched comparison group. These analyses assessed whether the association between POAG status and the risk of Alzheimer's disease differed by sex or age. Interaction terms were incorporated into the Cox proportional hazards models and likelihood‐ratio tests (LRT) were used to assess their statistical significance. Backward selection was performed starting with the most complex model. Significant interaction terms (p < 0.05) were included in the final model to account for effect modification, ensuring that the differential effects of cohort across sex and age groups were adequately captured.
Additionally, the study considered death as a competing risk, which is particularly relevant in populations with high mortality rates or chronic comorbid conditions. To address this, the Fine–Gray sub‐distribution hazards model was used to account for death as a competing risk for Alzheimer's disease.
All statistical analyses were carried out using Stata version 18.5 (StataCorp LLC, College Station, TX, USA).
2.8. Ethical considerations
This study is a subproject within the Danish Excellence Centre in Ophthalmic Epidemiology (DECODE‐EYE) initiative, conducted at the Research Unit of Ophthalmology, Department of Clinical Research, University of Southern Denmark (Grauslund et al., 2020).
According to Danish regulations, registry‐based studies are exempt from the requirement for informed consent and ethical approval from the Danish National Committee on Health Research Ethics (The Danish National Center for Ethics, 2024). Approval for data processing was obtained from the Region of Southern Denmark's record of data processing activities (Journal no. 19/15996) and access to registry data was granted by the Danish Health Authorities (FSEID‐00004302).
This study was conducted in accordance with the principles outlined in the Declaration of Helsinki (World Medical Association, 2013). Personal information was handled in strict compliance with the General Data Protection Regulation (GDPR) and the Danish Health Care Act (The Danish Parliament, 2010, 2018).
3. RESULTS
At the index date, the cohort comprised 61 829 individuals with POAG and 306 794 matched controls, resulting in an average of 4.96 per case. The study population included 157 161 (42.63%) males and 211 462 (57.37%) females, with a median age of 75.22 years (IQR: 70.35–80.63). Compared to controls, individuals with POAG had a slightly lower burden of systemic comorbidity (Table 1).
TABLE 1.
Baseline characteristics of the study population, individuals with POAG and controls at index date.
| All (n = 368 623) | Individuals with POAG (n = 61 829) | Controls (n = 306 794) | |
|---|---|---|---|
| Sex, n (%) | |||
| Male | 157 161 (42.63) | 26 382 (42.67) | 130 779 (42.63) |
| Female | 211 462 (57.37) | 35 447 (57.33) | 176 015 (57.37) |
| Age, years Q (IQR) | 75.22 (70.35; 80.63) | 75.19 (70.28; 80.61) | 75.22 (70.36; 80.64) |
| CCI, n (%) | |||
| 0 = low | 279 434 (75.80) | 47 948 (77.55) | 231 486 (75.45) |
| 1 = moderately low | 30 573 (8.29) | 5352 (8.66) | 25 221 (8.22) |
| 2 = moderately high | 41 528 (11.27) | 6295 (10.18) | 35 233 (11.48) |
| 3 ≥ high | 17 088 (4.64) | 2234 (3.61) | 14 854 (4.84) |
Note: Data are presented as numbers and percentages.
Abbreviations: CCI, Charlson comorbidity index; IQR, interquartile range; POAG, primary open‐angle glaucoma; Q, median.
During follow‐up, 1919 individuals with POAG (3.10%) and 9097 matched controls (2.97%) were diagnosed with Alzheimer's disease (Figure 1). The annual IRs of Alzheimer's disease were similar between cohorts, with an annual risk of 4.32 per 1.000 person‐years among cases and 4.30 per 1.000 person‐years among controls (IR rate 1.00, 95% CI: 0.96–1.06). Cox proportional hazards regression analyses further supported these observations (HR 1.00, 95% CI: 0.96–1.05). After adjusting for potential confounders, the association remained statistically non‐significant (HR 0.98, 95% CI: 0.93–1.03), suggesting no clear evidence of an increased hazard of Alzheimer's disease among individuals with POAG (Table 2).
FIGURE 1.
Flowchart of outcomes in individuals with primary open‐angle glaucoma and controls. †Registration in the DNPR with primary open‐angle glaucoma (ICD‐10 = H40.1*) as the most recent glaucoma or ocular hypertension diagnosis (ICD‐10 = H40*), from an ophthalmology department. ‡Registration in DNPrR for medical treatment of glaucoma (ATC = S01E*), with at least four unique dates of redeemed prescriptions within a one‐year period, without registration in the DNPR with primary angle‐closure glaucoma, secondary forms of glaucoma or other specified glaucoma (ICD‐10 = H40.2*, H40.3, H40.4, H40.5*, H40.6, H40.8) recorded as the most recent glaucoma or ocular hypertension diagnosis by an ophthalmology department. §Registration in DNPR with an Alzheimer's disease diagnosis code (ICD‐10 = G30*, F00*), from a neurological, geriatric or psychiatric department.
TABLE 2.
Number of events, incident rates and hazard ratios.
| Individuals with POAG | Controls | Crude HR (95% CI) | Adjusted b HR (95% CI) | |||
|---|---|---|---|---|---|---|
| AD/PYR | IR a (95% CI) | AD/PYR | IR a (95% CI) | |||
| Total | 1919/444 085.34 | 4.32 (4.13–4.52) | 9097/2 113 415.25 | 4.30 (4.22–4.39) | 1.00 (0.96–1.05) | 0.98 (0.93–1.03) |
Abbreviations: AD, Alzheimer's disease; CI, confidence interval; HR, hazard ratio; IR, incidence rate; POAG, primary open‐angle glaucoma; PYR, person‐years at risk.
Per 1.000 person‐years.
Model adjusted for age, sex, systemic comorbidity using modified Charlson comorbidity index (CCI) score.
Likelihood ratio tests showed that removing the interaction between cohort and sex resulted in a significantly poorer fit (p = 0.03), as did the removal of the interaction between cohort and age (p = 0.01). These results indicate that both sex and age modified the association between POAG and Alzheimer's disease, and interaction terms were therefore retained in the final model.
The Cox proportional hazards regression analysis with the interaction terms showed no statistically significant difference in the hazard of the outcome between individuals with POAG and the matched controls in either sex or age (men: HR 1.08, 95% CI: 0.99–1.17; women: HR 0.97, 95% CI: 0.90–1.03; age: HR 0.99, 95% CI: 0.98–1.17).
In the competing risk analysis using the Fine–Grey model, which accounted for death as a competing event, a slightly increased risk of Alzheimer's disease was observed among men with POAG compared to matched controls (sHR 1.12, 95% CI: 1.03–1.21). In contrast, no association was found among women with POAG and their matched controls (sHR 1.00, 95% CI: 0.94–1.07). Furthermore, there was no statistically significant difference in the hazard of the outcome between individuals with POAG and matched controls across age (sHR 1.00, 95% CI: 0.99–1.00; Table 3).
TABLE 3.
Number of events, incident rates and subdistribution hazard ratios from competing risk analysis.
| Individuals with POAG | Controls | Adjusted c sHR (95% CI) | |||
|---|---|---|---|---|---|
| Deaths a /PYR | IR b (95% CI) | Deaths a /PYR | IR b (95% CI) | ||
| Total | 26 233/444 085.34 | 59.07 (58.36–59.79) | 136 829/2 113 415.25 | 64.74 (64.40–65.09) | – |
| Male | 11 594/178 062.70 | 65.11 (63.94–66.31) | 60 195/852 292.25 | 70.63 (70.07–71.19) | 1.12 (1.03–1.21) |
| Female | 14 639/266 022.64 | 55.03 (54.14–55.93) | 76 634/1 261 123.00 | 60.34 (60.34–61.21) | 1.00 (0.94–1.07) |
| Age | – | – | 1.00 (0.99–1.00) | ||
Abbreviations: CI, confidence interval; POAG, primary open‐angle glaucoma; sHR, subdistribution hazard ratio.
Deaths prior to the onset of Alzheimer's disease.
Per 1.000 person‐years.
Competing risk model adjusted for age, sex, and systemic comorbidity using modified Charlson Comorbidity Index (CCI) score.
4. DISCUSSION
This study investigated whether POAG is associated with an increased long‐term risk of developing Alzheimer's disease. In this large Danish population‐based matched cohort of adults aged 65 years and older, no clear overall association was observed between POAG and Alzheimer's disease over a 20‐year follow‐up. However, when accounting for death as a competing risk, a small but statistically significant increased risk of Alzheimer's disease was observed among men with POAG. No such association was seen among women nor across age.
Despite the lack of a strong association in this study, a biological link between POAG and Alzheimer's disease remains plausible, as both conditions involve neurodegeneration and share overlapping pathological features, such as abnormal aggregation of Aβ and pTau (Ashok et al., 2020; Chiu et al., 2012). Given that the retina is considered an extension of the central nervous system, it is possible that retinal changes in glaucoma reflect neurodegenerative processes similar to those seen in Alzheimer's disease.
This is among the largest cohort studies published examining the association between POAG and Alzheimer's disease, encompassing 61 829 individuals with POAG and 306 794 matched controls. The large scale of this study allows for robust statistical analyses, providing a comprehensive view of the relationship between these two conditions within a population‐based setting.
The overall findings of the present study are consistent with those of Keenan et al. (2015), who reported no significant association between POAG and subsequent Alzheimer's disease in a large population‐based cohort study conducted in the United Kingdom (rate ratio 1.01, 95% CI: 0.96–1.06). While their use of rate ratios limits direct comparison with the hazard ratios in the present study, the alignment in findings suggests that POAG does not appear to be a strong independent risk factor for Alzheimer's disease. Keenan et al.'s study included a slightly larger POAG cohort (n = 87 658) but relied exclusively on hospital admission and day‐case data, potentially missing cases managed in outpatient settings and thereby underestimating the true burden of POAG. Notably, the present study did observe a small increased risk of Alzheimer's disease among men with POAG, a potential sex‐specific effect that was not explored in the Keenan et al. study, suggesting that sex‐specific pathways may play a role and warrant further investigation.
In contrast to the findings of both the present study and Keenan et al., a recent Swedish cohort study by Crump et al. (2024) reported a modest statistically significant association between POAG and an increased risk of Alzheimer's disease (adjusted HR 1.31, 95% CI: 1.27–1.36). Using comprehensive Swedish national registries, the study included a large sample of 324 730 individuals with glaucoma and followed them over an extensive 22‐year period (1995–2017), offering valuable long‐term insights into this potential relationship. While the authors conducted sub‐analyses by glaucoma subtype, the exact number of individuals with POAG was not specified, which presents some limitations for direct comparisons. The present study and the Swedish study by Crump et al. share similarities, as they were using high‐quality national health registries conducted on Scandinavian populations over an overlapping period of approximately 20 years. Despite these similarities, the present study did not find evidence of an overall increased risk of Alzheimer's disease among individuals with POAG. This discrepancy may reflect differences in study design. Notably, the Swedish study included a broader range of covariates, including various socioeconomic factors. Although Crump et al. adjusted for a broader set of covariates, their exclusive reporting of adjusted hazard ratios limits direct comparability with the present study, given the differences in covariate selection. Another methodological distinction is that the Swedish study did not appear to account for death as a competing risk, despite the relatively high median age (74 years) at glaucoma diagnosis in their cohort. In high‐mortality populations, failing to account for competing risks could overestimate the incidence of the primary event, as death would preclude its occurrence.
However, another recent Swedish study by Ekström et al. (2021) offered a complementary perspective. This study followed more than 1500 individuals with POAG for up to 35 years, requiring a visual field defect for a definite POAG diagnosis. Importantly, competing events such as death were accounted for. Although considerably smaller than registry‐based studies, the rigorous clinical ascertainment enhances internal validity and shows how methodological choices—clinical verification versus registry‐based coding—can influence interpretation.
These findings highlight the complexity of studying the relationship between POAG and Alzheimer's disease across different populations and study designs. These differences underscore both the challenges and the value of using nationwide registry data to explore the long‐term relationship between POAG and Alzheimer's disease.
The strengths of this study include its large cohort, comprising 61 829 individuals with POAG and 306 786 matched controls, drawn from comprehensive national registry data. The use of multiple data sources, including the Danish National Patient Registry, Danish National Prescription Registry and the Civil Registration System, ensures completeness and accuracy in identifying individuals with POAG and Alzheimer's disease and provides robust, population‐based data. Additionally, the long follow‐up period from 1998 to 2018 enhances the reliability of the findings by allowing for an investigation into the long‐term association between POAG and the risk of Alzheimer's disease. The data linkage also facilitated adjustment for important confounders, such as comorbidities, age and sex, which strengthened the validity of the analyses. Moreover, the inclusion of a competing risk analysis accounted for the high mortality rates in this elderly population, ensuring that the associations between POAG and Alzheimer's disease were not overestimated due to deaths occurring before the development of Alzheimer's disease.
However, this study is not without limitations. The use of registry‐based data may involve misclassification, particularly in distinguishing between POAG and ocular hypertension as individuals with POAG were primarily identified from redeemed prescriptions. Such non‐differential misclassification would most likely dilute rather than exaggerate a true association with Alzheimer's disease. Although the study adjusted for several potential confounders, other unmeasured factors, such as lifestyle factors, genetic predispositions and socio‐economic conditions, could influence the observed associations and contribute to differences seen across countries. Moreover, in large register‐based studies, even small increases in risk may reach statistical significance due to the high statistical power provided by very large sample sizes. In this context, the observed 12% increased risk of Alzheimer's disease among men should be interpreted with caution. Although the association remained statistically significant after accounting for the competing risk of death, its magnitude is modest and unlikely to represent a clinically meaningful difference at the individual level. Because competing risk analyses tend to attenuate associations by considering mortality as an alternative outcome, the result likely reflects a subtle rather than a strong causal relationship. Thus, while statistically robust, the finding should primarily be viewed as an indicator of a potential link rather than a clinically important risk elevation. Nevertheless, the observed sex‐specific difference may point to underlying biological mechanisms that differ between men and women. Given the shared neurodegenerative features of POAG and Alzheimer's disease, further research is warranted to explore this possible link more deeply.
FUNDING INFORMATION
This work was supported by the P. Carl Petersen Foundation (grant number 2410) and the Synoptik Foundation (grant number 24030032).
ACKNOWLEDGEMENTS
The authors have no acknowledgements to report.
REFERENCES
- Alzheimer's Association . (2024) 2024 Alzheimer's disease facts and figures. 20 . [DOI] [PMC free article] [PubMed]
- Ashok, A. , Singh, N. , Chaudhary, S. , Bellamkonda, V. , Kritikos, A.E. , Wise, A.S. et al. (2020) Retinal degeneration and Alzheimer's disease: an evolving link. International Journal of Molecular Sciences, 21, 7290. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bertram, L. & Tanzi, R.E. (2005) The genetic epidemiology of neurodegenerative disease. The Journal of Clinical Investigation, 115, 1449–1457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chiu, K. , Chan, T.F. , Wu, A. , Leung, I.Y. , So, K.F. & Chang, R.C. (2012) Neurodegeneration of the retina in mouse models of Alzheimer's disease: what can we learn from the retina? Age (Dordrecht, Netherlands), 34, 633–649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Crump, C. , Sundquist, J. , Sieh, W. & Sundquist, K. (2024) Risk of Alzheimer's disease and related dementias in persons with glaucoma: a National Cohort Study. Ophthalmology, 131, 302–309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ekström, C. , Puhto, I. & Kilander, L. (2021) Association between open‐angle glaucoma and Alzheimer's disease in Sweden: a long‐term population‐based follow‐up study. Upsala Journal of Medical Sciences, 126, e7819. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Grauslund, J. , Stokholm, L. , Kyvik, K.O. , Dornonville de la Cour, M. , Kessel, L. & Rubin, K.H. (2020) Interactions between ocular and systemic disease using national register‐based data in the Danish excellence Centre in Ophthalmic Epidemiology (DECODE‐EYE): study perspective. Acta Ophthalmologica, 98, 573–578. [DOI] [PubMed] [Google Scholar]
- Jayaram, H. , Kolko, M. , Friedman, D.S. & Gazzard, G. (2023) Glaucoma: now and beyond. Lancet, 402, 1788–1801. [DOI] [PubMed] [Google Scholar]
- Kang, J.M. & Tanna, A.P. (2021) Glaucoma. The Medical Clinics of North America, 105, 493–510. [DOI] [PubMed] [Google Scholar]
- Keenan, T.D. , Goldacre, R. & Goldacre, M.J. (2015) Associations between primary open angle glaucoma, Alzheimer's disease and vascular dementia: record linkage study. British Journal of Ophthalmology, 99, 524–527. [DOI] [PubMed] [Google Scholar]
- Mainz, J. , Hess, M.H. & Johnsen, S.P. (2019) The Danish unique personal identifier and the Danish civil registration system as a tool for research and quality improvement. International Journal for Quality in Health Care, 31, 717–720. [DOI] [PubMed] [Google Scholar]
- Pottegård, A. , Schmidt, S.A.J. , Wallach‐Kildemoes, H. , Sørensen, H.T. , Hallas, J. & Schmidt, M. (2017) Data resource profile: the Danish National Prescription Registry. International Journal of Epidemiology, 46, 798–798f. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Quan, H. , Li, B. , Couris, C.M. , Fushimi, K. , Graham, P. , Hider, P. et al. (2011) Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. American Journal of Epidemiology, 173, 676–682. [DOI] [PubMed] [Google Scholar]
- Quigley, H.A. & Broman, A.T. (2006) The number of people with glaucoma worldwide in 2010 and 2020. The British Journal of Ophthalmology, 90, 262–267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Scheltens, P. , De Strooper, B. , Kivipelto, M. , Holstege, H. , Chételat, G. , Teunissen, C.E. et al. (2021) Alzheimer's disease. Lancet, 397, 1577–1590. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schmidt, M. , Schmidt, S.A. , Sandegaard, J.L. , Ehrenstein, V. , Pedersen, L. & Sørensen, H.T. (2015) The Danish National Patient Registry: a review of content, data quality, and research potential. Clinical Epidemiology, 7, 449–490. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tham, Y.C. , Li, X. , Wong, T.Y. , Quigley, H.A. , Aung, T. & Cheng, C.Y. (2014) Global prevalence of glaucoma and projections of glaucoma burden through 2040: a systematic review and meta‐analysis. Ophthalmology, 121, 2081–2090. [DOI] [PubMed] [Google Scholar]
- The Danish National Center for Ethics . (2024) Registry Research Projects without the Use of Biological Material .
- The Danish Parliament . (2010) Danish Health Care Act .
- The Danish Parliament . (2018) Danish Data Protection Act .
- Weinreb, R.N. & Khaw, P.T. (2004) Primary open‐angle glaucoma. Lancet, 363, 1711–1720. [DOI] [PubMed] [Google Scholar]
- World Medical Association . (2013) World medical association declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA, 310, 2191–2194. [DOI] [PubMed] [Google Scholar]
- Xu, L. , Jiang, S. , Bi, Y. , Zhang, Y. , Du, Y. & Jia, J. (2019) Association between glaucoma and the risk of Alzheimer's disease: a systematic review of observational studies. Acta Ophthalmologica, 97, e480–e489. [DOI] [PubMed] [Google Scholar]
- Zhao, W. , Lv, X. , Wu, G. , Zhou, X. , Tian, H. , Qu, X. et al. (2021) Glaucoma is not associated with Alzheimer's disease or dementia: a meta‐analysis of cohort studies. Frontiers in Medicine, 8, 669975. [DOI] [PMC free article] [PubMed] [Google Scholar]