Free
Clinical and Epidemiologic Research  |   March 2015
Diabetes, Cardiovascular Morbidity, and Risk of Age-Related Macular Degeneration in a Primary Care Population
Author Affiliations & Notes
  • Zdravko P. Vassilev
    Bayer Healthcare Pharmaceuticals, Whippany, New Jersey, United States
  • Ana Ruigómez
    Spanish Centre for Pharmacoepidemiologic Research (CEIFE), Madrid, Spain
  • Montse Soriano-Gabarró
    Bayer Pharma AG, Berlin, Germany
  • Luis A. García Rodríguez
    Spanish Centre for Pharmacoepidemiologic Research (CEIFE), Madrid, Spain
  • Correspondence: Zdravko P. Vassilev, Bayer Healthcare Pharmaceuticals, 100 Bayer Boulevard, Whippany, NJ 07981, USA; [email protected]
Investigative Ophthalmology & Visual Science March 2015, Vol.56, 1585-1592. doi:https://doi.org/10.1167/iovs.14-16271
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Zdravko P. Vassilev, Ana Ruigómez, Montse Soriano-Gabarró, Luis A. García Rodríguez; Diabetes, Cardiovascular Morbidity, and Risk of Age-Related Macular Degeneration in a Primary Care Population. Invest. Ophthalmol. Vis. Sci. 2015;56(3):1585-1592. https://doi.org/10.1167/iovs.14-16271.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: Age-related macular degeneration (AMD) is the most common cause of legal blindness in Western patients over 65 years of age. We aimed to establish the incidence of AMD, and the association of diabetes, and cardiovascular and eye diseases with the risk of AMD, in a large cohort of primary care patients in the United Kingdom.

Methods.: Using data from The Health Improvement Network database in the United Kingdom, all individuals with a first recorded diagnosis of AMD from 2004 to 2010 were identified (N = 10,516) and frequency-matched to 19,389 AMD-free individuals by age, sex, and calendar year of AMD occurrence. Logistic regression was used to examine comorbidities and risk factors for AMD.

Results.: The incidence of AMD was 18.08 (95% confidence interval [CI], 17.74–18.43) per 10,000 person-years. A positive association with AMD was observed for smoking, a high frequency of primary care visits, and referrals. Diabetes and use of antidiabetic drugs were associated with an increased risk of AMD. Prevalence of cardiovascular diseases among AMD patients was slightly higher than in controls, with a small increased risk of AMD among patients with myocardial infarction, heart failure, or hyperlipidemia. Positive associations were observed between prior eye diseases and risk of AMD, in particular for chorioretinal disorders.

Conclusions.: The incidence of AMD in the United Kingdom is in line with previously reported incidence rates from population-based studies. The study suggests an association between diabetes, prior eye diseases, cardiovascular comorbidities and AMD risk, and a link between AMD and higher healthcare utilization.

Introduction
Age-related macular degeneration (AMD) is the most common cause of legal blindness in Western patients over the age of 65.1 The disease greatly impacts on patients' visual acuity and their quality of life.27 There are two types of AMD: dry (atrophic) and wet (neo-vascular). Dry AMD is the most frequent type of AMD, while wet AMD represents approximately 10% of all AMD patients, and it is responsible for the majority of cases with severe visual impairment and legal blindness. 
Previous studies have reported various risk factors for AMD, including older age,8 white race,9 family history of AMD,8,10,11 and smoking.9,1216 Also, it has been suggested that maintaining a normal BMI and preventing hypertension may reduce the risk of developing AMD,9,17 and associations with AMD have been postulated with some therapeutic agents (e.g., antacids, statins, antithyroids, and hormone treatments).18,19 However, reports of these potential risk factors often have been inconsistent.12,2023 
Although the epidemiology and risk factors of AMD have been described in different patient populations and ethnic groups, information on patient characteristics, comorbidities, risk factors, and healthcare use in primary care patients in large population-based studies is limited. This study aimed to expand the current knowledge and understanding of the natural history of AMD by evaluating the potential association of cardiovascular diseases, diabetes, and other factors that may be associated with the development of the disease in a large cohort of primary care patients in the United Kingdom. The study protocol was reviewed and approved by an independent scientific review committee (SRC, reference number 11-042V). 
Materials and Methods
Data Source
Data for this study were extracted from The Health Improvement Network (THIN), a population-based computerized medical research database that contains anonymized patient data systematically recorded by primary care physicians (PCPs). The database holds over 80 million patient years of computerized prospectively collected data and covers approximately 6% of the UK population. The PCPs record data as part of their routine patient care, including information on patient's demographics and lifestyle factors (e.g., alcohol use, smoking status), consultations, referrals, hospital admissions, laboratory test results, and diagnoses, and supply this information to THIN for anonymization and use in research projects. Prescriptions issued by the PCPs are recorded automatically in the database and the Read classification is used to code specific diagnoses.24 Additional information can also be entered by the PCP as free-text. THIN has been used widely for pharmacoepidemiological research and, to date, more than 380 research articles have been published using data from this source.25 
Study Population and Follow-Up
THIN was used to identify a cohort of patients aged 50 to 89 years between January 1, 2004 and December 31, 2009 (study period). Study subjects were required to have been enrolled with their PCP for at least 2 years and to have a computerized prescription history of at least 1 year before the start of the study. To avoid the inclusion of subjects with incomplete and invalid data records, patients were excluded if they were aged over 70 years with a follow-up longer than one year, but also had fewer than two recorded PCP consultations during their entire follow-up. Individuals in the study cohort were followed from the start date (January 1, 2004) until the first of the following endpoints: first recorded diagnosis of AMD, reaching the age of 90 years, death, or the end of the study period (December 31, 2009). All individuals with a Read code for AMD before the start date were excluded. 
Ascertainment of AMD Patients
The initial ascertainment of AMD patients was performed with a computer search of patients' electronic records, based on Read codes used in previous AMD research19,26 (see Supplementary Table S1). The PCPs record AMD diagnoses based on reports received from specialists/hospital care, and choose the Read code closest to the specific diagnosis. The date of recorded AMD diagnosis was considered to be their index date. Using AMD codes recorded at the index date and up to 365 days thereafter, patients were classified into three independent subgroups: wet AMD, dry AMD, and undefined AMD (when the type of AMD was not specified). Patients diagnosed with drusen alone were not included in the final AMD study cohort. 
Validation of AMD Diagnoses
To validate the recorded AMD diagnosis, a random sample of 140 patients from the computer-identified AMD patients was selected and all clinical information from one year prior to one year after the index date, and all AMD test procedures within 30 days of the index date, were reviewed. In addition, the free-text comments for these patients (including discharge letters and specific diagnostic procedures) were manually reviewed along with their electronic records by one researcher and an external ophthalmologist. The size of the validation sample was based on resources, and corresponds to 1% of all computer-identified AMD cases. 
The validation of patient records with free-text comments revealed the confirmation rate of incident AMD to be very high (97%) and similar across the different diagnostic AMD Read codes.27 The information entered as free-text proved to be of great additional value to correctly classifying AMD subgroups in the sample validation. Patients with a record of wet AMD in one eye and dry in the other were classified as having wet AMD. If there was no mention of wet or dry in the free-text and the diagnostic Read code was unspecific AMD, the patient was classified as undefined AMD. 
Control Selection
A comparison group of 19,389 AMD-free individuals was randomly sampled from the same source general population and frequency-matched by year of birth (±1 year), sex, and year of AMD occurrence to AMD patients. A random date during the study period was assigned to each member of the comparison population; the random date was used as the index date in the nested case-control analysis. In the sampling of controls, the same eligibility criteria for AMD patients were applied with the additional condition of being free of AMD at their random date. 
Ascertainment of Patient Information
Information on patient demographics, comorbidities, lifestyle factors, and comedications were extracted from THIN any time before the index date. We evaluated general comorbidity and focused on diabetes and cardiovascular diseases. Lifestyle factors evaluated included smoking, alcohol use (units per week), and body mass index (BMI [kg/m2]). Information on healthcare use was extracted from the database any time in the year before the index date, and included the number of PCP visits and referrals. Information on drug exposure was classified as follows: current use, when the supply of the most recent prescription lasted until the index date or ended in the 30 days before the index date; recent use, when supply of the most recent prescription ended 31 to 365 days before the index date; and nonuse, when supply of the most recent prescription ended more than 365 days before the index date or there was no recorded use at any time before index date. 
Statistical Analyses
Incidence rates of AMD were calculated using the number of newly diagnosed AMD patients as the numerator and the number of corresponding person-years at risk in the study group as the denominator, stratified by sex, and year of diagnosis. Nested case–control analyses were performed for all AMD patients (dry, wet, and undefined) compared to the set of 19,389 matched controls. All covariates in the analyses were measured at any time before the index date. All statistical analyses were performed using STATA package, version 11.0 (Stata Corp LP, College Station, TX, USA). 
Results
The total study population consisted of 1,319,867 subjects. Of these, 12,986 patients were excluded because they had a recorded code of AMD before the start of the study, and 1853 patients were excluded because they were diagnosed only with drusen and had no other recorded AMD diagnosis. The automatic computer search identified 10,516 newly diagnosed AMD patients in the study population. 
Incidence of AMD
The incidence rate of any type of AMD was 18.08 (95% confidence interval [CI], 17.74–18.43) per 10,000 person-years. Incidence rates were 1.43 (95% CI, 1.34–1.53) for wet AMD, 3.07 (95% CI, 2.94–3.22) for dry AMD, and 13.58 (95% CI, 13.28–13.88) for undefined AMD, per 10,000 person-years. Based on our manual review of the sample of patients with additional free text comments, most undefined AMD would be dry AMD. The incidence rates stratified by age group are shown in the Figure. The incidence rate of AMD increased with age during the study period; for undefined AMD, the incidence rate increased exponentially after 60 years of age, starting at less than 2 per 10,000 person-years for individuals under 60 years of age and reaching 61 per 10,000 person-years among individuals aged 80 to 90 years. Females had a slightly higher incidence of any type of AMD 20.71 (95% CI, 20.21–21.22) than men 15.07 (95% CI, 14.62–15.54) across all ages. 
Figure
 
Incidence rate of AMD per 10,000 person-years by type of AMD and age group.
Figure
 
Incidence rate of AMD per 10,000 person-years by type of AMD and age group.
Characteristics of AMD Patients
General characteristics, lifestyle factors, and healthcare use of all AMD patients are shown in Table 1 by type of AMD, along with their association with the development of AMD. Wet AMD was recorded in 832 (7.9%) patients, and 1788 (17.5%) patients were diagnosed with dry AMD. A third of AMD patients (7896; 75.1%), were classified as having undefined AMD. The mean age at presentation of AMD was 77 years for all AMD types. 
Table 1
 
Frequency Distribution of General Characteristics, Lifestyle Factors, and Healthcare Use Among AMD Cases by Subgroup, and Association With Development of AMD Compared to AMD-Free Controls
Table 1
 
Frequency Distribution of General Characteristics, Lifestyle Factors, and Healthcare Use Among AMD Cases by Subgroup, and Association With Development of AMD Compared to AMD-Free Controls
Controls,N= 19,389,n(%) Cohort of AMD Patients, N = 10,516 OR*, 95% CI, All AMD vs. Controls
Wet AMD,N= 832,n(%) Dry AMD,N= 1788,n(%) Undefined AMD,N= 7896,n(%)
Sex
 Male 7,698 (39.7) 329 (39.5) 716 (40.0) 3037 (38.5) N/A
 Female 11,691 (60.3) 503 (60.5) 1072 (60.0) 4859 (61.5)
Age, y
 50–59 1,150 (5.9) 40 (4.8) 51 (2.9) 303 (3.8) N/A
 60–69 3,008 (15.5) 101 (12.1) 228 (12.8) 1047 (13.3)
 70–79 7,490 (38.6) 350 (42.1) 667 (37.3) 3075 (38.9)
 80–85 7,741 (39.9) 341 (41.0) 842 (47.1) 3471 (44.0)
Smoking
 Nonsmoker 9,437 (48.7) 342 (41.1) 741 (41.4) 3415 (43.2) 1.0
 Smoker 2,061 (10.6) 92 (11.1) 161 (9.0) 925 (11.7) 1.29 (1.19–1.40)
 Ex-smoker 7,430 (38.3) 390 (46.9) 859 (48.0) 3443 (43.6) 1.27 (1.20–1.34)
 Unknown 461 (2.4) 8 (1.0) 27 (1.5) 113 (1.4) 0.87 (0.71–1.05)
BMI, kg/m2
 15–19 929 (4.8) 43 (5.2) 93 (5.2) 373 (4.7) 0.95 (0.85–1.07)
 20–24 5,793 (29.9) 228 (27.4) 568 (31.8) 2404 (30.4) 1.0
 25–29 6,807 (35.1) 295 (35.5) 625 (35.0) 2801 (35.5) 0.97 (0.92–1.03)
 ≥30 3,607 (18.6) 191 (23.0) 366 (20.5) 1609 (20.4) 1.05 (0.98–1.13)
 Unknown 2,253 (11.6) 75 (9.0) 136 (7.6) 709 (9.0) 0.83 (0.75–0.90)
Alcohol use, units†/w
 None 9,632 (49.7) 410 (49.3) 928 (51.9) 4135 (52.4) 1.0
 1–4 3,288 (17.0) 160 (19.2) 289 (16.2) 1375 (17.4) 1.01 (0.95–1.08)
 5–15 3,162 (16.3) 137 (16.5) 311 (17.4) 1246 (15.8) 0.99 (0.92-1.06)
 ≥16 1,147 (5.9) 58 (7.0) 118 (6.6) 501 (6.3) 1.14 (1.03–1.27)
 None 2,160 (11.1) 67 (8.1) 142 (7.9) 639 (8.1) 0.80 (0.73–0.88)
PCP visits‡
 0–3 2,419 (12.5) 35 (4.2) 69 (3.9) 419 (5.3) 1.0
 4–9 5,727 (29.5) 177 (21.3) 428 (23.9) 1981 (25.1) 2.04 (1.83–2.26)
 ≥10 11,243 (58.0) 620 (74.5) 1291 (72.2) 5496 (69.6) 2.95 (2.67–3.26)
Referrals‡
 None 6,464 (33.3) 72 (8.7) 211 (11.8) 1021 (12.9) 1.0
 1–3 7,180 (37.0) 310 (37.3) 775 (43.3) 3371 (42.7) 2.85 (2.65–3.07)
 ≥4 5,745 (29.6) 450 (54.1) 802 (44.9) 3504 (44.4) 3.73 (3.45–4.03)
The prevalence of lifestyle factors and healthcare use did not differ substantially across AMD subgroups. A notable difference was the higher frequency of healthcare service use (visits to PCPs and referrals) among AMD patients compared to controls. Current and past smoking was associated with an increased risk of AMD (odds ratio [OR], 1.29; 95% CI, 1.19–1.40 and OR, 1.27; 95% CI, 1.20–1.34, respectively). No differences between AMD subgroups and controls were found regarding socioeconomic characteristics as measured by socioeconomic deprivation (Townsend Index), or residing in an urban or rural area (data not shown). 
Comorbidity and Drug Use
Diabetes prevalence was higher among patients with wet AMD than patients with other types of AMD. Type I diabetes carried an increased risk of AMD (OR, 2.25; 95% CI, 1.62–3.12), with the risk higher among patients with wet AMD (OR, 4.58; 95% CI, 2.56–8.19; Table 2). Patients with type II diabetes also carried an increased risk of AMD (OR, 1.38; 95% CI, 1.29–1.47). Prevalence of cardiovascular diseases among AMD patients was slightly higher than among controls. We observed a small increased risk of AMD among patients with certain cardiovascular comorbidities: OR 1.11 (95% CI, 1.02–1.22) for myocardial infarction; OR 1.14 (95% CI, 1.03–1.26) for heart failure, and OR 1.08 (95% CI, 1.02–1.15) for hyperlipidemia. 
Table 2
 
Prevalence of Selected Comorbidity Among AMD Cases by Subgroup, and Association With Development of AMD Compared to AMD-Free Controls
Table 2
 
Prevalence of Selected Comorbidity Among AMD Cases by Subgroup, and Association With Development of AMD Compared to AMD-Free Controls
Controls,N= 19,389,n(%) Cohort of AMD Patients,N= 10,516 OR*, 95% CI, All AMD vs. Controls
Wet AMD,N= 832,n(%) Dry AMD,N= 1788,n(%) Undefined AMD,N= 7896,n(%)
Endocrine diseases
 Diabetes
  Type I 66 (0.3) 15 (1.8) 13 (0.7) 54 (0.7) 2.25 (1.62–3.12)
  Type II 2388 (12.3) 138 (16.6) 314 (17.6) 1464 (18.5) 1.38 (1.29–1.47)
 Hypothyroidism 1839 (9.5) 110 (13.2) 197 (11.0) 858 (10.9) 1.06 (0.98–1.15)
 Hyperthyroidism 392 (2.0) 18 (2.2) 51 (2.9) 185 (2.3) 1.11 (0.94–1.30)
Cardiovascular diseases
 IHD 180 (21.6) 442 (24.7) 1864 (23.6) 180 (21.6) 1.04 (0.98–1.10)
 MI 63 (7.6) 155 (8.7) 685 (8.7) 63 (7.6) 1.11 (1.02–1.22)
 AF 75 (9.0) 228 (12.8) 876 (11.1) 75 (9.0) 1.05 (0.97–1.13)
 HF 38 (4.6) 135 (7.6) 546 (6.9) 38 (4.6) 1.14 (1.03–1.26)
 HT 488 (58.7) 1033 (57.8) 4635 (58.7) 488 (58.7) 1.01 (0.96–1.07)
 Hyperlipidemia 201 (24.2) 454 (25.4) 1932 (24.5) 201 (24.2) 1.08 (1.02–1.15)
Eye diseases
 Glaucoma 1336 (6.9) 75 (9.0) 161 (9.0) 736 (9.3) 1.27 (1.17–1.39)
 Retinal diseases 1293 (6.7) 197 (23.7) 215 (12.0) 1189 (15.1) 2.29 (2.12–2.48)
 Cataracts 3701 (19.1) 249 (29.9) 617 (34.5) 2657 (33.6) 1.98 (1.87–2.10)
 Refractory/accommodation and visual disturbances 1048 (5.4) 123 (14.8) 179 (10.0) 954 (12.1) 2.22 (2.04–2.42)
 Eye examinations, tests and procedures† 6980 (36.0) 507 (60.9) 1054 (58.9) 4449 (56.3) 2.16 (2.05–2.27)
Cataracts and other eye disorders that were evaluated were associated with an increased risk of AMD (Table 2). Retinal diseases were recorded for 23.7% of patients with wet AMD compared to 12.0% for dry AMD and 15.1% for undefined AMD. More than half of AMD patients (57.2%) had a record of at least one AMD diagnostic procedure, eye examination or investigative test. Either weak or no associations with AMD were found for all other general comorbidities evaluated in the study (Table 2). 
The frequency of comedications used in patients with AMD and in controls is shown in Table 3. Current use of antidiabetic treatment drugs was associated with an increased risk of AMD (OR, 1.39; 95% CI, 1.28–1.50), in line with our finding for diabetes. We did not find a clear association between AMD and use of aspirin or lipid-lowering therapy. All other drug treatments studied were either not associated with AMD or the magnitude of association was weak. 
Table 3
 
Frequency Distribution of Comedications Among AMD Patients by Subgroup and Association With Development of AMD Compared to AMD-Free Controls
Table 3
 
Frequency Distribution of Comedications Among AMD Patients by Subgroup and Association With Development of AMD Compared to AMD-Free Controls
Controls,N= 19,389,n(%) Cohort of AMD Patients, N = 10,516 OR,* 95% CI, All AMD vs. Controls
Wet AMD,N= 832,n(%) Dry AMD,N= 1788,n(%) Undefined AMD,N= 7896,n(%)
Eye drops
 Current use† 985 (5.1) 59 (7.1) 143 (8.0) 615 (7.8) 1.43 (1.30–1.58)
 Recent use‡ 1,416 (7.3) 96 (11.5) 172 (9.6) 770 (9.8) 1.27 (1.17–1.39)
Anti-diabetics
 Current use† 1,653 (8.5) 113 (13.6) 230 (12.9) 1024 (13.0) 1.39 (1.28–1.50)
 Recent use‡ 89 (0.5) 1 (0.1) 6 (0.3) 53 (0.7) 1.15 (0.83–1.60)
Oral corticosteroids
 Current use† 689 (3.6) 19 (2.3) 76 (4.3) 364 (4.6) 1.05 (0.93–1.18)
 Recent use‡ 801 (4.1) 53 (6.4) 94 (5.3) 387 (4.9) 1.08 (0.96–1.21)
Aspirin
 Current use† 5,606 (28.9) 273 (32.8) 618 (34.6) 2541 (32.2) 1.03 (0.98–1.09)
 Recent use‡ 901 (4.6) 52 (6.2) 99 (5.5) 435 (5.5) 1.08 (0.96–1.20)
Oral-anticoagulants
 Current use† 1,074 (5.5) 50 (6.0) 132 (7.4) 532 (6.7) 1.03 (0.93–1.13)
 Recent use‡ 168 (0.9) 9 (1.1) 13 (0.7) 76 (1.0) 0.89 (0.69–1.15)
Lipid-lowering drugs
 Current use† 6,451 (33.3) 323 (38.8) 696 (38.9) 2993 (37.9) 1.06 (1.01–1.12)
 Recent use‡ 617 (3.2) 31 (3.7) 69 (3.9) 254 (3.2) 0.98 (0.86–1.13)
Antihypertensives
 Current use† 11,644 (60.1) 549 (66.0) 1216 (68.0) 5212 (66.0) 1.01 (0.95–1.07)
 Recent use‡ 760 (3.9) 38 (4.6) 53 (3.0) 285 (3.6) 0.84 (0.74–0.96)
Discussion
This study established the incidence of AMD in a large cohort of primary care patients in the United Kingdom and investigated the association of diabetes, cardiovascular diseases, and prior eye diseases as potential risk factors for the development of AMD. The incidence rates in this study appeared to be in line with previously reported cumulative incidence rates from population-based studies in predominantly white populations.28 Our findings confirmed that increasing age is the main risk factor for AMD, and showed a positive association between AMD and smoking, consistent with reports from others.8,9,1317,2931 As found by Cho et al.,32 our findings also do not support an inverse relationship between moderate alcohol use and risk of AMD. 
Findings from this study indicated that diabetic patients carry a higher risk of AMD compared to nondiabetics, with type I diabetics having an approximately 2-fold increased risk and type II diabetics having a 1.4-fold increased risk. These results are consistent with previous research. A 1.8-fold increased risk was reported for neovascular AMD by the Age-Related Eye Disease Study (AREDS) research group in a multiclinic-based prospective cohort study in the United States,9 while a borderline significant 1.7-fold increased risk for late AMD was reported associated with diabetes based on pooled data from four prospective cohort studies.8 In a hospital record-linkage study in the United Kingdom, Goldacre et al.33 found diabetes to be associated with an even greater risk of AMD, reporting a 3.5-fold increased risk among a diabetic cohort compared to a reference cohort. The investigators postulate that potential mechanisms may involve damage to the choroidal circulation, Bruch's membrane, and the retinal pigment epithelium. However, in contrast to these findings, a large population-based cross-sectional study of 14,352 people in Korea concluded that, following adjustment for major confounders, any type of AMD was less prevalent among diabetic patients, with a reported 26% reduction in risk among people in this patient group.34 
A strong positive association also was found between AMD and the eye diseases that were evaluated. Few studies have investigated this association, although analysis of pooled data from three cohort studies reported cataract surgery to be associated with a 3-fold increased risk of AMD.8 
Age-related macular degeneration long has been associated with generalized atherosclerosis.8,28 The vasculature of the eye and the heart share several common characteristics and it is plausible that risk factors for arteriosclerosis, such as diabetes and dyslipidemia also are risk factors for eye diseases.35 Cardiovascular comorbidities were evaluated in this study, and small increased risks of AMD were found among patients with myocardial infarction, heart failure, or hyperlipidemia. The Alienor study, a smaller population-based study in France,36 found that patients with elevated high-density lipoprotein levels were associated with approximately double the risk of AMD; however, they found no relation between systolic or diastolic blood pressure, hypertension, or use of antihypertensive medications.37 Similarly in the current study, hypertension was not a risk factor for AMD, yet findings from a case–control study by the AREDS Research Group17 found hypertension to be associated with a significant 1.5-fold increased risk of AMD. A recent study by Wang et al.38 reported an independent association of coronary stenosis with early AMD, suggesting that patients with coronary heart disease could be screened for detection of early AMD. Klein et al.39 similarly reported presence of carotid plaques and carotid artery intima-media thickness to be weakly associated with incidence of late AMD, and suggest that this may reflect reduced blood flow in choroidal blood vessels, affecting outer retinal/choroidal function. However, the investigators found no associations between angina, myocardial infarction (MI), and stroke and AMD risk. Although the increased risks of AMD among patients with cardiovascular comorbidities were small in our study, these findings could have some implications when assessing potential risks of AMD therapies. They suggest that any safety evaluation of AMD treatment should take into account patients' cardiovascular comorbidities, as well as the presence of diabetes, at the start of therapy. 
A noteworthy finding of potential importance was the higher frequency of primary care visits and referrals in patients with AMD compared to controls, in line with findings by others.40 Patients receive a diagnosis of AMD following referral to specialist care for investigations, and these visits will contribute to increased health care costs associated with these patients. Previous publications have reported an association between visual impairment and rate of hospital admission that could be attributed to higher levels of comorbidity.41 Our results also suggested that exposure to some comedications, such as antidiabetics (used as a proxy measure for diabetes), could be associated with an increased risk of AMD. Based on the potential association between AMD, and cardiovascular disease and dyslipidemia, it has been suggested that aspirin and statins also may offer protection against AMD.42 Findings of a protective effect of aspirin have so far been mixed, with some studies demonstrating protection,43,44 some reporting an increased risk of AMD,45,46 and others finding no association.47 Our findings for lipid-lowering drugs (nearly all accounted for by statin use) and aspirin were compatible with no risk of AMD, or a small increased risk, similar to previous studies.18,40,46,4850 This study was unable to evaluate other potential risk factors, such as race, family history, and other genetic factors as these are not widely recorded in THIN. Therefore, we could not control their potential effect on identified risk factors for AMD. 
The main strength of this study was the inclusion of a large, population-based, incident cohort of AMD patients in the real-world setting, using clinical information recorded by PCPs in routine daily practice. However, the study also has limitations. Specifically, in a large number of patients, the type of AMD could not be determined, as the PCP may not always record all clinical information received from specialist and hospital discharge letters as part of routine patient care. Because the classification of AMD patients was performed based on the Read codes recorded by the PCP without reviewing the full patient record, some patients recorded as “undefined AMD” could have been either wet or dry AMD patients. We did, however, address this weakness by validating the diagnoses through manual review of the medical records of a random sample of patients, including the review of free-text comments. A confirmation rate of 97% was obtained, which is in line with a previous validation study of AMD using the General Practice Research Database, a similar database to THIN, where the authors reported a high confirmation rate of 93.5%.48 We also found that a small number of AMD cases had a prior record of a chorioretinal disorder, which could reflect an initial presentation of AMD in most of these patients. Among controls, 39 individuals had a record of a prior chorioretinal disorder, and, therefore, could have potentially been misclassified cases of AMD. 
This study in a large cohort of primary care patients in the United Kingdom identified diabetes and smoking as significant risk factors for AMD. Having prior eye diseases also was identified as a significant risk factor for AMD, while some of the other comorbidities examined, predominantly cardiovascular comorbidities, carried a small increased risk of AMD. Further prospective evaluation of these relationships is warranted, as these findings could have therapeutic implications and prompt screening for early AMD monitoring in these patient subgroups. Additional studies are needed to further quantify the impact of suggested risk factors of AMD, as well as the higher frequency of healthcare use in these patients. 
Acknowledgments
Supported by Bayer Pharma AG, Berlin, Germany. Medical writing assistance was provided by Susan Bromley (EpiMed Communications Ltd, Oxford, UK), and funded by Bayer Pharma AG. 
Disclosure: Z.P. Vassilev, Bayer Pharma AG (E); A. Ruigómez, Bayer Pharma AG (F); M. Soriano-Gabarró, Bayer Pharma AG (E); L.A. García Rodríguez, Bayer Pharma AG (F, R) 
References
Ambati J Ambati BK Yoo SH Ianchulev S Adamis AP. Age-related macular degeneration: etiology, pathogenesis, and therapeutic strategies. Surv Ophthalmol. 2003; 48: 257–293. [CrossRef] [PubMed]
Brody BL Gamst AC Williams RA Depression, visual acuity, comorbidity, and disability associated with age-related macular degeneration. Ophthalmology. 2001; 108: 1893–1900, discussion 1900. [CrossRef] [PubMed]
DeCarlo DK Scilley K Wells J Owsley C. Driving habits and health-related quality of life in patients with age-related maculopathy. Optom Vis Sci. 2003; 80: 207–213. [CrossRef] [PubMed]
Williams RA Brody BL Thomas RG Kaplan RM Brown SI. The psychosocial impact of macular degeneration. Arch Ophthalmol. 1998; 116: 514–520. [CrossRef] [PubMed]
Mangione CM Gutierrez PR Lowe G Orav EJ Seddon JM. Influence of age-related maculopathy on visual functioning and health-related quality of life. Am J Ophthalmol. 1999; 128: 45–53. [CrossRef] [PubMed]
Rovner BW Casten RJ Tasman WS. Effect of depression on vision function in age-related macular degeneration. Arch Ophthalmol. 2002; 120: 1041–1044. [CrossRef] [PubMed]
Scott IU Feuer WJ Jacko JA. Impact of visual function on computer task accuracy and reaction time in a cohort of patients with age-related macular degeneration. Am J Ophthalmol. 2002; 133: 350–357. [CrossRef] [PubMed]
Chakravarthy U Wong TY Fletcher A Clinical risk factors for age-related macular degeneration: a systematic review and meta-analysis. BMC Ophthalmol. 2010; 10: 31. [CrossRef] [PubMed]
Clemons TE Milton RC Klein R Seddon JM Ferris FL III. Risk factors for the incidence of advanced age-related macular degeneration in the Age-Related Eye Disease Study (AREDS) AREDS report no. 19. Ophthalmology. 2005; 112: 533–539. [CrossRef] [PubMed]
Goodman DM Parmet S Lynm C Livingston EH. JAMA patient page. Age-related macular degeneration. JAMA. 2012; 308: 1702. [CrossRef] [PubMed]
van de Ven JP Smailhodzic D Boon CJ Association analysis of genetic and environmental risk factors in the cuticular drusen subtype of age-related macular degeneration. Mol Vis. 2012; 18: 2271–2278. [PubMed]
Smith W Assink J Klein R Risk factors for age-related macular degeneration: pooled findings from three continents. Ophthalmology. 2001; 108: 697–704. [CrossRef] [PubMed]
Seddon JM George S Rosner B. Cigarette smoking, fish consumption, omega-3 fatty acid intake, and associations with age-related macular degeneration: the US Twin Study of Age-Related Macular Degeneration. Arch Ophthalmol. 2006; 124: 995–1001. [CrossRef] [PubMed]
Evans JR Fletcher AE Wormald RP. 28,000 Cases of age related macular degeneration causing visual loss in people aged 75 years and above in the United Kingdom may be attributable to smoking. Br J Ophthalmol. 2005; 89: 550–553. [CrossRef] [PubMed]
Khan JC Thurlby DA Shahid H Smoking and age related macular degeneration: the number of pack years of cigarette smoking is a major determinant of risk for both geographic atrophy and choroidal neovascularisation. Br J Ophthalmol. 2006; 90: 75–80. [CrossRef] [PubMed]
Guymer RH Chong EW. Modifiable risk factors for age-related macular degeneration. Med J Aust. 2006; 184: 455–458. [PubMed]
Age-Related Eye Disease Study Research Group. Risk factors associated with age-related macular degeneration. A case-control study in the age-related eye disease study: Age-Related Eye Disease Study Report Number 3. Ophthalmology. 2000; 107: 2224–2232. [CrossRef] [PubMed]
Smeeth L Douglas I Hall AJ Hubbard R Evans S. Effect of statins on a wide range of health outcomes: a cohort study validated by comparison with randomized trials. Br J Clin Pharmacol. 2009; 67: 99–109. [CrossRef] [PubMed]
Douglas IJ Cook C Chakravarthy U Hubbard R Fletcher AE Smeeth L. A case-control study of drug risk factors for age-related macular degeneration. Ophthalmology. 2007; 114: 1164–1169. [CrossRef] [PubMed]
Wang JJ Foran S Smith W Mitchell P. Risk of age-related macular degeneration in eyes with macular drusen or hyperpigmentation: the Blue Mountains Eye Study cohort. Arch Ophthalmol. 2003; 121: 658–663. [CrossRef] [PubMed]
Klein R Klein BE Tomany SC Moss SE. Ten-year incidence of age-related maculopathy and smoking and drinking: the Beaver Dam Eye Study. Am J Epidemiol. 2002; 156: 589–598. [CrossRef] [PubMed]
Hyman L Neborsky R. Risk factors for age-related macular degeneration: an update. Curr Opin Ophthalmol. 2002; 13: 171–175. [CrossRef] [PubMed]
Miyazaki M Nakamura H Kubo M Risk factors for age related maculopathy in a Japanese population: the Hisayama study. Br J Ophthalmol. 2003; 87: 469–472. [CrossRef] [PubMed]
Stuart-Buttle CD Read JD Sanderson HF Sutton YM. A language of health in action: read codes, classifications and groupings. Proc AMIA Annu Fall Symp. 1996: 75–79.
CSD Medical Research UK. Available at: http://csdmruk.cegedim.com/. Accessed June 26, 2014.
Nitsch D Douglas I Smeeth L Fletcher A. Age-related macular degeneration and complement activation-related diseases: a population-based case-control study. Ophthalmology. 2008; 115: 1904–1910. [CrossRef] [PubMed]
Vassilev ZP García Rodríguez LA Soriano-Gabarró M Ruigómez A. Validation of diagnosis of age-related macular degeneration in a computer database of primary care (THIN). Poster presented at: 30th Anniversary International Conference on Pharmacoepidemiology and Therapeutic Risk Management; October 24–27, 2014; Taipei, Taiwan.
Tomany SC Wang JJ Van Leeuwen R Risk factors for incident age-related macular degeneration: pooled findings from 3 continents. Ophthalmology. 2004; 111: 1280–1287. [CrossRef] [PubMed]
De Jong PT Bergen AA Klaver CC Van Duijn CM Assink JM. Age-related maculopathy: its genetic basis. Eye. 2001; 15: 396–400. [CrossRef] [PubMed]
Mitchell P Wang JJ Foran S Smith W. Five-year incidence of age-related maculopathy lesions: the Blue Mountains Eye Study. Ophthalmology. 2002; 109: 1092–1097. [CrossRef] [PubMed]
Kabasawa S Mori K Horie-Inoue K Associations of cigarette smoking but not serum fatty acids with age-related macular degeneration in a Japanese population. Ophthalmology. 2011; 118: 1082–1088. [CrossRef] [PubMed]
Cho E Hankinson SE Willett WC Prospective study of alcohol consumption and the risk of age-related macular degeneration. Arch Ophthalmol. 2000; 118: 681–688. [CrossRef] [PubMed]
Goldacre MJ Wotton CJ Keenan TD. Risk of selected eye diseases in people admitted to hospital for hypertension or diabetes mellitus: record linkage studies. Br J Ophthalmol. 2012; 96: 872–876. [CrossRef] [PubMed]
Cho BJ Heo JW Shin JP Ahn J Kim TW Chung H. Epidemiological association between systemic diseases and age-related macular degeneration: the Korea National Health and Nutrition Examination Survey 2008–2011. Invest Ophthalmol Vis Sci. 2014; 55: 4430–4437. [CrossRef] [PubMed]
Flammer J Konieczka K Bruno RM Virdis A Flammer AJ Taddei S. The eye and the heart. Eur Heart J. 2013; 34: 1270–1278. [CrossRef] [PubMed]
Cougnard-Gregoire A Delyfer MN Korobelnik JF Elevated high-density lipoprotein cholesterol and age-related macular degeneration: the Alienor study. PLoS One. 2014; 9: e90973. [CrossRef] [PubMed]
Cougnard-Gregoire A Delyfer MN Korobelnik JF Long-term blood pressure and age-related macular degeneration: the ALIENOR study. Invest Ophthalmol Vis Sci. 2013; 54: 1905–1912. [CrossRef] [PubMed]
Wang SB Mitchell P Chiha J Severity of coronary artery disease is independently associated with the frequency of early age-related macular degeneration. Br J Ophthalmol. 2015 Mar; 99: 365–370.
Klein R Cruickshanks KJ Myers CE The relationship of atherosclerosis to the 10-year cumulative incidence of age-related macular degeneration: the Beaver Dam studies. Ophthalmology. 2013; 120: 1012–1019. [CrossRef] [PubMed]
Shalev V Sror M Goldshtein I Kokia E Chodick G. Statin use and the risk of age related macular degeneration in a large health organization in Israel. Ophthalmic Epidemiol. 2011; 18: 83–90. [PubMed]
Evans JR Smeeth L Fletcher AE. Hospital admissions in older people with visual impairment in Britain. BMC Ophthalmol. 2008; 8: 16. [CrossRef] [PubMed]
Cheung CM Wong TY. Is age-related macular degeneration a manifestation of systemic disease? New prospects for early intervention and treatment. J Intern Med. 2014; 276: 140–153. [CrossRef] [PubMed]
Cheung N Tay WT Cheung GC Wang JJ Mitchell P Wong TY. Is aspirin intake associated with early age-related macular degeneration? The Singapore Indian Eye Study. Br J Ophthalmol. 2013; 97: 785–788. [CrossRef] [PubMed]
Wilson HL Schwartz DM Bhatt HR McCulloch CE Duncan JL. Statin and aspirin therapy are associated with decreased rates of choroidal neovascularization among patients with age-related macular degeneration. Am J Ophthalmol. 2004; 137: 615–624. [PubMed]
de Jong PT Chakravarthy U Rahu M Associations between aspirin use and aging macula disorder: the European Eye Study. Ophthalmology. 2012; 119: 112–118. [CrossRef] [PubMed]
Klein BE Howard KP Gangnon RE Dreyer JO Lee KE Klein R. Long-term use of aspirin and age-related macular degeneration. JAMA. 2012; 308: 2469–2478. [CrossRef] [PubMed]
Tan JS Mitchell P Smith W Wang JJ. Cardiovascular risk factors and the long-term incidence of age-related macular degeneration: the Blue Mountains Eye Study. Ophthalmology. 2007; 114: 1143–1150. [CrossRef] [PubMed]
Smeeth L Cook C Chakravarthy U Hubbard R Fletcher AE. A case control study of age related macular degeneration and use of statins. Br J Ophthalmol. 2005; 89: 1171–1175. [CrossRef] [PubMed]
Liew G Mitchell P Wong TY Rochtchina E Wang JJ. The association of aspirin use with age-related macular degeneration. JAMA Intern Med. 2013; 173: 258–264. [CrossRef] [PubMed]
Klein R Knudtson MD Klein BE. Statin use and the five-year incidence and progression of age-related macular degeneration. Am J Ophthalmol. 2007; 144: 1–6. [CrossRef] [PubMed]
Figure
 
Incidence rate of AMD per 10,000 person-years by type of AMD and age group.
Figure
 
Incidence rate of AMD per 10,000 person-years by type of AMD and age group.
Table 1
 
Frequency Distribution of General Characteristics, Lifestyle Factors, and Healthcare Use Among AMD Cases by Subgroup, and Association With Development of AMD Compared to AMD-Free Controls
Table 1
 
Frequency Distribution of General Characteristics, Lifestyle Factors, and Healthcare Use Among AMD Cases by Subgroup, and Association With Development of AMD Compared to AMD-Free Controls
Controls,N= 19,389,n(%) Cohort of AMD Patients, N = 10,516 OR*, 95% CI, All AMD vs. Controls
Wet AMD,N= 832,n(%) Dry AMD,N= 1788,n(%) Undefined AMD,N= 7896,n(%)
Sex
 Male 7,698 (39.7) 329 (39.5) 716 (40.0) 3037 (38.5) N/A
 Female 11,691 (60.3) 503 (60.5) 1072 (60.0) 4859 (61.5)
Age, y
 50–59 1,150 (5.9) 40 (4.8) 51 (2.9) 303 (3.8) N/A
 60–69 3,008 (15.5) 101 (12.1) 228 (12.8) 1047 (13.3)
 70–79 7,490 (38.6) 350 (42.1) 667 (37.3) 3075 (38.9)
 80–85 7,741 (39.9) 341 (41.0) 842 (47.1) 3471 (44.0)
Smoking
 Nonsmoker 9,437 (48.7) 342 (41.1) 741 (41.4) 3415 (43.2) 1.0
 Smoker 2,061 (10.6) 92 (11.1) 161 (9.0) 925 (11.7) 1.29 (1.19–1.40)
 Ex-smoker 7,430 (38.3) 390 (46.9) 859 (48.0) 3443 (43.6) 1.27 (1.20–1.34)
 Unknown 461 (2.4) 8 (1.0) 27 (1.5) 113 (1.4) 0.87 (0.71–1.05)
BMI, kg/m2
 15–19 929 (4.8) 43 (5.2) 93 (5.2) 373 (4.7) 0.95 (0.85–1.07)
 20–24 5,793 (29.9) 228 (27.4) 568 (31.8) 2404 (30.4) 1.0
 25–29 6,807 (35.1) 295 (35.5) 625 (35.0) 2801 (35.5) 0.97 (0.92–1.03)
 ≥30 3,607 (18.6) 191 (23.0) 366 (20.5) 1609 (20.4) 1.05 (0.98–1.13)
 Unknown 2,253 (11.6) 75 (9.0) 136 (7.6) 709 (9.0) 0.83 (0.75–0.90)
Alcohol use, units†/w
 None 9,632 (49.7) 410 (49.3) 928 (51.9) 4135 (52.4) 1.0
 1–4 3,288 (17.0) 160 (19.2) 289 (16.2) 1375 (17.4) 1.01 (0.95–1.08)
 5–15 3,162 (16.3) 137 (16.5) 311 (17.4) 1246 (15.8) 0.99 (0.92-1.06)
 ≥16 1,147 (5.9) 58 (7.0) 118 (6.6) 501 (6.3) 1.14 (1.03–1.27)
 None 2,160 (11.1) 67 (8.1) 142 (7.9) 639 (8.1) 0.80 (0.73–0.88)
PCP visits‡
 0–3 2,419 (12.5) 35 (4.2) 69 (3.9) 419 (5.3) 1.0
 4–9 5,727 (29.5) 177 (21.3) 428 (23.9) 1981 (25.1) 2.04 (1.83–2.26)
 ≥10 11,243 (58.0) 620 (74.5) 1291 (72.2) 5496 (69.6) 2.95 (2.67–3.26)
Referrals‡
 None 6,464 (33.3) 72 (8.7) 211 (11.8) 1021 (12.9) 1.0
 1–3 7,180 (37.0) 310 (37.3) 775 (43.3) 3371 (42.7) 2.85 (2.65–3.07)
 ≥4 5,745 (29.6) 450 (54.1) 802 (44.9) 3504 (44.4) 3.73 (3.45–4.03)
Table 2
 
Prevalence of Selected Comorbidity Among AMD Cases by Subgroup, and Association With Development of AMD Compared to AMD-Free Controls
Table 2
 
Prevalence of Selected Comorbidity Among AMD Cases by Subgroup, and Association With Development of AMD Compared to AMD-Free Controls
Controls,N= 19,389,n(%) Cohort of AMD Patients,N= 10,516 OR*, 95% CI, All AMD vs. Controls
Wet AMD,N= 832,n(%) Dry AMD,N= 1788,n(%) Undefined AMD,N= 7896,n(%)
Endocrine diseases
 Diabetes
  Type I 66 (0.3) 15 (1.8) 13 (0.7) 54 (0.7) 2.25 (1.62–3.12)
  Type II 2388 (12.3) 138 (16.6) 314 (17.6) 1464 (18.5) 1.38 (1.29–1.47)
 Hypothyroidism 1839 (9.5) 110 (13.2) 197 (11.0) 858 (10.9) 1.06 (0.98–1.15)
 Hyperthyroidism 392 (2.0) 18 (2.2) 51 (2.9) 185 (2.3) 1.11 (0.94–1.30)
Cardiovascular diseases
 IHD 180 (21.6) 442 (24.7) 1864 (23.6) 180 (21.6) 1.04 (0.98–1.10)
 MI 63 (7.6) 155 (8.7) 685 (8.7) 63 (7.6) 1.11 (1.02–1.22)
 AF 75 (9.0) 228 (12.8) 876 (11.1) 75 (9.0) 1.05 (0.97–1.13)
 HF 38 (4.6) 135 (7.6) 546 (6.9) 38 (4.6) 1.14 (1.03–1.26)
 HT 488 (58.7) 1033 (57.8) 4635 (58.7) 488 (58.7) 1.01 (0.96–1.07)
 Hyperlipidemia 201 (24.2) 454 (25.4) 1932 (24.5) 201 (24.2) 1.08 (1.02–1.15)
Eye diseases
 Glaucoma 1336 (6.9) 75 (9.0) 161 (9.0) 736 (9.3) 1.27 (1.17–1.39)
 Retinal diseases 1293 (6.7) 197 (23.7) 215 (12.0) 1189 (15.1) 2.29 (2.12–2.48)
 Cataracts 3701 (19.1) 249 (29.9) 617 (34.5) 2657 (33.6) 1.98 (1.87–2.10)
 Refractory/accommodation and visual disturbances 1048 (5.4) 123 (14.8) 179 (10.0) 954 (12.1) 2.22 (2.04–2.42)
 Eye examinations, tests and procedures† 6980 (36.0) 507 (60.9) 1054 (58.9) 4449 (56.3) 2.16 (2.05–2.27)
Table 3
 
Frequency Distribution of Comedications Among AMD Patients by Subgroup and Association With Development of AMD Compared to AMD-Free Controls
Table 3
 
Frequency Distribution of Comedications Among AMD Patients by Subgroup and Association With Development of AMD Compared to AMD-Free Controls
Controls,N= 19,389,n(%) Cohort of AMD Patients, N = 10,516 OR,* 95% CI, All AMD vs. Controls
Wet AMD,N= 832,n(%) Dry AMD,N= 1788,n(%) Undefined AMD,N= 7896,n(%)
Eye drops
 Current use† 985 (5.1) 59 (7.1) 143 (8.0) 615 (7.8) 1.43 (1.30–1.58)
 Recent use‡ 1,416 (7.3) 96 (11.5) 172 (9.6) 770 (9.8) 1.27 (1.17–1.39)
Anti-diabetics
 Current use† 1,653 (8.5) 113 (13.6) 230 (12.9) 1024 (13.0) 1.39 (1.28–1.50)
 Recent use‡ 89 (0.5) 1 (0.1) 6 (0.3) 53 (0.7) 1.15 (0.83–1.60)
Oral corticosteroids
 Current use† 689 (3.6) 19 (2.3) 76 (4.3) 364 (4.6) 1.05 (0.93–1.18)
 Recent use‡ 801 (4.1) 53 (6.4) 94 (5.3) 387 (4.9) 1.08 (0.96–1.21)
Aspirin
 Current use† 5,606 (28.9) 273 (32.8) 618 (34.6) 2541 (32.2) 1.03 (0.98–1.09)
 Recent use‡ 901 (4.6) 52 (6.2) 99 (5.5) 435 (5.5) 1.08 (0.96–1.20)
Oral-anticoagulants
 Current use† 1,074 (5.5) 50 (6.0) 132 (7.4) 532 (6.7) 1.03 (0.93–1.13)
 Recent use‡ 168 (0.9) 9 (1.1) 13 (0.7) 76 (1.0) 0.89 (0.69–1.15)
Lipid-lowering drugs
 Current use† 6,451 (33.3) 323 (38.8) 696 (38.9) 2993 (37.9) 1.06 (1.01–1.12)
 Recent use‡ 617 (3.2) 31 (3.7) 69 (3.9) 254 (3.2) 0.98 (0.86–1.13)
Antihypertensives
 Current use† 11,644 (60.1) 549 (66.0) 1216 (68.0) 5212 (66.0) 1.01 (0.95–1.07)
 Recent use‡ 760 (3.9) 38 (4.6) 53 (3.0) 285 (3.6) 0.84 (0.74–0.96)
Supplementary Table S1
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×