Portal de Periódicos da CAPES

Visual Impairment and quality of life in the Older European Population, the EUREYE study

2009; Wiley; Volume: 89; Issue: 7 Linguagem: Inglês

10.1111/j.1755-3768.2009.01794.x

1755-3768

Johan H. Seland, Johannes R. Vingerling, C Augood, Graham Bentham, Usha Chakravarthy, P. deJong, Mati Rahu, G. Soubrane, Laura Tomazzoli, Fotis Topouzis, Astrid Fletcher,

Retinal Diseases and Treatments

Purpose: To determine the prevalence of visual impairment (VI) in populations 65 year or older from six European countries and describe the association with vision-related quality of life. VI was defined according to WHO as best corrected visual acuity 0,48 (World Health Organization (1992): International Statistical Classification of Diseases and Related Health Problems, 10th revised ed. Vol 1. Geneva). Methods: 4166 participants in The European Eye study, 65 years and older selected randomly from the general census in the participating centres, were interviewed for vision-related quality of life and underwent an eye exam including distance visual acuity, refraction and fundus photography. Results: The prevalence of VI rose with increasing age and more so in women. There was a pattern of a higher prevalence of VI in the Mediterranean countries compared to Northern European countries with the exception of Tallinn (Estonia) which had higher VI prevalence rates than the other north European centres. The prevalence of low vision was 3% or less in all centres. Blindness prevalence varied from 2% to less than half a per cent. Vision-related quality of life was strongly associated with visual acuity and the presence of bilateral age-related macular degeneration. Conclusion: The prevalence of visual impairment in the examined ageing European populations shows a definite increasing trend from north to south. An increasing prevalence of impaired visual function with increasing age is a phenomenon which has been documented in many worldwide studies (Leibowitz et al. 1980; Klein et al. 1991; Attebo et al. 1996; Klaver et al. 1998; Muñoz et al. 2000; van der Pols et al. 2000; Evans et al. 2002, 2004; Resnikoff et al. 2004). The considerable range of prevalences reported are attributed to a wide variation of causes and reflects a combination of demography, cultural traditions, presence of eye pathology combined with availability and quality of eye-health service as well as a number of poorly understood environmental and genetic factors. In a global perspective, poor refraction, cataract and glaucoma are the most important pathological causes of impaired vision in the aged population, while in Europe age-related macular degeneration (AMD) and cataract have been identified either singly or in a combination as important causes for visual impairment in older age groups. (Abou-Gareeb et al. 2001; Congdon et al.2004; Evans et al. 2004). Traditionally, visual function is assessed by measuring the corrected distance acuity with optimal contrast in the better eye, and sometimes this information is supplemented with visual fields. The universally accepted criteria for visual impairment are defined by WHO and consist of best corrected visual acuity 0,48. (WHO 1992) Visual impairment may be further categorised as low vision (<6/18–3/60/logMAR 0,48-1,3) and blind ( 1,3). The European Eye study (EUREYE) (Augood et al. 2004) was designed to study the prevalence of and risk factors for early and late AMD in seven European countries. In this article, we report the prevalence of visual impairment (VI) and its impact on quality of life in six of the original seven participating countries. The EUREYE study is a cross-sectional population-based study in persons aged 65 years and over in seven centres located from north to south Europe. The principal objective of the EUREYE study was to estimate the prevalence of AMD and investigate possible risk factors for AMD with a focus on sunlight and antioxidants. A detailed description of the EUREYE methods has been reported previously (Augood et al. 2004; 2006). In brief, participants were recruited from random sampling of the population aged over 65 years in the centres: Bergen (Norway), Tallinn (Estonia), Belfast (UK), Paris-Creteil (France), Verona (Italy) and Thessaloniki (Greece). Participants were interviewed by fieldworkers, underwent an eye examination including visual acuity measurement, lens and fundus photography and gave a blood sample. Study participants gave informed written consent. The study was approved by national ethical committees and met the criteria of the Helsinki declaration. Data were also collected from Spain (Alicante), but these were omitted from this report, because analysis of our data revealed a discrepancy in the manner of visual acuity recording between the Spanish centre and the remaining study sites. This was because of a difference in the interpretation of the visual acuity measurement protocol. The interview was conducted before the eye examination. Trained fieldworkers used structured questionnaires to collect data on smoking and alcohol use, brief medical history, dietary habits (food frequency questionnaire), outdoor occupational and leisure behaviour and vision-related quality of life. For quality of life, we used a shortened version of the National Eye Institute Visual Function Questionnaire-25 (NEIVFQ) (Mangione et al. 2001). Because the study population consisted of elderly people and because of the time constraints imposed by other data collection, we omitted eight items from the NEIVFQ 25. We omitted the sections on role difficulties (two questions) and dependency (three questions), because we were concerned that these questions which were developed in the American context might not be culturally appropriate in the European setting. We retained the item on worry about eye sight from the mental health subscale but excluded three others. All remaining 17 items were kept, administered and scored in accordance with previously published information. We did not consider that removal of these items would impact on the responsiveness of the NEIVFQ as a major review of the NEIVFQ has concluded that there was considerable redundancy within this instrument (Massof & Rubin 2001). The items in the NEIVFQ were translated and back-translated into the languages of the participating centres. Distance visual acuity was tested separately in each eye by optometrists or ophthalmologists using the EDTRS log MAR chart (Ferris et al. 1982). The testing distance was 4 m, and if a participant was unable to read 20 letters at this distance, the test was repeated at 1 m. Any participant who was unable to achieve 0.3 log MAR (30 letters equivalent to Snellen 6/12 or 20/40) in either eye, underwent automated or manual retinoscopy followed by refraction and recording of best corrected acuity. If vision was improved by refraction, participants were advised that they needed glasses or a new prescription for their spectacles. If vision was not improved, arrangements were made for follow-up examination with a collaborating ophthalmologist. All centres were equipped with a digital Topcon fundus camera (TRC-50EX; Topcon Corporation, Tokyo, Japan). Following pupillary dilation, for each eye, one photograph of the pupillary area was taken to show the status of the lens followed by two 35° nonsimultaneous stereoscopic colour fundus images, centred on the fovea. Fundus images were graded at a single reading centre (PTVMJ, JRV). Using the International Classification System for age-related maculopathy (ARM) (Bird et al. 1995) with five mutually exclusive grades: Grade 0: No early or late AMD. Grade 1: Soft distinct drusen (≥63 μm and <125 μm) only or pigmentary irregularities only. Grade 2: Soft indistinct (≥125 μm) or reticular drusen only or soft distinct drusen with pigmentary irregularities. Grade 3: Soft indistinct or reticular drusen with pigmentary irregularities. Grade 4: Neovascular-AMD (presence of any of the following: serous or haemorrhagic retinal or retinal pigment epithelial detachment, subretinal neovascular membrane, peri-retinal fibrous scar) or geographical atrophy, GA (well-demarcated area of retinal pigment atrophy with visible choroidal vessels). Any other fundus abnormality that appeared on the fundus and anterior segment images were graded by the Rotterdam Study graders and adjudicated by an experienced ophthalmologist for final diagnosis (JRV). The lens and fundus images of all those with visual impairment were closely reviewed (JRV) to ascertain the primary cause of vision impairment. We used the refracted visual acuity in the best eye and the following cut points and definitions: 6/6 (0,0 log MAR) or better to define normal vision, 0,0 ≥ 0,3) for very mild visual impairment, and 0,3 ≥ 0,48) for mild visual impairment, 0,48≥ 1,3) as low vision and 1,3 log Mar) as blind. For the purposes of comparison with other studies, we have used the term visual impairment to include both low vision and blindness. Scoring of the abbreviated NEIVFQ-25 was undertaken using the published algorithm for the VFQ-25 (Mangione et al. 2001) to generate the following nine subscales: general vision, worry over eye sight, ocular pain, near activities, distance activities, peripheral vision, vision-specific social functioning, colour vision and driving. Although administered, we did not analyse the data from the item on General Health, because we did not consider it relevant to vision. We used the Likert scoring method to convert the data from each subscale into a score from between 0 to 100 where 100 represents good quality of life. Statistical analysis was carried out using STATA 10 (StataCorp LP, 4905 Lakeway Dr. College Station, TX, USA). In analyses of visual acuity, we used age and sex standardisation using the study population as the standard (direct standardisation) for prevalence by centre. Sex standardisation was used to compare prevalence by age group (65–69, 70–74, 75+) and age standardisation for prevalence by sex. In multivariable models examining differences in prevalence of visual impairment by age, sex and country, we calculated prevalence ratios and 95% confidence intervals using Poisson regression with robust variance estimators. We used linear regression with robust standard errors to compare vision-related quality of life scores by categories of visual acuity and grade of AMD. Four thousand and one hundred and sixty-six people (1857 men and 2309 women) underwent visual acuity testing and photography. The average age of the study population was 73.2 years (SD = 5.6). Visual acuity levels varied by centre (Table 1). Over half the participants in Bergen, Belfast and Verona were classified as having normal vision in the better seeing eye compared to around a quarter in Thessaloniki. A substantial proportion of participants in each centre had acuity in the range 0.0–0.3); thus, over 90% were in the normal or very mild impairment group i.e. VA >6/12 (<log MAR 0.3). The prevalence of low vision was 3% or less in all centres. Blindness prevalence varied from 2% to less than half a per cent. In all centres, prevalence rates of visual impairment rose steeply with age with the highest rates observed in the 75+ year age group both for men and women (Table 2). In all centres except Belfast, visual impairment was higher in women compared to men, with an age and sex-standardised prevalence rate ratio varying from 1.15 to 3.45, lowest in Belfast and highest in Thessaloniki. In 107 people with vision impairment ( 0.48), the principal causes were AMD (n = 51, 47.8%) and cataract (n = 23, 21.5%). Other causes were as follows: diabetic retinopathy (n = 9, 8.4%), myopic macular degeneration (n = 6, 5.6%) and other conditions (n = 11, 10.3%) including glaucomatous optic neuropathy (n = 2), corneal abnormalities (n=2), chorioretinal scar not because of AMD (n = 3), pucker cellophane maculopathy (n = 2), coloboma posterior pole (n = 1) and macular hole (n = 1). In seven people, the cause of vision impairment could not be ascertained. The distribution of cataract cases by center and as a % of visual impairment was Bergen (1) 7.7%, Tallinn (14) 37.8%, Belfast (0), Paris (4) 26.7%, Verona (2) 11.1% and Thessaloniki (2) 12.5%. Vision-related quality of life data (QOL) were available in 4133 of the 4166 participants with visual acuity measurements. The section on driving was completed by 44.5% (n = 1839) of people who were either ex- or current drivers (22% of women and 72% of men). All scores showed highly significant trends according to category of visual acuity with lower scores indicating reduction in QOL in those with poorer vision (Table 3). For example, compared to those with the best vision (≥6/6), the mean score in the vision-specific social function scale was 38 points lower for those with VA <6/18–3/60 and 53 points lower for those with VA <3/60. The differences were smaller for those with less vision reduction, e.g. 12 points lower for those with VA <6/12–6/18 and only a 2.1 point reduction for those with very mild VI (<6/6–6/12) Vision-related quality of life was lowest in those with late AMD (grade 4 of the ARM grading scale (Table 4). People with late AMD had around a 20-point difference compared to those with early AMD for many of the scales. Scores were not related to the early AMD stages namely, grades 0–3 and were uniformly high (apart from the question on general vision and worry about eyesight) indicating minimal disruption to vision quality of life. The countries included in this study today all have a well-developed ophthalmological health service, and differences observed are probably caused by local factors – environmental, dietary or possibly genetic. The aged population of Estonia had a higher prevalence of visual impairment than the other Northern European countries. However, a third of the VI at this centre could be attributed to untreated cataract cases. A similar high prevalence of visual impairment was also observed in the Mediterranean countries (Tables 1 and 2), but here, the number of untreated cataract were less. The strong relationship between age and visual disability has been described in many previous publications and is confirmed in our study. Many studies have also documented the higher prevalence of visual impairment in women compared to men (Abou-Gareeb et al. 2001). Reduction in vision-related QOL caused by AMD was not substantial (Table 4) until the classifying (worse) eye fell into the category of late AMD (stage 4), and even then 58% reported able to drive a car, indicating that this disease usually affects the two eyes to different degrees. Comparing our results with previously published findings is of some value despite variability in protocol and design between studies of this nature. The Beaver Dam (Klein et al. 1991), The Blue Mountain (Attebo et al. 1996) and Rotterdam Studies (Klaver et al. 1998) all investigated the age group 65–75 and found that respectively 4.7%, 2.1% and 0.83% had visual acuity equal to or worse than 6/12 (between log Mar 0.3 and logMar1.0). The American Eye Disease Prevalence Research group pooled the results from eight population studies, based in the United States, Europe and Australia, to estimate the magnitude of visual impairment in United States (Congdon et al. 2004). They estimated a prevalence of best corrected VA <6/12 of 1.47% in 65–69 age group, 2.60% in 70–74, 5.03% in the 75–79 and 23.73% in 80+. We found that the prevalence of VA 0,3logMAR) was 5.9% (95% CI 2.8–9.0) in our over 65 years population. In the age group, over 75 years this rose to 12.6% (95% CI 5.2–19.9) and to 32.9% (95% CI 10.8–55.1) in those over 85 years. On considering only the 65–74 age group, 2.8% (95%CI 1.1–4.5) had VA 0,3). These results seem to be in the same range as those quoted by the US review. The Melbourne study (Taylor et al. 1997) which investigated a younger age group (60–69 years) found a fairly low prevalence of VA <6/12 i.e. 0,93%. A study in Britain using pinhole showed VA <6/12 ranging from 3,1% in the 65–74 years age group rising to 35,5% in the 85+ (van der Pols et al. 2000). On the other hand, there are reports from middle-income countries (e.g. Iran, a lower middle-income country) on visual impairment affecting some 20% of the population older than 60 years of age, and this was strongly linked to the level of education (Fotouhi et al. 2004). Racial differences have been described in countries with mixed populations. (Klaver et al. 1998), but we did not collect data on this in our study. In most industrialised countries, legislation sets a minimum visual acuity level for holding a driving licence to log Mar 0.3 (6/12) in the best eye. However, a quarter of our participants with VA <6/12 were continuing to drive, although their scores on the driving section of the vision-related quality of life questionnaire indicated that they were having difficulties. Our study also provides new data on the impact of poor vision and AMD on vision-related quality of life across a range of European populations following a common protocol. Considerable human and economic resources are allocated to implement rehabilitation services by most national governments. Prevention and treatment visual impairments have been shown to be a good public health investment both from a socio-economic point of view but more importantly from an individual quality of life point of view (Taylor et al. 2007). None of the authors have any proprietary interests relevant to this article to declare. EUREYE was supported by the European Commission Vth Framework (QLK6-CT-1999-02094). Additional funding for cameras was provided by the Macular Disease Society UK. M. Rahu was financed by the Estonian Ministry of Education and Science (target funding 01921112s02 & SF0940026s07).