- Email: [email protected]
Physical Activity and Age-related Macular Degeneration: A Systematic Literature Review and Meta-analysis
Accepted Manuscript Physical Activity and Age-related Macular Degeneration: A systematic literature review and meta-analysis Myra B. McGuinness, Jerome Le, Paul Mitchell, Bamini Gopinath, Ester Cerin, Nicole T.M. Saksens, Tina Schick, Carel B. Hoyng, Robyn H. Guymer, Robert P. Finger PII:
S0002-9394(17)30218-0
DOI:
10.1016/j.ajo.2017.05.016
Reference:
AJOPHT 10146
To appear in:
American Journal of Ophthalmology
Received Date: 5 January 2017 Revised Date:
13 May 2017
Accepted Date: 15 May 2017
Please cite this article as: McGuinness MB, Le J, Mitchell P, Gopinath B, Cerin E, Saksens NTM, Schick T, Hoyng CB, Guymer RH, Finger RP, Physical Activity and Age-related Macular Degeneration: A systematic literature review and meta-analysis, American Journal of Ophthalmology (2017), doi: 10.1016/j.ajo.2017.05.016. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Abstract Purpose: Age-related macular degeneration (AMD) is the main cause of irreversible severe vision loss in developed countries. It has been suggested that a healthy lifestyle may assist in delaying the onset and progression of AMD, however evidence for an association between
Design: Systematic review and meta-analysis
RI PT
physical activity and age-related macular degeneration (AMD) remains inconclusive.
Methods: Medline, EMBASE and Google Scholar were systematically searched for studies
SC
up to May 2015. Reference lists of published articles were hand searched and study authors were contacted to provide additional data. Those in the lowest category of activity in each
M AN U
study were compared with all other participants to assess the association between physical activity and both early and late AMD using random effects meta-analysis. Results: Nine studies (age range 30-97 years) were included in the meta-analysis. Physical activity was found to have a protective association with both early AMD (8 studies, n =
TE D
38,112, odds ratio (OR) 0.92 95% confidence interval (CI) 0.86-0.98) and late AMD (7 studies, n = 28,854, OR 0.59 95% CI 0.49-0.72). Conclusions: Physical activity is associated with lower odds of early and late AMD in
EP
Caucasian populations. These findings have important implications, reinforcing the public health message of staying active throughout life. However, further longitudinal studies are
AC C
required to confirm and further characterize a protective effect of physical activity on the onset and/or progression of AMD.
ACCEPTED MANUSCRIPT Physical Activity and Age-related Macular Degeneration: A systematic literature review and meta-analysis
Myra B. McGuinness,1 Jerome Le,1 Paul Mitchell2, Bamini Gopinath,2 Ester Cerin,3 Nicole T. M. Saksens,4 Tina Schick,5 Carel B Hoyng,4 Robyn H. Guymer,1 Robert P.
1
RI PT
Finger1,6 Centre for Eye Research Australia, Royal Victorian Eye and Ear Hospital & Ophthalmology,
Department of Surgery, University of Melbourne, Melbourne, Australia 2
Centre for Vision Research, Department of Ophthalmology and Westmead Millennium
Institute, University of Sydney, Sydney, Australia Australian Catholic University, Melbourne, Australia
4
Ophthalmology, Radboud University Medical Centre, Nijmegen, The Netherlands
5
Department of Ophthalmology, University Hospital of Cologne, Cologne, Germany
6
Department of Ophthalmology, University of Bonn, Bonn, Germany
Corresponding author:
University of Bonn Ernst-Abbe-Str. 2 D-53127 Bonn, Germany
TE D
Robert P. Finger Department of Ophthalmology
M AN U
Short title: Physical activity & AMD
SC
3
EP
Email: [email protected]
AC C
(W) +49 228 287 15505 (F) +49 228 287 11518
Supplemental Material available at AJO.com
1
ACCEPTED MANUSCRIPT
RI PT
Age-related macular degeneration (AMD) is the most common cause of irreversible severe vision loss among Caucasians in developed countries.1 The early, non-sight threatening stages of AMD affect up to 1 of 3 persons over the age of 70, of whom 10-20% progress to late-stage disease and are at risk of severe vision loss.2 Whilst smoking cessation and daily supplementation of high dose antioxidant vitamins and trace elements were reported to reduce the risk of progression from early asymptomatic disease to sight-threatening late stage AMD in up to one third of affected persons, there is currently no specific intervention to prevent AMD or to delay its progression.3, 4 Due to current demographic trends, the number of people living with AMD is projected to increase considerably.5, 6 Thus, there is a large and unmet need to identify modifiable risk factors and develop targeted interventions which prevent the development of AMD or delay its progression. AMD is a complex disease with genetic and environmental risk factors. Whilst there is solid evidence of genetic and some lifestyle risk factors, such as smoking, contributing to AMD, they only account for a limited proportion of the overall risk.7, 8 Thus there is likely to be undetermined environmental and/or lifestyle risk factors where modifications may be possible and potentially beneficial.
M AN U
SC
A growing body of evidence supports the positive impact of physical activity on successful aging, i.e. lowering morbidity and mortality, a better maintenance of physical and cognitive functions, avoidance of disability, and lowering biomarkers of aging.9-12 The current evidence on the effect of physical activity on age-related ocular disorders such as AMD suggests a protective effect.10-22 However, available evidence is inconclusive as past studies were conducted in differing populations, used differing AMD classification systems and activity was not assessed uniformly.9-12, 18 Many of these studies reported wide confidence intervals and thus there is a lack of certainty surrounding the accuracy of each estimate.
METHODS Literature Search
TE D
In the absence of disability, physical activity represents lifestyle choices that are modifiable for a majority of older persons, making it an ideal modifiable risk factor to assess further as currently available evidence is conflicting. Against this background we conducted a systematic review and meta-analysis of the available literature to better understand the association of physical activity and AMD among Caucasians.
AC C
EP
A systematic review of literature and meta-analysis was conducted. Our search strategy, selection of publications, and reporting of results were conducted in accordance with MOOSE (Meta-analysis of Observational Studies in Epidemiology) guidelines, and further details are provided in Supplemental Table 1 (available at AJO.com).23 In brief, Ovid Medline (In-Process & Other Non-Indexed Citations), Ovid Embase, and Google Scholar were searched by J.L. A search for articles with at least one diagnosis term (such as agerelated macular degeneration) and one exposure term (such as physical activity or lifestyle) was conducted without restrictions. Reference lists of papers reviewed in full were hand searched for additional articles. Conference abstracts were also considered in the review. Study Selection and eligibility criteria Full text articles were reviewed by J.L. and M.B.M. and included on the basis of pre-defined criteria,namely: i) a reported outcome of AMD; ii) inclusion of physical activity as a study variable/risk factor; and iii) reported odds ratio (OR) with measure of variance, or aggregate data which allowed the calculation of an OR. AMD differs in regards to its genetic makeup as well as phenotype and affects predominantly Caucasians.24 Subsequently all non-Caucasian populations were excluded. Where sufficient data were not reported in the publication, authors were contacted for more information. Authors involved with cohorts known to have collected relevant data were also contacted. 2
ACCEPTED MANUSCRIPT Data collection and risk of bias assessment
RI PT
Data from eligible studies to be included in the review were extracted independently by J.L. amd M.B.M. The risk of bias was assessed both qualitatively and numerically using the Newcastle Ottawa Scale.25 Two authors (J.L. and M.B.M.) independently reviewed data and graded articles obtained from the literature search to assess quality. To accommodate for various study designs, an adapted version of the Newcastle Ottawa Scale was used for each study type (i.e. case-control, cohort, and cross-sectional). A study was designated high quality if presenting sampling methods, a justification for sample size, adequate adjustments for potential confounders, and comparison of differences between respondents and nonrespondents and if using validated tools to define AMD and capture physical activity. The intra-class correlation coefficient was calculated to quantify inter-rater reliability on grading using a two-way mixed effects model.
SC
Data synthesis and analysis
M AN U
In order to overcome the variation in activity assessment between studies, participants were categorized as sedentary if they fell in the lowest activity category in a specific study, and as active otherwise.
TE D
In cases where participant level data were supplied by the author, logistic regression was used to calculated odds ratios adjusting for age, sex and smoking status for each study. Study-specific estimates were calculated from unadjusted aggregate level data only in cases where unadjusted cross-sectional estimates and participant level data were not available. Methods presented by Hamling et al. were employed for studies that reported ordinal levels of physical activity in order to obtain a single adjusted estimate for each level of AMD.26 These calculations were performed in Microsoft Excel using macros obtained from www.pnlee.co.uk/software.htm.
EP
Pooled odds ratios comparing AMD prevalence between sedentary and active individuals were estimated using meta-analysis with random effects models. Statistical heterogeneity between studies was quantified using the ܫଶ and ߬ ଶ statistics.27 Funnel plots were used to examine small-study effects. A sensitivity analysis to investigate potential bias was performed by excluding studies for which adjusted estimates could not be obtained.
AC C
Location of study (USA, Europe or Australia) and measure of activity (categorized as metabolic equivalent of task score, weekly days or hours of activity, or other activity measure) were considered to be potential sources of heterogeneity and were investigated using univariable random-effects meta-regression. Analyses were performed using Stata 13.1 (StataCorp LP, College Station, TX,
USA).
RESULTS
Literature Search The literature search yielded 620 articles of which 21 articles were considered potentially eligible on the basis of title and abstract (Figure 1). Of these, three were removed as more recent papers on the same study populations had been published, and three did not present relevant data. Full text papers were obtained for the remaining 15. Of the studies reviewed in full, one study was excluded based on the lack of objective verification of AMD status as it was based on participants’ self-report.12 Another study was 3
ACCEPTED MANUSCRIPT excluded as insufficient data were presented in the text and more detailed information could not be obtained.28 Five studies were excluded as they reported data for non-Caucasian populations (Figure 1). Patient level data and information on study design were obtained from the Melbourne Collaborative Cohort Study following personal contact with authors; analysis of this data has since been published.29 Nine studies were included in the review.
RI PT
All but two studies reported cross-sectional data. Baseline patient level data was therefore obtained for the Blue Mountains Eye Study cohort and baseline aggregate level case and exposure counts were obtained for the Beaver Dam Eye Study in order to calculate estimates of AMD prevalence for synthesis.10, 30 Adjusted estimates were obtained from the authors involved in European Genetic Database study as their published estimates are stratified by genetic subgroups.21
Study characteristics and risk of bias
M AN U
SC
All included studies reported having obtained ethics approval and adhered to the Declaration of Helsinki. Characteristics of included studies are summarized in Table 1. Year of publication ranged from 1992 to 2016 and a total of 40,879 participants were included with an age range from 21-97 years. Four of the studies were conducted in the United States10, 13, 16, 19 , three in Europe11, 21, 22 and two in Australia.29, 30 One study recruited female participants only,16 while the others included both sexes.
TE D
The Inter99 Study originated as a randomized trial on healthy lifestyle interventions, however only cross-sectional data from the baseline visit were examined in this study.11 Both the Eye Disease Case Control Study (EDCCS) and the European Genetic Database Study (EUGENDA) were conducted as case control studies,13, 21 while patient selection for the Beaver Dam, Blue Mountains and Tromsø studies was population-based.10, 22, 30 Participants in the Beaver Dam Offspring Study were invited to participate owing to at least one parent in the Beaver Dam Eye Study, and those in the Carotenoids in Age-related Eye Disease Study (CAREDS) were recruited from the Women’s Health Initiative Observational Study based on their reported intake of antioxidants.16, 19 The Melbourne Collaborative Cohort Study prospectively recruited volunteers from the community.29
AC C
EP
Participants in the Melbourne Collaborative Cohort Study underwent fundus photography a median of 11 years following assessment of physical activity, and CAREDS participants had retinal imaging an average of six years after lifestyle ascertainment. As the EDCCS was a case-control study, AMD status was ascertained prior to measurement of activity levels. Fundus photography and level of activity were captured concurrently in the remaining studies. In the studies that recruited participants prospectively, exclusion due to poor quality imaging was low, ranging from less than one percent to almost nine percent. On average, the participants in the Inter99 and Beaver Dam Offspring studies were considerably younger than those in the other studies, as seen in Table 1.11, 19 Estimates from each study were adjusted for age, sex and smoking to address bias from confounding apart from two; estimates from the Beaver Dam Offspring Study were adjusted for age and sex alone, and only unadjusted level aggregate data was available from the Beaver Dam Study. Estimates from the Tromsø Study and the EDCCS were adjusted for additional potential confounders.13, 22 Estimates and sample sizes for each study can be seen in Table 2. Quality grading was performed using the Newcastle Ottawa Scale by two graders and inter-rater reliability was good (intra-class correlation coefficient: ߩ = 0.77, 95% CI 0.270.94). Overall quality was deemed to be moderate for three papers, high for four papers and very high for two papers as seen in Supplemental Table 2 (available at AJO.com).
4
ACCEPTED MANUSCRIPT Assessment of Age-related Macular Degeneration (AMD)
RI PT
AMD status was ascertained by grading color fundus photographs in all nine studies. The classification of late AMD (geographic atrophy and neovascular AMD) was consistent across all studies while the definition of early AMD was more heterogeneous (Table 1). All studies followed established grading protocols for AMD with most using the Age-related Eye Disease Study Grading System (AREDSGS), the Wisconsin Age-Related Maculopathy Grading System (WARMGS) and the classification used in the International Age-Related Maculopathy Epidemiological Study (IARMES).4, 31, 32 The AREDS grading system is derivative of the WARMGS classification, both of which are comparable to the IARMES classification.4 Despite the differences in classification, AMD could be categorized into the distinct stages of early and late AMD for all studies.
SC
Assessment of Physical Activity
M AN U
All studies captured physical activity through self-report using a self-administered or interviewer administered questionnaires with variation in the assessment of type, intensity, amount and context of activity as seen Tables 1 and 2. Early AMD
Seven of the eight estimates from the contributing studies were suggestive of a protective association between physical activity and AMD. The pooled estimate is indicative of a small reduction in the odds of early AMD for persons with an active lifestyle compared to a sedentary lifestyle (n = 38,112, OR 0.92, 95% CI 0.86-0.98, p =0.013; Figure 2). All studyspecific estimates for early AMD were adjusted for age, gender and smoking.
TE D
Despite differences in study design, statistical heterogeneity was low ( ܫଶ =4.8%). Examination of the funnel plot revealed no evidence of bias due to small-study effects (Figure 3).
EP
Late AMD
AC C
Estimates from the Tromsø study were reported separately for males and females and both were included in the meta-analysis. Of all estimates included in the meta-analysis, three reported a statistically-significant protective association between physical activity and late AMD and the point estimates were less than unity for all except one. Pooling across studies, the odds of late AMD was estimated to be 41% lower for persons with an active lifestyle compared to those who were sedentary (n = 28,854, OR 0.59, 95% CI 0.49-0.72; p < 0.001, Figure 4). There was a low level of statistical heterogeneity ( ܫଶ = 23.5%) and the funnel plot showed no bias due to small-study effects (Figure 5).
Sensitivity analysis No multivariate adjusted estimate could be obtained for The Beaver Dam Eye Study. This elevates the risk of bias and therefore sensitivity analyses were conducted by excluding this study from the meta-analysis. Following the exclusion of this study, the pooled estimate for early AMD was similar but not as precise and inferences remained unchanged for late AMD (Table 3).
5
ACCEPTED MANUSCRIPT Meta-regression Neither differences in study location nor type of activity contributed substantially to the levels of statistical heterogeneity for early or late AMD (Table 4).
DISCUSSION
RI PT
In this meta-analysis we found a more active lifestyle to be independently associated with lower odds of both early and late AMD, with the effect being more pronounced for late AMD. This supports the evidence of a protective association of physical activity on incident AMD as well as AMD progression reported for two prospective cohort studies.10, 30 Our findings reinforce the public health message of staying active throughout life.
TE D
M AN U
SC
This meta-analysis is the first to assess the evidence for an association between physical activity and AMD. There is considerable evidence that regular physical activity leads to more healthy aging when compared to a sedentary lifestyle.33 Furthermore, engagement in regular exercise has been associated with lower mortality and greater longevity.34 A study excluded from this review based on AMD status being ascertained by self-report found a dose-reponse relationship between running and incidence of a self-reported diagnosis of AMD in a large cohort of runners.35 This lends further support to our findings. However, trying to assess the impact of physical activity on aging and/or development of disease is challenging as it is heavily confounded by lifestyle patterns. Persons who exercise tend to eat well, do not smoke, drink in moderation, are more health aware, and utilize health services for prevention and early detection of disease.11 Thus, all studies have to be interpreted with caution as this multitude of confounders potentially distorts results and is difficult to control for. The majority of studies included in this meta-analysis controlled for age and gender, and smoking. In an entirely Caucasian population and the absence of genetic data, these risk factors have consistently been shown to be the most influential in determining a person’s risk for AMD.36 However, additional confounders in the lifestyle-AMD relationship such as diet and education are also likely to be important and have not been controlled for in this analysis. We assume health outcomes such as body composition and markers of cardiovascular health to be post-exposure variables and as such should be treated as mediators rather than confounders of the relationship in question.37 Insufficient data are available to conduct a meta-analysis of this mediation effect.
AC C
EP
The amount of activity classified as an active lifestyle in this meta-analysis is as little as three hours of moderate to low intensity physical activity per week which highlights that already a small amount of physical activity might be sufficient in confering a beneficial effect. Moderate amounts of physical activity have repeatedly been shown to lower all-cause mortality. For example, a recent prospective cohort study demonstrated that 15 minutes per day or 90 minutes per week of moderate intensity exercise lowered mortality compared to those not exercising.38 An increase in physical activity by such a small amount should be achievable for a large proportion of the elderly and might be sufficient for a multiltude of positive longer term health outcomes.39 Regular exercise has been shown to increase antioxidant enzyme activity and increase resistance to oxidative stress.40 Oxidative stress has long been theorized to contribute to aging and is thought to be one of the key components in the pathogenesis of AMD.41 As one of several possible pathways influenced by physical activity, the increase in oxidative resistance conferred by regular physical activity may lead to a prevention of AMD or delay of its progression. The strengths of this study are a synthesis of all available evidence and utilization of existing best practice guidelines. There are two main limitations that affect interpretation of the results. First, self-reported physical activity was not captured uniformly across studies 6
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
and rarely the duration and intensity per session and/or context of exercise, i.e. occupational, competitive sport or leisure, was captured and accounted for. This shortcoming did not permit the accurate assessment of a dose-response relationship between accrued volume of physical activity and the odds of AMD. Yet, we note that studies which examined a wider range of physical activities as opposed to only leisure-time activity and attempted to more accurately quantify the total weekly amount of physical activity tended to report larger effects. The first study which employed an accelerometer to objectively capture physical activity also reported larger effects, in particular for late AMD.42 Using accelerometers in future studies may allow for a much more precise quantification of physical activity and thus delineation of its association with AMD. Unfortunately, these data are currently unavailable. Second, the data included in this meta-analysis were crosssectional, allowing only for assessment of an association not of causality. Assessing the relationship between late AMD and physical activity is difficult as late AMD may cause visual loss and thus reduce physical activity, causing a reversal of causality. However, this would not be the case for early AMD which causes next to no visual symptoms. Therefore, our final results may overestimate the protective effect of physical activity, but are unlikely to represent a spurious effect, as the final data set was very large. In contrast, survival bias resulting from the removal of less robust individuals from among the non-exercisers may have led to an underestimate of the protective effect of physical activity.43 Similarly, we found no evidence of publication bias among the included studies. In conclusion, physical activity is associated with lower odds of both early and late AMD in Caucasian populations which has potentially important public health implications. A small to moderate amount of physical activity might already be sufficient to confer this health benefit. Further longitudinal data, however, are required to confirm a protective effect of physical activity on the development and/or progression of AMD, and a more detailed and objective assessment of physical activity is needed to assess the exact dose-response relationship.
7
ACCEPTED MANUSCRIPT Acknowledgements a. Funding: Some funding support was provided the Lloyd and Kathleen Ansell Ophthalmology Foundation and the Mankiewicz-Zelkin Fellowship of the University of Melbourne to RPF. The Centre for Eye Research Australia receives Operational Infrastructure
RI PT
Support from the Victorian Government. The Beaver Dam Eye Study is funded by the National Institutes of Health and partly by the Research to Prevent Blindness.
a. Financial disclosures:
SC
No financial disclosures
b. Other acknowledgments:
M AN U
We are very grateful to Chelsea Myers and Ronald Klein for providing additional data
AC C
EP
TE D
from the Beaver Dam Eye Study
8
ACCEPTED MANUSCRIPT
References
AC C
EP
TE D
M AN U
SC
RI PT
1. Klein R. The prevalence of age-related eye diseases and visual impairment in aging: current estimates. Invest Ophthalmol Vis Sci 2013;54:ORSF5-ORSF13. 2. Klein R, Klein BEK, Linton KLP. Prevalence of age-related maculopathy: The Beaver Dam Eye Study. Ophthalmology 1992;99:933-943. 3. Thornton J, Edwards R, Mitchell P, Harrison RA, Buchan I, Kelly SP. Smoking and age-related macular degeneration: a review of association. Eye (London, England) 2005;19:935-944. 4. Age-Related Eye Disease Study Research Group. The Age-related Eye Disease Study system for classifying age-related macular degeneration from stereoscopic color fundus photographs: The Age-Related Eye Disease Study Report Number 6. Am J Ophthalmol 2001;132:668-681. 5. Pascolini D, Mariotti SP. Global estimates of visual impairment: 2010. The British journal of ophthalmology 2012;96:614-8. 6. Wong WL, Su X, Li X, et al. Global prevalence of age-related macular degeneration and disease burden projection for 2020 and 2040: a systematic review and meta-analysis. Lancet Glob Health 2014;2:e106-e116. 7. Klaver CCW, Kliffen M, van Duijn CM, et al. Genetic association of apolipoprotein E with agerelated macular degeneration. Am J Hum Genet 1998;63:200-206. 8. Seddon JM, Cote J, Page WF, Aggen SH, Neale MC. The US twin study of age-related macular degeneration - Relative roles of genetic and environmental influences. Arch Ophthalmol 2005;123:321-327. 9. Klein R, Lee KE, Gangnon RE, Klein BEK. Relation of smoking, drinking, and physical activity to changes in vision over a 20-year period: The Beaver Dam Eye Study. Ophthalmology 2014;121:12201228. 10. Knudtson MD, Klein R, Klein BEK. Physical activity and the 15-year cumulative incidence of age-related macular degeneration: the Beaver Dam Eye Study. The British journal of ophthalmology 2006;90:1461-1463. 11. Munch IC, Linneberg A, Larsen M. Precursors of age-related macular degeneration: Associations with physical activity, obesity, and serum lipids in the Inter99 Eye Study. Invest Ophthalmol Vis Sci 2013;54:3932-3940. 12. Williams PT. Prospective study of incident age-related macular degeneration in relation to vigorous physical activity during a 7-year follow-up. Invest Ophthalmol Vis Sci 2009;50:101-106. 13. Eye Disease Case-Control Study Group. Risk factors for neovascular age-related macular degeneration. Arch Ophthalmol 1992;110:1701-1708. 14. Song SJ, Youm DJ, Chang Y, Yu HG. Age-related macular degeneration in a screened South Korean population: prevalence, risk factors, and subtypes. Ophthalmic Epidemiol 2009;16:304-310. 15. Nidhi B, Mamatha BS, Padmaprabhu CA, Pallavi P, Vallikannan B. Dietary and lifestyle risk factors associated with age-related macular degeneration: A hospital based study. Indian J Ophthalmol 2013;61:722-727. 16. Mares JA, Voland RP, Sondel SA, et al. Healthy lifestyles related to subsequent prevalence of age-related macular degeneration. Arch Ophthalmol 2011;129:470-480. 17. Cho B-J, Heo JW, Kim TW, Ahn J, Chung H. Prevalence and risk factors of age-related macular degeneration in Korea: The Korea National Health and Nutrition Examination Survey 2010-2011. Invest Ophthalmol Vis Sci 2014;55:1101-1108. 18. Moon BG, Joe SG, Hwang J-u, Kim HK, Choe J, Yoon YH. Prevalence and risk factors of earlystage age-related macular degeneration in patients examined at a health promotion center in Korea. J Korean Med Sci 2012;27:537-541. 19. Klein R, Cruickshanks KJ, Nash SD, et al. The prevalence of age-related macular degeneration and associated risk factors. Arch Ophthalmol 2010;128:750-758. 9
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
20. Seddon JM, Cote J, Davis N, Rosner B. Progression of age-related macular degeneration: association with body mass index, waist circumference, and waist-hip ratio. Arch Ophthalmol 2003;121:785-792. 21. Saksens NT, Kersten E, Groenewoud JM, et al. Clinical characteristics of familial and sporadic age-related macular degeneration: differences and similarities. Invest Ophthalmol Vis Sci 2014;55:7085-92. 22. Erke MG, Bertelsen G, Peto T, Sjolie AK, Lindekleiv H, Njolstad I. Cardiovascular risk factors associated with age-related macular degeneration: the Tromso Study. Acta Ophthalmol 2014;92:662-669. 23. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology - A proposal for reporting. JAMA 2000;283:2008-2012. 24. Ding X, Patel M, Chan C-C. Molecular pathology of age-related macular degeneration. Prog Retin Eye Res 2009;28:1-18. 25. Wells G, Shea B, O'Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Canada, Accessed 2015. 26. Hamling J, Lee P, Weitkunat R, Ambuehl M. Facilitating meta-analyses by deriving relative effect and precision estimates for alternative comparisons from a set of estimates presented by exposure level or disease category. Stat Med 2008;27:954-970. 27. Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. Br Med J 2003;327:557-560. 28. Vinding T, Appleyard M, Nyboe J, Jensen G. Risk factor-analysis for atrophic and exudative age-related macular degeneration - an epidemiologic study of 1000 aged individuals. Acta Ophthalmologica 1992;70:66-72. 29. McGuinness MB, Karahalios A, Simpson JA, et al. Past physical activity and age-related macular degeneration: the Melbourne Collaborative Cohort Study. The British journal of ophthalmology 2016;100:1353-1358. 30. Gopinath B, Liew G, Burlutsky G, Mitchell P. Physical activity and the 15-year incidence of age-related macular degeneration. Invest Ophthalmol Vis Sci 2014;55:7799-7803. 31. Klein R, Davis MD, Magli YL, Segal P, Klein BEK, Hubbard L. The Wisconsin age-related maculopathy grading system. Ophthalmology 1991;98:1128-1134. 32. Bird AEC, Bressler NM, Bressler SB, et al. An international classification and grading system for age-related maculopathy and age-related macular degeneration. Surv Ophthalmol 1995;39:367374. 33. DiPietro L. Physical activity in aging: Changes in patterns and their relationship to health and function. J Gerontol A Biol Sci Med Sci 2001;56:13-22. 34. Paffenbarger Jr RS, Hyde R, Wing AL, Hsieh C-c. Physical activity, all-cause mortality, and longevity of college alumni. New Eng J Med 1986;314:605-613. 35. Williams PT. Prospective epidemiological cohort study of reduced risk for incident cataract with vigorous physical activity and cardiorespiratory fitness during a 7-year follow-up. Invest Ophthalmol Vis Sci 2009;50:95-100. 36. Smith W, Assink J, Klein R, et al. Risk factors for age-related macular degeneration: pooled findings from three continents. Ophthalmology 2001;108:697-704. 37. Bauman AE, Sallis JF, Dzewaltowski DA, Owen N. Toward a better understanding of the influences on physical activity: the role of determinants, correlates, causal variables, mediators, moderators, and confounders. Am J Prev Med 2002;23:5-14. 38. Wen CP, Wai JPM, Tsai MK, et al. Minimum amount of physical activity for reduced mortality and extended life expectancy: a prospective cohort study. Lancet 2011;378:1244-1253. 39. Nigam A, Juneau M. Survival benefit associated with low-level physical activity. Lancet 2011;378:1202-1203. 40. Radak Z, Taylor AW, Ohno H, Goto S. Adaptation to exercise-induced oxidative stress: From muscle to brain. Exerc Immunol Rev 2001;7:90-107.
10
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
41. Khandhadia S, Lotery A. Oxidation and age-related macular degeneration: insights from molecular biology. Expert Rev Mol Med 2010;12:28. 42. Loprinzi PD, Swenor BK, Ramulu PY. Age-related macular degeneration is associated with less physical activity among US adults: Cross-sectional study. PLoS One 2015;10. 43. McGuinness MB, Karahalios A, Kasza J, Guymer RH, Finger RP, Simpson JA. Survival Bias When Assessing Risk Factors for Age-Related Macular Degeneration: A Tutorial with Application to the Exposure of Smoking. Ophthalmic Epidemiol 2017;24:1-10.
11
ACCEPTED MANUSCRIPT Figure captions
FIGURE 1. Flowchart depicting the literature search of physical activity and age-related
RI PT
macular degeneration, 1992-2016.
FIGURE 2. Forest plot of meta-analysis of physical activity and the presence of early age-
SC
related macular degeneration (AMD), 1992-2016. Dashed line indicates overall estimate;
M AN U
bars indicate 95% confidence interval (CI).
FIGURE 3. Funnel plot for meta-analysis of physical activity and the presence of early agerelated macular degeneration, 1992-2016. Solid vertical line indicates pooled estimate,
TE D
dashed lines indicate pseudo-95% confidence interval.
FIGURE 4. Forest plot of meta-analysis of physical activity and the presence of late age-
EP
related macular degeneration (AMD), 1992-2016. Separate estimates were reported for males and females by Erke. EDCCSG = Eye Disease Case-Control Study Group; dashed
AC C
line = overall estimate; bars = 95% confidence interval (CI).
FIGURE 5. Funnel plot for meta-analysis of physical activity and the presence of late agerelated macular degeneration, 1992-2016. Solid vertical line indicates pooled estimate, dashed lines indicate pseudo-95% confidence interval.
12
ACCEPTED MANUSCRIPT
Age range (years)
Eye Disease Case Control Study 13 Group (1992)
Eye Disease Case Control Study (USA)
Case control
55-80
Tromsø Study (Norway)
Crosssectional
65-80
19
Klein (2010)
Knudtson (2006)
10
16
Mares (2011)
McGuinness (2016) 11
Munch (2013)
21
Saksens (2014)
29
Blue Mountains Eye Study (Australia) Beaver Dam Offspring Study (USA) Beaver Dam Eye Study (USA)
Cohort
45-97
Crosssectional
21-84
Cohort
43-86
Carotenoids in Age-related Eye Diseases (USA)
Crosssectional
Melbourne Collaborative Cohort Study (Australia)
Cohort
Inter99 Study (Denmark)
Crosssectional
European Genetic Database Study (Netherlands and Germany)
METs = metabolic equivalent of task score
55-74
40-69
Case-control
30-60
68-72
Measure physical activity
N/A
Likert scale of activity level
Covariate adjustment
Drusen > 125 µm no pigment abnormalities Drusen > 125 µm with pigment abnormalities
Hours of activity per week
Age, sex, smoking, clinic, biochemical measures, ocular variables and estrogen use Age, sex, smoking and blood pressure
METs per week
Age, sex and smoking
Drusen 250 µm with pigment abnormalities
Any weekly exercise
Age and sex
Drusen ≥ 250 µm with pigment abnormalities Drusen ≥ 125 µm no pigment abnormalities OR Drusen ≥ 63 µm with pigment abnormalities Drusen > 63 µm with or without pigment abnormalities Drusen > 63 µm no pigment abnormalities >9 drusen < 63 µm with pigment abnormalities OR Drusen ≥ 63 µm and <125 µm with or without pigment abnormalities
Days of activity per week
None
METs per week
Age and smoking (all female)
TE D
30
EP
Gopinath (2014)
AC C
22
Erke (2014)
Features of Early AMD
SC
Study group (country)
M AN U
Study design
First author (year)
RI PT
Table 1. Characteristics of Studies used in meta-analysis of physical activity and age-related macular degeneration (1992-2016)
Days of vigorous activity per week () Hours of activity per week Days of activity per week
Age, sex and smoking Age, sex and smoking
Age, sex and smoking
ACCEPTED MANUSCRIPT
Table 2. Estimates included in meta-analysis of physical activity and age-related macular degeneration, 1992-2016 Classification of sedentary behaviour Subjectively below average
Calculated from published ordinal estimates
19
Beaver Dam Offspring Study (Klein, 2010) 10
Beaver Dam Eye Study (Knudtson, 2006)
21
European Genetic Database (Saksens, 2014)
(0.39,0.77)
2,539
Early
0.97
(0.79,1.19)
Late (Males)
1.12
(0.55,2.29)
1,134
Late (Females)
0.73
(0.31,1.71)
2,522
Early
0.85
(0.56,1.28)
2,455
Late
0.80
(0.42,1.52)
862
2,806
Early
1.25
(0.81,1.92)
Activity on < 3 days
Unadjusted aggregate level data supplied by author
4,682
Early
0.93
(0.78,1.11)
3,892
Late
0.47
(0.22,0.93)
1,312
Early
0.76
(0.51,1.13)
21,008
Early
0.92
(0.86,0.99)
14,139
Late
0.72
(0.43,1.22)
825
Early
0.54
(0.32,0.91)
2,418
Early
0.95
(0.78,1.16)
Calculated from published ordinal estimates
AC C
11
0.55
As published
Vigorous activity on < 1 day per week
Inter99 (Munch, 2013)
(95% CI)
Exercise on < 1 day per week
≤ 2.2 METs per week
Melbourne Collaborative 29 Cohort Study (McGuinness, 2016)
Calculated from patient level data supplied by author
EP
Carotenoids in Age-related 16 Eye Disease Study (Mares, 2011)
< 198 METs per week
TE D
30
Blue Mountains Eye Study (Gopinath, 2014)
As published
SC
< 1 hour vigorous activity or less vigorous activity on < 2 days
OR
1,036
AMD stage of cases Late
n
RI PT
22
Tromsø (Erke, 2014)
Source of OR estimate
M AN U
Study (Author, Year) Eye Disease Case-control 13 Study (EDCCSG, 1992)
Calculated from patient level data supplied by author
< 2 hours activity per week
Calculated from published ordinal estimates
Exercise on < 1 day per week
Adjusted estimates provided by author
0.50 (0.42,0.60) 2,668 Late EDCCSG = Eye Disease Case-control Study Group; METs = metabolic equivalent of task score; n = number of participants in model including controls; OR = odds ratio.
ACCEPTED MANUSCRIPT Table 3. Summary of pooled estimates for meta-analyses of physical activity and age-related macular degeneration, primary and sensitivity analyses AMD Status
Primary analysis
Sensitivity analysis
ࡱ
ࡼ
OR
(95% CI)
ࡵ (%)
ࡱ
ࡼ
OR
(95% CI)
ࡵ (%)
Early
8
35,695
0.92
(0.86, 0.98)
4.8
7
31,013
0.92
(0.84, 1.01)
18.3
Late
7
23,519
0.59
(0.49, 0.72)
23.5
6
15,176
0.61
(0.49, 0.77)
34.5
AC C
EP
TE D
M AN U
SC
RI PT
CI = confidence interval; ݊ா = number of estimates; ݊ = number of participants; OR = odds ratio; ܫଶ = measure of heterogeneity
ACCEPTED MANUSCRIPT
Table 4. Univariable meta-regression of meta-analysis of physical activity and age-related macular degeneration, 1992 to 2016
Late
OR
(95% CI)
Meta-analysis estimate Study location USA Europe Australia Measure of activity Weekly activity (hours or days) METs Other
8
0.92
(0.86,0.98)
4.8
3 3 2
0.94 0.88 0.92
(0.77,1.16) (0.70,1.11) (0.86,0.98)
28.1 54.3 0.0
5 2 1
0.92 0.80 1.25
(0.86,0.99) (0.60,1.07) (0.81,1.93)
7
0.59
2 3 2 5 1 1
a
࣎
Meta-regression Ratio of (95% CI) ORs
0.0006
(0.71,1.36) (0.75,1.28)
0.0000
8.9 0.0
0.0007 0.0000
1.00 0.87 1.36
(0.59,1.28) (0.77,2,40)
0.0000
(0.49,0.72)
23.5
0.0158
0.53 0.68 0.75
(0.39,0.73) (0.40,1.14) (0.50,1.13)
0.0 61.4 0.0
0.0000 0.1336 0.0000
1.00 1.17 1.43
(0.52,2.63) (0.56,3.68)
0.0420
0.62 0.80 0.55
(0.46,0.83) (0.42,1.52) (0.39,0.77)
38.7
0.0428
1.00 1.30 0.89
(0.40,4.17) (0.37,2.13)
0.0459
SC
1.00 0.98 0.98
AMD = age-related macular degeneration; METs = metabolic equivalent of task score; OR = odds ratio; CI = confidence Interval a No measure of ܫଶ is possible when only one study estimate is available
AC C
࣎
0.0109 0.0208 0.0000
M AN U
Meta-analysis estimate Study location USA Europe Australia Measure of activity Weekly activity (hours or days) METs Other
ࡵ (%)
RI PT
Number of estimates
TE D
Early
Study Characteristic
EP
AMD Status
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0002-9394(17)30218-0
DOI:
10.1016/j.ajo.2017.05.016
Reference:
AJOPHT 10146
To appear in:
American Journal of Ophthalmology
Received Date: 5 January 2017 Revised Date:
13 May 2017
Accepted Date: 15 May 2017
Please cite this article as: McGuinness MB, Le J, Mitchell P, Gopinath B, Cerin E, Saksens NTM, Schick T, Hoyng CB, Guymer RH, Finger RP, Physical Activity and Age-related Macular Degeneration: A systematic literature review and meta-analysis, American Journal of Ophthalmology (2017), doi: 10.1016/j.ajo.2017.05.016. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Abstract Purpose: Age-related macular degeneration (AMD) is the main cause of irreversible severe vision loss in developed countries. It has been suggested that a healthy lifestyle may assist in delaying the onset and progression of AMD, however evidence for an association between
Design: Systematic review and meta-analysis
RI PT
physical activity and age-related macular degeneration (AMD) remains inconclusive.
Methods: Medline, EMBASE and Google Scholar were systematically searched for studies
SC
up to May 2015. Reference lists of published articles were hand searched and study authors were contacted to provide additional data. Those in the lowest category of activity in each
M AN U
study were compared with all other participants to assess the association between physical activity and both early and late AMD using random effects meta-analysis. Results: Nine studies (age range 30-97 years) were included in the meta-analysis. Physical activity was found to have a protective association with both early AMD (8 studies, n =
TE D
38,112, odds ratio (OR) 0.92 95% confidence interval (CI) 0.86-0.98) and late AMD (7 studies, n = 28,854, OR 0.59 95% CI 0.49-0.72). Conclusions: Physical activity is associated with lower odds of early and late AMD in
EP
Caucasian populations. These findings have important implications, reinforcing the public health message of staying active throughout life. However, further longitudinal studies are
AC C
required to confirm and further characterize a protective effect of physical activity on the onset and/or progression of AMD.
ACCEPTED MANUSCRIPT Physical Activity and Age-related Macular Degeneration: A systematic literature review and meta-analysis
Myra B. McGuinness,1 Jerome Le,1 Paul Mitchell2, Bamini Gopinath,2 Ester Cerin,3 Nicole T. M. Saksens,4 Tina Schick,5 Carel B Hoyng,4 Robyn H. Guymer,1 Robert P.
1
RI PT
Finger1,6 Centre for Eye Research Australia, Royal Victorian Eye and Ear Hospital & Ophthalmology,
Department of Surgery, University of Melbourne, Melbourne, Australia 2
Centre for Vision Research, Department of Ophthalmology and Westmead Millennium
Institute, University of Sydney, Sydney, Australia Australian Catholic University, Melbourne, Australia
4
Ophthalmology, Radboud University Medical Centre, Nijmegen, The Netherlands
5
Department of Ophthalmology, University Hospital of Cologne, Cologne, Germany
6
Department of Ophthalmology, University of Bonn, Bonn, Germany
Corresponding author:
University of Bonn Ernst-Abbe-Str. 2 D-53127 Bonn, Germany
TE D
Robert P. Finger Department of Ophthalmology
M AN U
Short title: Physical activity & AMD
SC
3
EP
Email: [email protected]
AC C
(W) +49 228 287 15505 (F) +49 228 287 11518
Supplemental Material available at AJO.com
1
ACCEPTED MANUSCRIPT
RI PT
Age-related macular degeneration (AMD) is the most common cause of irreversible severe vision loss among Caucasians in developed countries.1 The early, non-sight threatening stages of AMD affect up to 1 of 3 persons over the age of 70, of whom 10-20% progress to late-stage disease and are at risk of severe vision loss.2 Whilst smoking cessation and daily supplementation of high dose antioxidant vitamins and trace elements were reported to reduce the risk of progression from early asymptomatic disease to sight-threatening late stage AMD in up to one third of affected persons, there is currently no specific intervention to prevent AMD or to delay its progression.3, 4 Due to current demographic trends, the number of people living with AMD is projected to increase considerably.5, 6 Thus, there is a large and unmet need to identify modifiable risk factors and develop targeted interventions which prevent the development of AMD or delay its progression. AMD is a complex disease with genetic and environmental risk factors. Whilst there is solid evidence of genetic and some lifestyle risk factors, such as smoking, contributing to AMD, they only account for a limited proportion of the overall risk.7, 8 Thus there is likely to be undetermined environmental and/or lifestyle risk factors where modifications may be possible and potentially beneficial.
M AN U
SC
A growing body of evidence supports the positive impact of physical activity on successful aging, i.e. lowering morbidity and mortality, a better maintenance of physical and cognitive functions, avoidance of disability, and lowering biomarkers of aging.9-12 The current evidence on the effect of physical activity on age-related ocular disorders such as AMD suggests a protective effect.10-22 However, available evidence is inconclusive as past studies were conducted in differing populations, used differing AMD classification systems and activity was not assessed uniformly.9-12, 18 Many of these studies reported wide confidence intervals and thus there is a lack of certainty surrounding the accuracy of each estimate.
METHODS Literature Search
TE D
In the absence of disability, physical activity represents lifestyle choices that are modifiable for a majority of older persons, making it an ideal modifiable risk factor to assess further as currently available evidence is conflicting. Against this background we conducted a systematic review and meta-analysis of the available literature to better understand the association of physical activity and AMD among Caucasians.
AC C
EP
A systematic review of literature and meta-analysis was conducted. Our search strategy, selection of publications, and reporting of results were conducted in accordance with MOOSE (Meta-analysis of Observational Studies in Epidemiology) guidelines, and further details are provided in Supplemental Table 1 (available at AJO.com).23 In brief, Ovid Medline (In-Process & Other Non-Indexed Citations), Ovid Embase, and Google Scholar were searched by J.L. A search for articles with at least one diagnosis term (such as agerelated macular degeneration) and one exposure term (such as physical activity or lifestyle) was conducted without restrictions. Reference lists of papers reviewed in full were hand searched for additional articles. Conference abstracts were also considered in the review. Study Selection and eligibility criteria Full text articles were reviewed by J.L. and M.B.M. and included on the basis of pre-defined criteria,namely: i) a reported outcome of AMD; ii) inclusion of physical activity as a study variable/risk factor; and iii) reported odds ratio (OR) with measure of variance, or aggregate data which allowed the calculation of an OR. AMD differs in regards to its genetic makeup as well as phenotype and affects predominantly Caucasians.24 Subsequently all non-Caucasian populations were excluded. Where sufficient data were not reported in the publication, authors were contacted for more information. Authors involved with cohorts known to have collected relevant data were also contacted. 2
ACCEPTED MANUSCRIPT Data collection and risk of bias assessment
RI PT
Data from eligible studies to be included in the review were extracted independently by J.L. amd M.B.M. The risk of bias was assessed both qualitatively and numerically using the Newcastle Ottawa Scale.25 Two authors (J.L. and M.B.M.) independently reviewed data and graded articles obtained from the literature search to assess quality. To accommodate for various study designs, an adapted version of the Newcastle Ottawa Scale was used for each study type (i.e. case-control, cohort, and cross-sectional). A study was designated high quality if presenting sampling methods, a justification for sample size, adequate adjustments for potential confounders, and comparison of differences between respondents and nonrespondents and if using validated tools to define AMD and capture physical activity. The intra-class correlation coefficient was calculated to quantify inter-rater reliability on grading using a two-way mixed effects model.
SC
Data synthesis and analysis
M AN U
In order to overcome the variation in activity assessment between studies, participants were categorized as sedentary if they fell in the lowest activity category in a specific study, and as active otherwise.
TE D
In cases where participant level data were supplied by the author, logistic regression was used to calculated odds ratios adjusting for age, sex and smoking status for each study. Study-specific estimates were calculated from unadjusted aggregate level data only in cases where unadjusted cross-sectional estimates and participant level data were not available. Methods presented by Hamling et al. were employed for studies that reported ordinal levels of physical activity in order to obtain a single adjusted estimate for each level of AMD.26 These calculations were performed in Microsoft Excel using macros obtained from www.pnlee.co.uk/software.htm.
EP
Pooled odds ratios comparing AMD prevalence between sedentary and active individuals were estimated using meta-analysis with random effects models. Statistical heterogeneity between studies was quantified using the ܫଶ and ߬ ଶ statistics.27 Funnel plots were used to examine small-study effects. A sensitivity analysis to investigate potential bias was performed by excluding studies for which adjusted estimates could not be obtained.
AC C
Location of study (USA, Europe or Australia) and measure of activity (categorized as metabolic equivalent of task score, weekly days or hours of activity, or other activity measure) were considered to be potential sources of heterogeneity and were investigated using univariable random-effects meta-regression. Analyses were performed using Stata 13.1 (StataCorp LP, College Station, TX,
USA).
RESULTS
Literature Search The literature search yielded 620 articles of which 21 articles were considered potentially eligible on the basis of title and abstract (Figure 1). Of these, three were removed as more recent papers on the same study populations had been published, and three did not present relevant data. Full text papers were obtained for the remaining 15. Of the studies reviewed in full, one study was excluded based on the lack of objective verification of AMD status as it was based on participants’ self-report.12 Another study was 3
ACCEPTED MANUSCRIPT excluded as insufficient data were presented in the text and more detailed information could not be obtained.28 Five studies were excluded as they reported data for non-Caucasian populations (Figure 1). Patient level data and information on study design were obtained from the Melbourne Collaborative Cohort Study following personal contact with authors; analysis of this data has since been published.29 Nine studies were included in the review.
RI PT
All but two studies reported cross-sectional data. Baseline patient level data was therefore obtained for the Blue Mountains Eye Study cohort and baseline aggregate level case and exposure counts were obtained for the Beaver Dam Eye Study in order to calculate estimates of AMD prevalence for synthesis.10, 30 Adjusted estimates were obtained from the authors involved in European Genetic Database study as their published estimates are stratified by genetic subgroups.21
Study characteristics and risk of bias
M AN U
SC
All included studies reported having obtained ethics approval and adhered to the Declaration of Helsinki. Characteristics of included studies are summarized in Table 1. Year of publication ranged from 1992 to 2016 and a total of 40,879 participants were included with an age range from 21-97 years. Four of the studies were conducted in the United States10, 13, 16, 19 , three in Europe11, 21, 22 and two in Australia.29, 30 One study recruited female participants only,16 while the others included both sexes.
TE D
The Inter99 Study originated as a randomized trial on healthy lifestyle interventions, however only cross-sectional data from the baseline visit were examined in this study.11 Both the Eye Disease Case Control Study (EDCCS) and the European Genetic Database Study (EUGENDA) were conducted as case control studies,13, 21 while patient selection for the Beaver Dam, Blue Mountains and Tromsø studies was population-based.10, 22, 30 Participants in the Beaver Dam Offspring Study were invited to participate owing to at least one parent in the Beaver Dam Eye Study, and those in the Carotenoids in Age-related Eye Disease Study (CAREDS) were recruited from the Women’s Health Initiative Observational Study based on their reported intake of antioxidants.16, 19 The Melbourne Collaborative Cohort Study prospectively recruited volunteers from the community.29
AC C
EP
Participants in the Melbourne Collaborative Cohort Study underwent fundus photography a median of 11 years following assessment of physical activity, and CAREDS participants had retinal imaging an average of six years after lifestyle ascertainment. As the EDCCS was a case-control study, AMD status was ascertained prior to measurement of activity levels. Fundus photography and level of activity were captured concurrently in the remaining studies. In the studies that recruited participants prospectively, exclusion due to poor quality imaging was low, ranging from less than one percent to almost nine percent. On average, the participants in the Inter99 and Beaver Dam Offspring studies were considerably younger than those in the other studies, as seen in Table 1.11, 19 Estimates from each study were adjusted for age, sex and smoking to address bias from confounding apart from two; estimates from the Beaver Dam Offspring Study were adjusted for age and sex alone, and only unadjusted level aggregate data was available from the Beaver Dam Study. Estimates from the Tromsø Study and the EDCCS were adjusted for additional potential confounders.13, 22 Estimates and sample sizes for each study can be seen in Table 2. Quality grading was performed using the Newcastle Ottawa Scale by two graders and inter-rater reliability was good (intra-class correlation coefficient: ߩ = 0.77, 95% CI 0.270.94). Overall quality was deemed to be moderate for three papers, high for four papers and very high for two papers as seen in Supplemental Table 2 (available at AJO.com).
4
ACCEPTED MANUSCRIPT Assessment of Age-related Macular Degeneration (AMD)
RI PT
AMD status was ascertained by grading color fundus photographs in all nine studies. The classification of late AMD (geographic atrophy and neovascular AMD) was consistent across all studies while the definition of early AMD was more heterogeneous (Table 1). All studies followed established grading protocols for AMD with most using the Age-related Eye Disease Study Grading System (AREDSGS), the Wisconsin Age-Related Maculopathy Grading System (WARMGS) and the classification used in the International Age-Related Maculopathy Epidemiological Study (IARMES).4, 31, 32 The AREDS grading system is derivative of the WARMGS classification, both of which are comparable to the IARMES classification.4 Despite the differences in classification, AMD could be categorized into the distinct stages of early and late AMD for all studies.
SC
Assessment of Physical Activity
M AN U
All studies captured physical activity through self-report using a self-administered or interviewer administered questionnaires with variation in the assessment of type, intensity, amount and context of activity as seen Tables 1 and 2. Early AMD
Seven of the eight estimates from the contributing studies were suggestive of a protective association between physical activity and AMD. The pooled estimate is indicative of a small reduction in the odds of early AMD for persons with an active lifestyle compared to a sedentary lifestyle (n = 38,112, OR 0.92, 95% CI 0.86-0.98, p =0.013; Figure 2). All studyspecific estimates for early AMD were adjusted for age, gender and smoking.
TE D
Despite differences in study design, statistical heterogeneity was low ( ܫଶ =4.8%). Examination of the funnel plot revealed no evidence of bias due to small-study effects (Figure 3).
EP
Late AMD
AC C
Estimates from the Tromsø study were reported separately for males and females and both were included in the meta-analysis. Of all estimates included in the meta-analysis, three reported a statistically-significant protective association between physical activity and late AMD and the point estimates were less than unity for all except one. Pooling across studies, the odds of late AMD was estimated to be 41% lower for persons with an active lifestyle compared to those who were sedentary (n = 28,854, OR 0.59, 95% CI 0.49-0.72; p < 0.001, Figure 4). There was a low level of statistical heterogeneity ( ܫଶ = 23.5%) and the funnel plot showed no bias due to small-study effects (Figure 5).
Sensitivity analysis No multivariate adjusted estimate could be obtained for The Beaver Dam Eye Study. This elevates the risk of bias and therefore sensitivity analyses were conducted by excluding this study from the meta-analysis. Following the exclusion of this study, the pooled estimate for early AMD was similar but not as precise and inferences remained unchanged for late AMD (Table 3).
5
ACCEPTED MANUSCRIPT Meta-regression Neither differences in study location nor type of activity contributed substantially to the levels of statistical heterogeneity for early or late AMD (Table 4).
DISCUSSION
RI PT
In this meta-analysis we found a more active lifestyle to be independently associated with lower odds of both early and late AMD, with the effect being more pronounced for late AMD. This supports the evidence of a protective association of physical activity on incident AMD as well as AMD progression reported for two prospective cohort studies.10, 30 Our findings reinforce the public health message of staying active throughout life.
TE D
M AN U
SC
This meta-analysis is the first to assess the evidence for an association between physical activity and AMD. There is considerable evidence that regular physical activity leads to more healthy aging when compared to a sedentary lifestyle.33 Furthermore, engagement in regular exercise has been associated with lower mortality and greater longevity.34 A study excluded from this review based on AMD status being ascertained by self-report found a dose-reponse relationship between running and incidence of a self-reported diagnosis of AMD in a large cohort of runners.35 This lends further support to our findings. However, trying to assess the impact of physical activity on aging and/or development of disease is challenging as it is heavily confounded by lifestyle patterns. Persons who exercise tend to eat well, do not smoke, drink in moderation, are more health aware, and utilize health services for prevention and early detection of disease.11 Thus, all studies have to be interpreted with caution as this multitude of confounders potentially distorts results and is difficult to control for. The majority of studies included in this meta-analysis controlled for age and gender, and smoking. In an entirely Caucasian population and the absence of genetic data, these risk factors have consistently been shown to be the most influential in determining a person’s risk for AMD.36 However, additional confounders in the lifestyle-AMD relationship such as diet and education are also likely to be important and have not been controlled for in this analysis. We assume health outcomes such as body composition and markers of cardiovascular health to be post-exposure variables and as such should be treated as mediators rather than confounders of the relationship in question.37 Insufficient data are available to conduct a meta-analysis of this mediation effect.
AC C
EP
The amount of activity classified as an active lifestyle in this meta-analysis is as little as three hours of moderate to low intensity physical activity per week which highlights that already a small amount of physical activity might be sufficient in confering a beneficial effect. Moderate amounts of physical activity have repeatedly been shown to lower all-cause mortality. For example, a recent prospective cohort study demonstrated that 15 minutes per day or 90 minutes per week of moderate intensity exercise lowered mortality compared to those not exercising.38 An increase in physical activity by such a small amount should be achievable for a large proportion of the elderly and might be sufficient for a multiltude of positive longer term health outcomes.39 Regular exercise has been shown to increase antioxidant enzyme activity and increase resistance to oxidative stress.40 Oxidative stress has long been theorized to contribute to aging and is thought to be one of the key components in the pathogenesis of AMD.41 As one of several possible pathways influenced by physical activity, the increase in oxidative resistance conferred by regular physical activity may lead to a prevention of AMD or delay of its progression. The strengths of this study are a synthesis of all available evidence and utilization of existing best practice guidelines. There are two main limitations that affect interpretation of the results. First, self-reported physical activity was not captured uniformly across studies 6
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
and rarely the duration and intensity per session and/or context of exercise, i.e. occupational, competitive sport or leisure, was captured and accounted for. This shortcoming did not permit the accurate assessment of a dose-response relationship between accrued volume of physical activity and the odds of AMD. Yet, we note that studies which examined a wider range of physical activities as opposed to only leisure-time activity and attempted to more accurately quantify the total weekly amount of physical activity tended to report larger effects. The first study which employed an accelerometer to objectively capture physical activity also reported larger effects, in particular for late AMD.42 Using accelerometers in future studies may allow for a much more precise quantification of physical activity and thus delineation of its association with AMD. Unfortunately, these data are currently unavailable. Second, the data included in this meta-analysis were crosssectional, allowing only for assessment of an association not of causality. Assessing the relationship between late AMD and physical activity is difficult as late AMD may cause visual loss and thus reduce physical activity, causing a reversal of causality. However, this would not be the case for early AMD which causes next to no visual symptoms. Therefore, our final results may overestimate the protective effect of physical activity, but are unlikely to represent a spurious effect, as the final data set was very large. In contrast, survival bias resulting from the removal of less robust individuals from among the non-exercisers may have led to an underestimate of the protective effect of physical activity.43 Similarly, we found no evidence of publication bias among the included studies. In conclusion, physical activity is associated with lower odds of both early and late AMD in Caucasian populations which has potentially important public health implications. A small to moderate amount of physical activity might already be sufficient to confer this health benefit. Further longitudinal data, however, are required to confirm a protective effect of physical activity on the development and/or progression of AMD, and a more detailed and objective assessment of physical activity is needed to assess the exact dose-response relationship.
7
ACCEPTED MANUSCRIPT Acknowledgements a. Funding: Some funding support was provided the Lloyd and Kathleen Ansell Ophthalmology Foundation and the Mankiewicz-Zelkin Fellowship of the University of Melbourne to RPF. The Centre for Eye Research Australia receives Operational Infrastructure
RI PT
Support from the Victorian Government. The Beaver Dam Eye Study is funded by the National Institutes of Health and partly by the Research to Prevent Blindness.
a. Financial disclosures:
SC
No financial disclosures
b. Other acknowledgments:
M AN U
We are very grateful to Chelsea Myers and Ronald Klein for providing additional data
AC C
EP
TE D
from the Beaver Dam Eye Study
8
ACCEPTED MANUSCRIPT
References
AC C
EP
TE D
M AN U
SC
RI PT
1. Klein R. The prevalence of age-related eye diseases and visual impairment in aging: current estimates. Invest Ophthalmol Vis Sci 2013;54:ORSF5-ORSF13. 2. Klein R, Klein BEK, Linton KLP. Prevalence of age-related maculopathy: The Beaver Dam Eye Study. Ophthalmology 1992;99:933-943. 3. Thornton J, Edwards R, Mitchell P, Harrison RA, Buchan I, Kelly SP. Smoking and age-related macular degeneration: a review of association. Eye (London, England) 2005;19:935-944. 4. Age-Related Eye Disease Study Research Group. The Age-related Eye Disease Study system for classifying age-related macular degeneration from stereoscopic color fundus photographs: The Age-Related Eye Disease Study Report Number 6. Am J Ophthalmol 2001;132:668-681. 5. Pascolini D, Mariotti SP. Global estimates of visual impairment: 2010. The British journal of ophthalmology 2012;96:614-8. 6. Wong WL, Su X, Li X, et al. Global prevalence of age-related macular degeneration and disease burden projection for 2020 and 2040: a systematic review and meta-analysis. Lancet Glob Health 2014;2:e106-e116. 7. Klaver CCW, Kliffen M, van Duijn CM, et al. Genetic association of apolipoprotein E with agerelated macular degeneration. Am J Hum Genet 1998;63:200-206. 8. Seddon JM, Cote J, Page WF, Aggen SH, Neale MC. The US twin study of age-related macular degeneration - Relative roles of genetic and environmental influences. Arch Ophthalmol 2005;123:321-327. 9. Klein R, Lee KE, Gangnon RE, Klein BEK. Relation of smoking, drinking, and physical activity to changes in vision over a 20-year period: The Beaver Dam Eye Study. Ophthalmology 2014;121:12201228. 10. Knudtson MD, Klein R, Klein BEK. Physical activity and the 15-year cumulative incidence of age-related macular degeneration: the Beaver Dam Eye Study. The British journal of ophthalmology 2006;90:1461-1463. 11. Munch IC, Linneberg A, Larsen M. Precursors of age-related macular degeneration: Associations with physical activity, obesity, and serum lipids in the Inter99 Eye Study. Invest Ophthalmol Vis Sci 2013;54:3932-3940. 12. Williams PT. Prospective study of incident age-related macular degeneration in relation to vigorous physical activity during a 7-year follow-up. Invest Ophthalmol Vis Sci 2009;50:101-106. 13. Eye Disease Case-Control Study Group. Risk factors for neovascular age-related macular degeneration. Arch Ophthalmol 1992;110:1701-1708. 14. Song SJ, Youm DJ, Chang Y, Yu HG. Age-related macular degeneration in a screened South Korean population: prevalence, risk factors, and subtypes. Ophthalmic Epidemiol 2009;16:304-310. 15. Nidhi B, Mamatha BS, Padmaprabhu CA, Pallavi P, Vallikannan B. Dietary and lifestyle risk factors associated with age-related macular degeneration: A hospital based study. Indian J Ophthalmol 2013;61:722-727. 16. Mares JA, Voland RP, Sondel SA, et al. Healthy lifestyles related to subsequent prevalence of age-related macular degeneration. Arch Ophthalmol 2011;129:470-480. 17. Cho B-J, Heo JW, Kim TW, Ahn J, Chung H. Prevalence and risk factors of age-related macular degeneration in Korea: The Korea National Health and Nutrition Examination Survey 2010-2011. Invest Ophthalmol Vis Sci 2014;55:1101-1108. 18. Moon BG, Joe SG, Hwang J-u, Kim HK, Choe J, Yoon YH. Prevalence and risk factors of earlystage age-related macular degeneration in patients examined at a health promotion center in Korea. J Korean Med Sci 2012;27:537-541. 19. Klein R, Cruickshanks KJ, Nash SD, et al. The prevalence of age-related macular degeneration and associated risk factors. Arch Ophthalmol 2010;128:750-758. 9
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
20. Seddon JM, Cote J, Davis N, Rosner B. Progression of age-related macular degeneration: association with body mass index, waist circumference, and waist-hip ratio. Arch Ophthalmol 2003;121:785-792. 21. Saksens NT, Kersten E, Groenewoud JM, et al. Clinical characteristics of familial and sporadic age-related macular degeneration: differences and similarities. Invest Ophthalmol Vis Sci 2014;55:7085-92. 22. Erke MG, Bertelsen G, Peto T, Sjolie AK, Lindekleiv H, Njolstad I. Cardiovascular risk factors associated with age-related macular degeneration: the Tromso Study. Acta Ophthalmol 2014;92:662-669. 23. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology - A proposal for reporting. JAMA 2000;283:2008-2012. 24. Ding X, Patel M, Chan C-C. Molecular pathology of age-related macular degeneration. Prog Retin Eye Res 2009;28:1-18. 25. Wells G, Shea B, O'Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. Canada, Accessed 2015. 26. Hamling J, Lee P, Weitkunat R, Ambuehl M. Facilitating meta-analyses by deriving relative effect and precision estimates for alternative comparisons from a set of estimates presented by exposure level or disease category. Stat Med 2008;27:954-970. 27. Higgins JPT, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. Br Med J 2003;327:557-560. 28. Vinding T, Appleyard M, Nyboe J, Jensen G. Risk factor-analysis for atrophic and exudative age-related macular degeneration - an epidemiologic study of 1000 aged individuals. Acta Ophthalmologica 1992;70:66-72. 29. McGuinness MB, Karahalios A, Simpson JA, et al. Past physical activity and age-related macular degeneration: the Melbourne Collaborative Cohort Study. The British journal of ophthalmology 2016;100:1353-1358. 30. Gopinath B, Liew G, Burlutsky G, Mitchell P. Physical activity and the 15-year incidence of age-related macular degeneration. Invest Ophthalmol Vis Sci 2014;55:7799-7803. 31. Klein R, Davis MD, Magli YL, Segal P, Klein BEK, Hubbard L. The Wisconsin age-related maculopathy grading system. Ophthalmology 1991;98:1128-1134. 32. Bird AEC, Bressler NM, Bressler SB, et al. An international classification and grading system for age-related maculopathy and age-related macular degeneration. Surv Ophthalmol 1995;39:367374. 33. DiPietro L. Physical activity in aging: Changes in patterns and their relationship to health and function. J Gerontol A Biol Sci Med Sci 2001;56:13-22. 34. Paffenbarger Jr RS, Hyde R, Wing AL, Hsieh C-c. Physical activity, all-cause mortality, and longevity of college alumni. New Eng J Med 1986;314:605-613. 35. Williams PT. Prospective epidemiological cohort study of reduced risk for incident cataract with vigorous physical activity and cardiorespiratory fitness during a 7-year follow-up. Invest Ophthalmol Vis Sci 2009;50:95-100. 36. Smith W, Assink J, Klein R, et al. Risk factors for age-related macular degeneration: pooled findings from three continents. Ophthalmology 2001;108:697-704. 37. Bauman AE, Sallis JF, Dzewaltowski DA, Owen N. Toward a better understanding of the influences on physical activity: the role of determinants, correlates, causal variables, mediators, moderators, and confounders. Am J Prev Med 2002;23:5-14. 38. Wen CP, Wai JPM, Tsai MK, et al. Minimum amount of physical activity for reduced mortality and extended life expectancy: a prospective cohort study. Lancet 2011;378:1244-1253. 39. Nigam A, Juneau M. Survival benefit associated with low-level physical activity. Lancet 2011;378:1202-1203. 40. Radak Z, Taylor AW, Ohno H, Goto S. Adaptation to exercise-induced oxidative stress: From muscle to brain. Exerc Immunol Rev 2001;7:90-107.
10
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
41. Khandhadia S, Lotery A. Oxidation and age-related macular degeneration: insights from molecular biology. Expert Rev Mol Med 2010;12:28. 42. Loprinzi PD, Swenor BK, Ramulu PY. Age-related macular degeneration is associated with less physical activity among US adults: Cross-sectional study. PLoS One 2015;10. 43. McGuinness MB, Karahalios A, Kasza J, Guymer RH, Finger RP, Simpson JA. Survival Bias When Assessing Risk Factors for Age-Related Macular Degeneration: A Tutorial with Application to the Exposure of Smoking. Ophthalmic Epidemiol 2017;24:1-10.
11
ACCEPTED MANUSCRIPT Figure captions
FIGURE 1. Flowchart depicting the literature search of physical activity and age-related
RI PT
macular degeneration, 1992-2016.
FIGURE 2. Forest plot of meta-analysis of physical activity and the presence of early age-
SC
related macular degeneration (AMD), 1992-2016. Dashed line indicates overall estimate;
M AN U
bars indicate 95% confidence interval (CI).
FIGURE 3. Funnel plot for meta-analysis of physical activity and the presence of early agerelated macular degeneration, 1992-2016. Solid vertical line indicates pooled estimate,
TE D
dashed lines indicate pseudo-95% confidence interval.
FIGURE 4. Forest plot of meta-analysis of physical activity and the presence of late age-
EP
related macular degeneration (AMD), 1992-2016. Separate estimates were reported for males and females by Erke. EDCCSG = Eye Disease Case-Control Study Group; dashed
AC C
line = overall estimate; bars = 95% confidence interval (CI).
FIGURE 5. Funnel plot for meta-analysis of physical activity and the presence of late agerelated macular degeneration, 1992-2016. Solid vertical line indicates pooled estimate, dashed lines indicate pseudo-95% confidence interval.
12
ACCEPTED MANUSCRIPT
Age range (years)
Eye Disease Case Control Study 13 Group (1992)
Eye Disease Case Control Study (USA)
Case control
55-80
Tromsø Study (Norway)
Crosssectional
65-80
19
Klein (2010)
Knudtson (2006)
10
16
Mares (2011)
McGuinness (2016) 11
Munch (2013)
21
Saksens (2014)
29
Blue Mountains Eye Study (Australia) Beaver Dam Offspring Study (USA) Beaver Dam Eye Study (USA)
Cohort
45-97
Crosssectional
21-84
Cohort
43-86
Carotenoids in Age-related Eye Diseases (USA)
Crosssectional
Melbourne Collaborative Cohort Study (Australia)
Cohort
Inter99 Study (Denmark)
Crosssectional
European Genetic Database Study (Netherlands and Germany)
METs = metabolic equivalent of task score
55-74
40-69
Case-control
30-60
68-72
Measure physical activity
N/A
Likert scale of activity level
Covariate adjustment
Drusen > 125 µm no pigment abnormalities Drusen > 125 µm with pigment abnormalities
Hours of activity per week
Age, sex, smoking, clinic, biochemical measures, ocular variables and estrogen use Age, sex, smoking and blood pressure
METs per week
Age, sex and smoking
Drusen 250 µm with pigment abnormalities
Any weekly exercise
Age and sex
Drusen ≥ 250 µm with pigment abnormalities Drusen ≥ 125 µm no pigment abnormalities OR Drusen ≥ 63 µm with pigment abnormalities Drusen > 63 µm with or without pigment abnormalities Drusen > 63 µm no pigment abnormalities >9 drusen < 63 µm with pigment abnormalities OR Drusen ≥ 63 µm and <125 µm with or without pigment abnormalities
Days of activity per week
None
METs per week
Age and smoking (all female)
TE D
30
EP
Gopinath (2014)
AC C
22
Erke (2014)
Features of Early AMD
SC
Study group (country)
M AN U
Study design
First author (year)
RI PT
Table 1. Characteristics of Studies used in meta-analysis of physical activity and age-related macular degeneration (1992-2016)
Days of vigorous activity per week () Hours of activity per week Days of activity per week
Age, sex and smoking Age, sex and smoking
Age, sex and smoking
ACCEPTED MANUSCRIPT
Table 2. Estimates included in meta-analysis of physical activity and age-related macular degeneration, 1992-2016 Classification of sedentary behaviour Subjectively below average
Calculated from published ordinal estimates
19
Beaver Dam Offspring Study (Klein, 2010) 10
Beaver Dam Eye Study (Knudtson, 2006)
21
European Genetic Database (Saksens, 2014)
(0.39,0.77)
2,539
Early
0.97
(0.79,1.19)
Late (Males)
1.12
(0.55,2.29)
1,134
Late (Females)
0.73
(0.31,1.71)
2,522
Early
0.85
(0.56,1.28)
2,455
Late
0.80
(0.42,1.52)
862
2,806
Early
1.25
(0.81,1.92)
Activity on < 3 days
Unadjusted aggregate level data supplied by author
4,682
Early
0.93
(0.78,1.11)
3,892
Late
0.47
(0.22,0.93)
1,312
Early
0.76
(0.51,1.13)
21,008
Early
0.92
(0.86,0.99)
14,139
Late
0.72
(0.43,1.22)
825
Early
0.54
(0.32,0.91)
2,418
Early
0.95
(0.78,1.16)
Calculated from published ordinal estimates
AC C
11
0.55
As published
Vigorous activity on < 1 day per week
Inter99 (Munch, 2013)
(95% CI)
Exercise on < 1 day per week
≤ 2.2 METs per week
Melbourne Collaborative 29 Cohort Study (McGuinness, 2016)
Calculated from patient level data supplied by author
EP
Carotenoids in Age-related 16 Eye Disease Study (Mares, 2011)
< 198 METs per week
TE D
30
Blue Mountains Eye Study (Gopinath, 2014)
As published
SC
< 1 hour vigorous activity or less vigorous activity on < 2 days
OR
1,036
AMD stage of cases Late
n
RI PT
22
Tromsø (Erke, 2014)
Source of OR estimate
M AN U
Study (Author, Year) Eye Disease Case-control 13 Study (EDCCSG, 1992)
Calculated from patient level data supplied by author
< 2 hours activity per week
Calculated from published ordinal estimates
Exercise on < 1 day per week
Adjusted estimates provided by author
0.50 (0.42,0.60) 2,668 Late EDCCSG = Eye Disease Case-control Study Group; METs = metabolic equivalent of task score; n = number of participants in model including controls; OR = odds ratio.
ACCEPTED MANUSCRIPT Table 3. Summary of pooled estimates for meta-analyses of physical activity and age-related macular degeneration, primary and sensitivity analyses AMD Status
Primary analysis
Sensitivity analysis
ࡱ
ࡼ
OR
(95% CI)
ࡵ (%)
ࡱ
ࡼ
OR
(95% CI)
ࡵ (%)
Early
8
35,695
0.92
(0.86, 0.98)
4.8
7
31,013
0.92
(0.84, 1.01)
18.3
Late
7
23,519
0.59
(0.49, 0.72)
23.5
6
15,176
0.61
(0.49, 0.77)
34.5
AC C
EP
TE D
M AN U
SC
RI PT
CI = confidence interval; ݊ா = number of estimates; ݊ = number of participants; OR = odds ratio; ܫଶ = measure of heterogeneity
ACCEPTED MANUSCRIPT
Table 4. Univariable meta-regression of meta-analysis of physical activity and age-related macular degeneration, 1992 to 2016
Late
OR
(95% CI)
Meta-analysis estimate Study location USA Europe Australia Measure of activity Weekly activity (hours or days) METs Other
8
0.92
(0.86,0.98)
4.8
3 3 2
0.94 0.88 0.92
(0.77,1.16) (0.70,1.11) (0.86,0.98)
28.1 54.3 0.0
5 2 1
0.92 0.80 1.25
(0.86,0.99) (0.60,1.07) (0.81,1.93)
7
0.59
2 3 2 5 1 1
a
࣎
Meta-regression Ratio of (95% CI) ORs
0.0006
(0.71,1.36) (0.75,1.28)
0.0000
8.9 0.0
0.0007 0.0000
1.00 0.87 1.36
(0.59,1.28) (0.77,2,40)
0.0000
(0.49,0.72)
23.5
0.0158
0.53 0.68 0.75
(0.39,0.73) (0.40,1.14) (0.50,1.13)
0.0 61.4 0.0
0.0000 0.1336 0.0000
1.00 1.17 1.43
(0.52,2.63) (0.56,3.68)
0.0420
0.62 0.80 0.55
(0.46,0.83) (0.42,1.52) (0.39,0.77)
38.7
0.0428
1.00 1.30 0.89
(0.40,4.17) (0.37,2.13)
0.0459
SC
1.00 0.98 0.98
AMD = age-related macular degeneration; METs = metabolic equivalent of task score; OR = odds ratio; CI = confidence Interval a No measure of ܫଶ is possible when only one study estimate is available
AC C
࣎
0.0109 0.0208 0.0000
M AN U
Meta-analysis estimate Study location USA Europe Australia Measure of activity Weekly activity (hours or days) METs Other
ࡵ (%)
RI PT
Number of estimates
TE D
Early
Study Characteristic
EP
AMD Status
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT