Customize your JAMA Network experience by selecting one or more topics from the list below.
Fenwick EK, Ong PG, Man REK, et al. Association of Vision Impairment and Major Eye Diseases With Mobility and Independence in a Chinese Population. JAMA Ophthalmol. 2016;134(10):1087–1093. doi:10.1001/jamaophthalmol.2016.2394
Copyright 2016 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
What is the association of vision impairment (VI) and major eye diseases with mobility and independence?
In this cross-sectional study of 3353 Chinese individuals, patients with severe VI in both eyes had a substantial and clinically meaningful reduction in their ability to perform tasks related to mobility and independence. Glaucoma and cataract were independently associated with worse mobility and independence.
Although these associations do not prove that public health strategies to prevent bilateral VI will improve mobility and independence, such interventions may be valuable from a population perspective.
Mobility limitations arising from vision impairment (VI) can result in loss of independence and reduced quality of life. However, few data are available on the association between VI and mobility limitations at a population-based level, particularly in Asian populations.
To assess the association of VI and major eye diseases with mobility and independence (M&I) in a Chinese population.
Design, Setting, and Participants
The Singapore Chinese Eye Study (February 9, 2009, to December 19, 2011) was a population-based, cross-sectional study of 3353 persons aged 40 to 80 years of Chinese ethnicity. Patients underwent visual acuity testing, and sociodemographic and medical data were collected from standardized questionnaires. Data analysis for this study was performed October 2015 to April 2016.
Presenting bilateral visual acuity (categorized as none, moderate, or severe VI) and major eye diseases (cataract, uncorrected refractive error, glaucoma, age-related macular degeneration, and diabetic retinopathy).
Main Outcomes and Measures
Patients answered questions on the M&I scale of the Impact of Vision Impairment questionnaire, validated using Rasch analysis. The composite M&I score (score range, −4.47 to 7.48 logits; higher scores indicate better M&I) and 11 individual item scores were the main outcomes. The association between bilateral VI and eye conditions and the composite and individual M&I item scores was assessed using linear regression models.
Of the 3353 patients, the mean (SD) age was 59.7 (9.9) years, and 1662 (49.6%) were male. The mean (SD) presenting visual acuity values in the better and worse eyes were 0.20 (0.21) and 0.39 (0.42) logMAR, respectively. A total of 1432 patients (42.7%) and 114 patients (3.4%) had moderate and severe bilateral VI, respectively. Mobility and independence systematically worsened as the severity of bilateral VI increased. There was a clinically meaningful reduction in M&I (20%; β, –1.44; 95% CI, –1.75 to –1.13) and all 11 M&I tasks in patients with severe bilateral VI compared with no VI. Glaucoma (13%; β, –0.94; 95% CI, –1.82 to –0.06) and cataract (6%; β, –0.43; 95% CI, –0.65 to –0.22) were independently associated with worse M&I, with patients with glaucoma particularly concerned about avoiding falling or tripping.
Conclusions and Relevance
Bilateral VI in this population was associated with substantial decrements in M&I, with glaucoma and cataract independently associated with worse M&I. Although these associations do not prove that preventing bilateral VI will improve M&I in this population, the results suggest that such interventions could be of tremendous value from this perspective.
Asia’s population is aging at an unprecedented rate because of increased life expectancy, better health care, and decreasing fertility.1 In Singapore, 1 in 5 people will be older than 65 years by 2030,2 and with changing family dynamics, more seniors will be living alone.3 Evidence suggests that the prevalence of mobility limitations has increased in adults 65 years or older in Singapore.4 Limitations in mobility can have disastrous consequences, resulting in loss of independence, functional decline in other daily activities, greater reliance on others, and reduced quality of life.4-6 In a study by Ng and colleagues7 in a population of Chinese adults in Singapore, reduced mobility and severe vision impairment (VI) were associated with nonsuccessful aging.
Vision impairment is a risk factor for reduced mobility and independence (M&I) and has been associated with slower walking speed, greater rate and risk of falls, early entry into residential care facilities, and increased difficulty driving.8-10 Age-related macular degeneration (AMD) and glaucoma, eye conditions that tend to affect central and peripheral vision, respectively, have been associated with reduced levels of physical activity and fewer excursions outside the home.11,12 However, few data are available on VI and mobility limitations at a population-based level. The Salisbury Eye Evaluation study13 found that those with VI were more likely to report mobility disability than those without. However, mobility was assessed using only 3 individual items (difficulty walking up 10 steps, down 10 steps, and 45 m), which precludes a comprehensive understanding of vision-related mobility disability. Moreover, the effect of VI on M&I in Asian populations is largely unknown. Differences in living conditions and lifestyle, cultural, and environmental habits may mean that people with VI in Asian populations face different mobility issues than in Western populations.
In this study, we investigated the association between VI and major eye diseases and M&I in patients from the population-based Singapore Chinese Eye Study (SCES). We used the validated M&I scale of the Impact of Vision Impairment (IVI) questionnaire,14 which examined 11 individual M&I tasks.
The SCES15,16 (February 9, 2009, through December 19, 2011) was a population-based, cross-sectional, and epidemiologic study of eye diseases in Chinese adults residing in Singapore aged 40 to 80 years. Details of the study design, sampling plan, and methods have been reported elsewhere.16 Briefly, patients were recruited using an age-stratified random sampling method, with 3353 of the 4605 eligible individuals participating in the SCES (72.8% response rate).17 The study was conducted in accordance with the Declaration of Helsinki. Ethics approval was obtained from the Singapore Eye Research Institutional Review Board, and written informed consent was obtained from each participant before enrollment in the study. Data were deidentified using unique identification numbers.
A logMAR chart (Lighthouse International) was used at a distance of 4 m. If no numbers were read at 4 m, the participant was moved to 3, 2, and then 1 m. If no numbers were identified on the chart, presenting visual acuity (PVA) was assessed as counting fingers, hand movements, perception of light, or no perception of light. Presenting visual acuity was measured in the left and right eyes separately with patients wearing their usual habitual optical correction (eg, spectacles or contact lenses). Best-corrected visual acuity was defined as best visual acuity measured at a distance after subjective refraction was determined for each eye and was based on the better seeing eye. Presenting visual acuity in the better eye was used in the current study because this acuity is considered to best represent the role of VI in patients’ performance of day-to-day tasks.18
Because both eyes are used by patients in real-world situations, we defined 3 categories of bilateral VI based on PVA: (1) none, defined as having 0.3 logMAR or less in both eyes; (2) moderate, defined as having 0.3 logMAR or less in one eye and greater than 0.3 logMAR in the other or greater than 0.3 to less than 1.0 logMAR in both eyes; and (3) severe, defined as having 1.0 or greater logMAR in one eye and greater than 0.3 to less than 1.0 logMAR in the other eye or 1.0 or greater logMAR in both eyes (0.3 logMAR equals Snellen equivalent 20/40, and 1.0 logMAR equals Snellen equivalent 20/200).
We determined the primary causes of vision loss using standardized definitions. Diabetic retinopathy and AMD were graded from retinal photographs using the modified Airlie House classification system from the Early Treatment Diabetic Retinopathy Study19 and the Wisconsin Age-Related Maculopathy Grading System, respectively.20 Glaucoma was diagnosed and classified using the International Society of Geographical and Epidemiological Ophthalmology scheme based on gonioscopy, optic disc characteristics, and visual field results.21 Uncorrected refractive error was defined as at least a 2-line difference between presenting and corrected logMAR in the better spherical eye and the presence of myopia or hyperopia.22 Lens status and the presence of aphakia or pseudophakia were determined with a slitlamp.
Mobility and independence were assessed using the M&I scale (Table 1) of the 32-item IVI, a vision-related quality-of-life questionnaire initially developed and validated in an Australian population with VI.14 The M&I items ask patients to rate their level of difficulty on mobility-related tasks (eg, negotiating steps and stairs, avoiding falling or tripping) and remaining independent (eg, needing help from other people, visiting friends or family). In the current study, the IVI was professionally translated into Mandarin and back-translated into English by 2 different translators fluent in Mandarin and English. Our group has recently validated the IVI in the SCES population using Rasch analysis23 and found the 11-item M&I scale to have satisfactory psychometric properties. Rasch analysis is a form of item response theory in which ordinal ratings of questionnaires are transformed to estimates of interval measures (expressed in log of the odd units or logits). Higher logit scores represent better M&I and vice versa. We also generated transformed individual person measures for all 11 items.
Characteristics of the study population were examined using proportions, means, medians, percentiles, and SDs. Key covariates included age (years), sex, educational level, income level, presence of comorbidity (yes or no), presence of eye condition (cataract, uncorrected refractive error, glaucoma, diabetic retinopathy, AMD, or other eye condition [yes or no]), and severity of bilateral VI.
We used χ2 statistics; the 2-sample, 2-tailed t test; and analysis of variance for univariate associations. The overall M&I and individual item scores were fitted to multiple linear regression modeling and t test–based 95% CIs for regression coefficients adjusted for sociodemographic characteristics, ocular and nonocular comorbidities, and other factors associated univariately with M&I. Multiple regression models were used to demonstrate the changes in β coefficients. To facilitate interpretation of the β coefficients in a clinically meaningful way, the coefficients were converted into adjusted means to obtain the relative changes and are presented in percentage form. The percentage was calculated by taking the β coefficient divided by the reference’s group M&I score ×100.
A difference in the M&I scores was estimated to be clinically important if the 95% CI limits of their β coefficients were approximately half the SD of the overall mean (0.60). All statistical analyses were performed with STATA statistical software, version 12.0 (StataCorp). Data analysis was performed October 2015 to April 2016.
The mean (SD) age of the 3353 patients was 59.7 (9.9) years, and 1662 (49.6%) were male (Table 2). The mean (SD) PVA values in the better and worse eyes were 0.20 (0.21) and 0.39 (0.42) logMAR, respectively. A total of 1805 patients had no bilateral VI, whereas 1432 patients (42.7%) and 114 patients (3.4%) had moderate and severe bilateral VI, respectively. The main cause of vision loss was uncorrected refractive error (11.3%) followed by cataract (7.3%). Those with an eye disease had substantially lower M&I scores compared with those without, and scores systematically worsened as severity of bilateral VI increased (P < .001) (Table 2).
The overall mean (SD) M&I score was 6.93 (1.44) logits (range, −4.47 to 7.48 logits). One-way analysis of variance found significant between-group effects on the overall score and all 11 individual M&I items. Scores for patients with glaucoma were substantially lower than the other 5 eye diseases, particularly for the individual items of go carefully to avoid falling or tripping and stopped you from doing the things you want to do (Table 1).
In our multiple regression model that adjusted for potential confounders (Table 3), compared with those with no bilateral VI, those with moderate bilateral VI had worse M&I (β, −0.12; 95% CI, −0.24 to −0.01). The association was clinically meaningful for the category of severe bilateral VI, with patients reporting a 20% (β, –1.44; 95% CI, −1.75 to −1.13) decrease in M&I compared with those with no VI. Glaucoma was associated with a 13% decrease in M&I (β, −0.94; 95% CI, −1.82 to −0.06) independent of VI. Similarly, cataract (β, −0.43; 95% CI, –0.65 to −0.22) and other eye diseases (β, −0.41; 95% CI, −0.79 to −0.02) were also independently associated with worse M&I (Table 3). Those 70 years or older were significantly more likely to report difficulty with M&I compared with those aged 40 to 49 years (eg, β, −0.30; 95% CI, −0.48 to −0.12) (Table 3).
An independent association with severe bilateral VI was found for all 11 M&I items (Table 4), with the greatest effect on spilling or breaking things (β, –1.60; 95% CI, −1.96 to −1.24; 23% reduction) and the least effect on visiting friends or family (β, –1.39; 95% CI, −1.76 to −1.03; 20% reduction), although the decrement was similar among all 11 M&I tasks. Four of the 11 items (general safety at home, general safety out of your home, stopped you from doing the things you want to do, and taking part in recreational activities) were not significantly associated with moderate bilateral VI (Table 4).
In our population-based study that assessed the effect of VI on self-reported mobility in Chinese adults living in Singapore, we found a clinically meaningful decrease in overall M&I and 11 vision-specific M&I tasks in those with severe bilateral VI. In addition, we found an independent association between cataract and glaucoma and reduced M&I even after controlling for bilateral VI, highlighting the multifaceted aspects of these 2 major eye diseases beyond central visual acuity loss. Our findings suggest that preventive strategies to stop or slow the progression to severe bilateral VI and vision-specific programs to optimize orientation and mobility for people living with severe VI are needed. Broad-ranging public initiatives to increase accessibility in community settings, public areas, and the transport system are essential to ensure that adults with VI can remain independent.
Ours is one of the few population-based studies to comprehensively explore the effect of VI on M&I measures rather than the more commonly studied construct of visual functioning. Our findings support those of the Salisbury Eye Evaluation study in which people with VI were significantly more likely to report difficulty walking up and down stairs and walking 150 ft than non–visually impaired people.13 Similarly, in a rural Indian population, mobility losses of 5.1, 10.2, and 23.4 points of 100 were found for people with VI (visual acuity, <6/18-6/60), low vision (visual acuity, <6/60-6/120), and blindness (visual acuity, <6/120), respectively.24 However, the mobility scale in this study contained only 3 items, 1 of which was more closely related to visual functioning than mobility (ie, doing your household chores).
Previous studies have found that, compared with normal vision, mild-moderate unilateral VI is associated with poor visual functioning25,26 or mobility.24 Although we also found an association between moderate bilateral VI and worse M&I, the decrement was small and not clinically significant. Moreover, 4 of the 11 M&I items were not significantly associated with moderate bilateral VI. These findings may suggest that the burden on M&I is limited in those with moderate bilateral VI in our Chinese population, highlighting the importance of preventing progression from moderate to severe VI. Early detection through regular eye screening prevents much of the vision loss associated with eye diseases, such as diabetic retinopathy27 and AMD.28 Our study suggests that societies with rapidly aging populations should understand the broader effect of VI and eye diseases on M&I beyond central vision alone. In Singapore, the importance of eye screening in older adults to decrease VI and blindness is already recognized by the Singapore Ministry of Health as part of their 2010 Functional Screening for Older Adults in the Community Clinical Practice Guidelines.29
Patients with glaucoma reported the largest reduction in M&I (13.4%) compared with those with other eye diseases, with the most affected task being avoiding falling or tripping. This finding suggests that patients with glaucoma perceive themselves to be at high risk of falling, which is supported by several studies that found an association between glaucoma and fearing of falling12,30 and increased risk of falls.31 In the Singapore Malay Eye Study,8 for example, having glaucoma was associated with a 4-fold increased risk of falling even when visual acuity was adjusted for. Our finding that glaucoma, cataract, and other eye conditions were associated with M&I is similar to that from a rural Indian population.24 The independent association between glaucoma and cataract and M&I suggests that other visual symptoms beyond visual acuity are important contributors to daily functioning.32 These symptoms may include loss of peripheral vision, difficulty with light and dark adaptation, contrast sensitivity for glaucoma,33 halos, stereo visual acuity, and contrast sensitivity for cataract.34 Indeed, a previous study35 reported associations between reduced contrast sensitivity and visual field mean deviation (MD) and vision-related quality of life in patients with glaucoma. Cataract extraction is highly successful in improving patient-centered outcomes for patients with cataract,36 whereas for patients with glaucoma, low vision rehabilitation may improve M&I.37 Our results also suggest that efforts to improve bilateral VI in eye conditions other than glaucoma and cataract may improve M&I in people with these eye diseases.
Older age was independently associated with worse M&I in our study, which, in conjunction with our finding that VI has a substantial effect on M&I, has important implications for policymaking given Singapore’s rapidly aging population. Indeed, to address the issues associated with an aging population and reduce the projected burden on Singapore’s health care system, the Ministry of Social and Family Development’s Committee on Aging Issues has launched a campaign to promote Successful Aging for Singapore. Part of their multifaceted long-term plan is to improve accessibility and mobility for older Singaporeans.2 Our findings suggest that an emphasis on universal design in new public developments that caters to all people regardless of age or disability and vision-specific aids, such as use of color strips to denote steps or uneven surfaces, would be beneficial.
Strengths of this study include its population-based design, large sample size, and the use of a well-validated, culturally appropriate questionnaire with a comprehensive measure of overall and task-specific M&I. The use of Rasch analysis to validate the IVI and provide estimated linear interval measures of M&I is another strong aspect of this study. One limitation is the cross-sectional nature of the study, which limits our capacity to investigate causality. Longitudinal changes in vision or eye disease in relation to a change in M&I could not be explored. We also did not take into account other aspects of visual function, such as peripheral vision (visual fields) or contrast sensitivity, which could have been associated with our M&I outcomes.32 We had visual field data, comprising MD and pattern SD (PSD), for half our sample (n = 1635); however, because the number of patients with severe bilateral VI was small in this subgroup, we were unable to conduct a subgroup analysis because of a lack of power. However, we ran 2 experiments to determine how visual field defects affected the association between VI and M&I observed in our study. First, because patients with MD and PSD data were those with suspected visual field defects, we assumed that those without MD and PSD data had no visual field defect and thus imputed a value of zero (indicating no visual field defect) for these individuals. After imputation, we reran the original regression models, including MD and PSD separately. The association among bilateral VI, glaucoma, cataract, other eye diseases, and M&I was unaffected, and MD and PSD were not significantly associated with M&I. Second, we reran the original regression models excluding those with glaucoma (n = 134); again, the observed association among bilateral VI, cataract, other eye conditions, and M&I was unchanged. This finding suggests that visual field defects are not likely to be driving the association between VI and M&I in our study.
Our study found that severe bilateral VI was associated with substantial and clinically meaningful decreases in overall measures of M&I and specific tasks relating to these broad categories in an adult Chinese population living in Singapore. Glaucoma was associated with a 13% decrease in M&I, with avoiding falling or tripping of particular concern to patients. The effects of cataract and glaucoma on M&I cannot be explained by central visual acuity alone. Although these associations do not prove that public health strategies to prevent bilateral VI and initiatives to improve accessibility for older people living with severe VI will improve M&I in this population, our results suggest that such interventions may be valuable from a population perspective.
Accepted for Publication: May 27, 2016.
Corresponding Author: Ecosse L. Lamoureux, PhD, Singapore Eye Research Institute, 20 College Rd, Level 6, Singapore 169856 (email@example.com).
Published Online: July 28, 2016. doi:10.1001/jamaophthalmol.2016.2394.
Author Contributions: Drs Fenwick and Lamoureux had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Fenwick, Cheng, Lamoureux.
Acquisition, analysis, or interpretation of data: Fenwick, Ong, Man, Sabanayagam, Wong, Lamoureux.
Drafting of the manuscript: Fenwick, Ong, Man.
Critical revision of the manuscript for important intellectual content: Fenwick, Man, Cheng, Sabanayagam, Wong, Lamoureux.
Statistical analysis: Fenwick, Ong, Sabanayagam.
Obtained funding: Cheng, Wong.
Administrative, technical, or material support: Cheng, Lamoureux.
Study supervision: Lamoureux.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Funding/Support: This study was supported by grant STaR/0003/2008 (Dr Wong) and grant CIRG/1417/2015 (Dr Cheng) from the National Medical Research Council, grant C-011/2006 from the Singapore Bio Imaging Consortium (Dr Wong), and grant 08/1/35/19/550 from the Biomedical Research Council (Dr Wong). Dr Fenwick is funded by grant 1072987 from the Australian National Health and Medical Research Council Early Career Fellowship. The Centre for Eye Research Australia receives Operational Infrastructure Support from the Victorian government.
Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.