- Email: [email protected]
Glaucoma and Associated Visual Acuity and Field Loss Significantly Affect Glaucoma-Specific Psychosocial Functioning
Glaucoma and Associated Visual Acuity and Field Loss Significantly Affect Glaucoma-Specific Psychosocial Functioning Errol W. Chan, MBBS, FRCOphth,1,2 Peggy P.C. Chiang, MIPH (Hons), PhD,2,3 Jiemin Liao, MA,1,2 Gwyneth Rees, MSc, PhD,3 Tien Y. Wong, FRCS, PhD,1,2,3,4 Janice S.H. Lam, MBBS,1,2 Tin Aung, FRCS, PhD,1,2,4 Ecosse Lamoureux, MSc, PhD2,3,4 Objective: To examine the impact of glaucoma and visual acuity (VA) and visual field (VF) losses on psychosocial functioning (PF). Design: Cross-sectional study. Participants: We compared PF between 192 participants with bilateral glaucoma with VA or VF losses and 40 controls from a tertiary eye hospital clinic in Singapore. Methods: Glaucoma was defined using the Hodapp-Anderson-Parish criteria. Four psychosocial outcomes of the Glaucoma Quality of Life 36 questionnaire were psychometrically assessed using Rasch analysis. Multivariate regression was performed to determine the independent impact of glaucoma and VA and VF losses on PF. The impact of VA and VF on PF were evaluated by restricted cubic spline analysis. Main Outcome Measures: Anxiety, self-image, psychological well-being, and confidence in health care. Results: The mean age of participants was 66.211.0 years, and 63% were male. In the better eye, VA and mean deviation were Snellen 20/25 and 8.896.52 dB, respectively. In multivariate models, glaucoma patients had 63.0% greater anxiety (95% confidence interval [CI], 66.0% to 61.2%; P < 0.001), 71.0% lower self-image (95% CI, 74.1% to 68.5%; P < 0.001), 38.3% less psychological well-being (95% CI, 37.4% to 39.0%; P < 0.001), and 32.4% reduced confidence in health care than patients without glaucoma. The worst VA and VF categories had the most reduced PF (range, 26.0% to 81.5%; P < 0.001 for all associations) compared with controls. With worsening VA, there was a linear increase in anxiety (P ¼ 0.009) and decrease in self-image (P ¼ 0.005). With worsening VF from 0 to 12.1 dB (P ¼ 0.003), anxiety increased before plateauing. Self-image decreased as VF worsened from 0 to 10 dB (P < 0.001), and confidence in health care decreased when VF worsened from 0 to 9.3 dB (P ¼ 0.008). However, self-image and confidence in health care actually improved at greater levels of VF loss beyond these thresholds. Conclusion: Glaucoma negatively affects PF. Early stage glaucoma with mild VF loss adversely affects anxiety, self-image, and confidence in health care. As VA worsens in advanced glaucoma, anxiety further increases and self-image deteriorates. Ophthalmologists and glaucoma patients need to be aware that both VA and VF losses at different stages of glaucoma negatively impact PF. Ophthalmology 2015;122:494-501 ª 2015 by the American Academy of Ophthalmology. Supplemental material is available at www.aaojournal.org
Glaucoma is a leading cause of visual impairment and blindness worldwide, and the prevalence is expected to increase.1,2 In Singapore, the prevalence of glaucoma ranges between 2% and 3.4%.3e5 Glaucoma adversely affects a patient’s overall health-related quality of life (HRQoL),6,7 as well as specific activities such as mobility, reading, and driving.8e12 However, the psychological, social, and emotional burdens of glaucoma are less clearly understood because of limited and equivocal evidence.13e22 Psychosocial functioning (PF) represents a key aspect of the World Health Organization’s conceptualization of HRQoL23 and refers to psychological and social effects of the disease on
494
2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
the patient. It distinctly differs from the ability to perform vision-related tasks,6e12 that is, vision-specific functioning, which has been the focus of HRQoL research in ophthalmology to date. Reduced PF is an increasingly recognized clinical consequence of chronic medical illness and negatively impacts an individual’s overall physical and mental health.24e26 Therefore, understanding how PF is affected in glaucoma patients is important and may determine when additional treatment and/or psychosocial therapy is indicated.27 Visual acuity (VA) and visual field (VF) losses are the most commonly used clinical indices to assess visual function in
http://dx.doi.org/10.1016/j.ophtha.2014.09.030 ISSN 0161-6420/14
Chan et al
Psychosocial Functioning in Glaucoma
glaucoma.28 However, although the extent of VF damage has been extensively used to characterize HRQoL in glaucoma,6e12 surprisingly limited corresponding evidence exists with glaucoma-induced VA loss.28e30 Quantitative changes in VA may more readily reflect HRQoL reduction in advanced glaucoma, where only a central island of vision is present.7,29,30 This study aims to investigate the impact of glaucoma and glaucoma-induced VA reduction and VF loss on 4 glaucoma-specific PF outcomes, namely, anxiety, self-image, psychological well-being, and confidence in health care, using the validated Glaucoma Quality of Life 36 (Glau-QoL-36) questionnaire.31
Methods Study Participants This cross-sectional study recruited patients aged 50 years and older from the glaucoma clinics of the Singapore National Eye Centre, a tertiary eye hospital in Singapore. Participants were excluded if they have had any prior laser procedure, hospitalization, or ocular or nonocular surgery in the last 2 months. Participants were also excluded if they had other eye disease, such as age-related macular degeneration, diabetic retinopathy, myopic macular degeneration, or significant cataract. Significant cataract was defined by the presence of nuclear opalescence of 2, cortical changes of 1, and any posterior subcapsular component, based on the Lens Opacity Classification System III in a dilated slit-lamp examination.32 Participants were recruited into 1 of 2 groups: (1) glaucoma cases and (2) controls. Glaucoma was defined based on the Hodapp-Anderson-Parish criteria33: vertical cup-to-disc ratio >0.7 and/or notching of the neuroretinal rim on clinical examination with a 78-diopter noncontact lens attributable to glaucoma associated with a reproducible glaucomatous VF defect on 2 occasions with the following features: (1) a glaucoma hemifield test outside normal limits, (2) a cluster of 3 nonedge, contiguous points not crossing the horizontal meridian on the pattern deviation plot, with a probability of <5% compared with age-matched normals (1 with a probability <1%), and (3) pattern standard deviation of <0.05. A VF was deemed reliable if it satisfied all of the following criteria: fixation losses of <20%, false positives and false negatives each <33%. The control group comprised patients from the same clinic population who were on follow-up solely as “glaucoma suspects” for optic disc evaluation (e.g., presence of a large cup-to-disc ratio [0.7]). None of these individuals had ocular hypertension or received any form of glaucoma-related medical, laser, or surgical treatment. All control participants had no VF defects (consistent with glaucoma or otherwise), had mean deviation (MD) better than 2 dB in each eye, and had logarithm of the minimum angle of resolution (logMAR) of 0.2 (Snellen 30/32). All participants had VA testing within the previous week and reliable VFs conducted in the last 3 months. During follow-up visits, VF testing was based on the Swedish Interactive Testing Algorithm (SITA FAST 24-2 algorithm) using a size III stimulus (Humphrey Visual Field Analyzer; Carl Zeiss Meditec, Inc., Dublin, CA). This study was approved by the Singhealth Institutional Research Board and conducted according to the tenets of the Declaration of Helsinki. Written informed consent was obtained from all participants.
Measurement of Covariates We measured VA monocularly in each eye with Early Treatment Diabetic Retinopathy Study charts at either 1 or 4 m.34 Participants used their habitual visual correction. Participants also underwent a refraction examination by study optometrists. Sociodemographic information (age, race, educational level, employment status, and number of persons residing in the same house) and data on medical comorbidities were collected from face-to-face interviews by trained research assistants.
Assessment of Psychosocial Functioning The Glau-QoL-36 questionnaire comprises 36 questions assessing the following 7 domains: anxiety, self-image, psychological wellbeing, confidence in health care, driving, burden of treatment, and daily life.31 Only the 4 domains pertaining to PF (anxiety, selfimage, psychological well-being, and confidence in health care) were evaluated in this study (Table 1). For all patients, masked research assistants administered the questionnaires before the clinical consult to eliminate the effect of clinician interaction on participants’ responses. All participants were asked to rate their difficulties and emotional responses to items, regardless of whether they were experienced recently. Responses were rated on a scale from 0 (not at all) to 4 (hugely) for anxiety, from 0 (completely true) to 3 (completely untrue) for the self-image domain, 0 (very often) to 4 (never) for the psychological well-being domain, and 0 (completely true) to 3 (completely untrue) in the confidence in health care domain (Table 1). The questionnaire was administered by trained bilingual research assistants in English and/or Mandarin.
Rasch Analysis Rasch analysis was conducted to determine the reliability, validity, and measurement characteristics of the 4 PF domains of the GlauQoL-36 questionnaire. Rasch analysis is a psychometric method that transforms raw ordinal scores into interval-level measurement (expressed in log of odds units [logits]). It calculates item difficulty (item measure) in relation to person ability (person measure) by placing both on the same linear continuum. A high person measure (in logits) indicates that the participant possesses a high level of the assessed latent trait (e.g., anxiety). Rasch analysis assesses the psychometric properties of a scale, such as how well items fit the underlying latent trait being measured (i.e., anxiety, self-image), how well items discriminate between the participants, how well item difficulty targets person ability, and the appropriateness of the response scale used.35 Winsteps (version 3.74; J.M. Linacre, Chicago, IL) was used to perform Rasch analysis on data from the 4 Glau-QoL-36 domains with the Andrich rating scale model. Each of the 4 PF domains of the Glau-QoL-36 scale was assessed for (1) item fit, (2) behavior of response categories (i.e., if higher categories represented better functioning), (3) measurement precision (minimum person separation index of 2.0), (4) unidimensionality, i.e., principal component analysis of >50% and an eigenvalue of <2.0, and (5) targeting items to the patient’s level of the 4 PF traits. Rasch scores (person measures) were generated after the structure of the responses and the items in the 4 PF domains of the Glau-QoL-36 questionnaire satisfied the five criteria of the Rasch model.
Statistical Analysis All statistical analyses were performed using STATA software version 12.1 (Stata Corp., College Station, TX). Characteristics of the study population and risk factors were examined using proportions, means, medians, percentiles, and standard deviations. Chi-square statistics and t tests were used for univariate associations. For each PF outcome, whether a nonlinear function resulted
495
Ophthalmology Volume 122, Number 3, March 2015 Table 1. Glaucoma Quality of Life-36 (Glau-QoL-36) Questionnaire: Dimensions, Item Content, and Response Options for Anxiety, Self-image, Psychological Well-being, and Confidence in Health Care Glau-QoL-36 Dimensions
Item Content
Anxiety
Feeling concerned about ocular pressure Fear of having to undergo an operation because of eye condition Problems in falling asleep because thinking about eye condition Thinking about the possibility of going blind
Self-image
Feeling that you didn’t get enough support when you first became ill Feeling of being treated differently Feeling physically or emotionally diminished Avoiding speaking about eye condition Feeling older because of eye condition Feeling that it is unfair that you are ill, rather than somebody else Feeling of being misunderstood Feeling of discouragement Feeling of frustration Feeling weak or vulnerable Feeling useless Regular check-ups with ophthalmologist being of help Following advice or instructions given by the ophthalmologist Medication being effective Having enough information about the eye disease and treatment
Psychological well-being
Confidence in health care
in a more reasonable fit than a linear trend was determined by inserting a quadratic term and determining significance using the F test. Visual predictors (i.e., glaucoma, VA, and VF MD in the better eye) were separately evaluated in multivariate linear or quadratic regression models adjusting for age, gender, other sociodemographic variables, and medical comorbidity. Where appropriate, b coefficients were converted into adjusted means and presented as percentages along with 95% confidence intervals (CIs) to facilitate interpretation. We used restricted cubic splines with knots corresponding to the 10th, 50th, and 90th percentiles36 of quantitative visual variables (i.e., VA and VF) significantly associated with PF outcomes to further characterize these relationships, namely, linear or quadratic. For quadratic associations, linear splines of the change in PF per unit change in VA or VF (i.e., PF/ logMAR or PF/MD slopes) were used to estimate specific VA or VF thresholds at which PF changed significantly. To ease the reader’s interpretation of logMAR VA, Snellen equivalents are shown (Table 2, available at www.aaojournal.org).
Results There were 192 bilateral glaucoma cases (82.7%) and 40 controls (17.3%). There were 108 primary open-angle glaucoma and 84 angle-closure glaucoma cases. Compared with our normal-sighted control group, those with glaucoma were generally older (mean age, 68.710.3 vs. 54.14.5 years; P < 0.001), were not working (71.5% vs. 40.0%; P < 0.001), reported lower educational levels (primary education or below, 57% vs. 29%; P < 0.001), and had a greater proportion with 2 medical comorbidities (16.4% vs. 2.5%; P ¼ 0.022; Table 3). As anticipated, compared with the control group, individuals with glaucoma had more severe VF loss in the better eye (MD, 8.89 vs. 1.19 dB; P < 0.001) and worse eye (17.70 vs. 2.82 dB; P < 0.001) and poorer logMAR VA in the better eye (0.11 vs. 0.01; P < 0.001) and worse eye (0.40 vs. 0.02; P < 0.001).
Validation of the 4 PF domains of the Glau-QoL-36 in Singapore Asians All participants answered all items in the Glau-QoL-36 questionnaire. For the anxiety and self-image domains, there was evidence
496
Response Options 0¼Not at all 1¼A little 2¼Moderately 3¼A lot 4¼Hugely 0¼Completely true for me 1¼Somewhat true for me 2¼Somewhat untrue for me 3¼Completely untrue for me 0¼Very often 1¼Often 2¼From time to time 3¼Seldom 4¼Never 0¼Completely true 1¼Somewhat true for me 2¼Somewhat untrue for me 3¼Completely untrue for me
of disordered thresholds (a sign that the response categories are not working as intended). Merging adjacent response categories subsequently resulted in ordered thresholds. Items that demonstrated misfit, that is, infit residual values of >1.3, were iteratively removed to improve fit statistics. These were Q13 (“Do you feel like you are the victim of an injustice?”), Q14 (“Do you feel misunderstood?”), and Q15 (“Do you feel discouraged?”) for the psychological well-being scale and Q25 (“Are you afraid of surgery?”) for the anxiety scale. The person separation index ranged from 0.68 (anxiety) to 1.27 (psychological well-being), suggesting that the 4 PF scales could only modestly discriminate between levels of difficulty experienced by the participants. However, the principal component analysis of the residuals for each of the 4 PF scales explained >50% of the raw variance (range, 51.2%e58.8%), and the unexplained variance by the first contrast of the residuals for all scales was <2.0 eigenvalues. These 2 parameters indicate that the 4 assessed PF traits were unidimensional (measuring a single latent trait). The mean value standard deviation of person and item fit residual values ranged from 4.842.82 (psychological well-being) to þ0.271.26 (anxiety). These data suggest that the scale item difficulty in relation to person ability was suboptimal and participants were too able for the level of item difficulty in the scale. There was no evidence of differential item functioning for any items across the 4 PF scales for age, gender, and systemic comorbidities.
Impact of Glaucoma, VA, and VF Losses in the Better Eye on PF Glaucoma cases, compared with controls, had significantly lower mean scores for anxiety (1.801.80 vs. 4.290.68; P < 0.001), self-image (1.461.67 vs. 4.090; P < 0.001), psychological well-being (4.402.92 vs. 6.950.35; P < 0.001), and confidence in health care (3.841.84 vs. 6.150; P < 0.001). In multivariate models adjusting for sociodemographic factors and medical comorbidities (Table 4), glaucoma in the better eye was independently associated with increased anxiety (b ¼ 3.09 [95% CI, 3.85 to 2.31]; P < 0.001), lower self-image (b ¼ 2.79 [95% CI, 3.40 to 2.18]; P < 0.001), psychological well-being
Chan et al
Psychosocial Functioning in Glaucoma
Table 3. Visual, Demographic, and Health Characteristics of Controls and Bilateral Glaucoma Cases Controls (n [ 40) Demographics Age (yrs) Female gender (%) Education level, no. (%) Primary and below Secondary/technical diploma Pre-university or higher Employed, no. (%) Housing Public, no. (%) Private, no. (%) No. of additional persons in same household Health (medical comorbidities) <2 2 Vision (logMAR VA) Worse eye Better eye VF loss MD (dB) Worse Eye Better Eye Glaucoma subtype, no. (%) Primary open-angle glaucoma Primary angle closure glaucoma
54.13 (4.48) 18 (45.0)
Glaucoma (n [ 192) 68.67 (10.29) 67 (34.9)
P <0.001 0.228
(57.0) (28.9) (14.1) (28.5)
<0.001
36 (90.0) 4 (10.0) 3.73 (1.15)
178 (92.7) 14 (7.3) 3.34 (1.40)
0.765 0.116
39 (97.5) 1 (2.5)
158 (83.6) 31 (16.4)
0.022
0.02 (0.04) 0.01 (0.03)
0.40 (0.81) 0.11 (0.11)
<0.001 <0.001
2.82 (5.55) 1.19 (3.02)
17.70 (8.99) 8.89 (6.52)
<0.001 <0.001
11 10 17 24
(29.0) (26.3) (44.7) (60.0)
e e
73 37 18 37
90 (46.9) 84 (43.8)
<0.001
-
Data are mean (standard deviation) unless otherwise indicated. logMAR ¼ logarithm of minimum angle of resolution; MD ¼ mean deviation; VA ¼ visual acuity; VF ¼ visual field.
(b ¼ 3.01 [95% CI, 4.19 to 1.82]; P < 0.001), and confidence in health care (b ¼ 1.66 [95% CI, 2.37 to 0.94]; P < 0.001) compared with controls. In terms of percentage reductions (Table 5, available at www.aaojournal.org), glaucoma was associated with reduced PF compared with controls (63.3% for anxiety, 71.0% for self-image, 38.3% for psychological wellbeing, and 32.4% for confidence in health care). In multivariate models, worse best-corrected VA in the better eye was linearly associated with lower self-image (b ¼ 0.27 [95% CI, 0.45 to 0.08]; P ¼ 0.005) and greater anxiety (b ¼ 0.30 [95% CI, 0.52 to 0.07]; P ¼ 0.009; Table 4). Compared with controls, glaucoma patients in each tertile of VA loss had significantly worse PF for all outcomes, ranging from 26.8% to 81.5% (all P < 0.001; Table 5; Fig 1, available at www.aaojournal.org). Greater VF loss in the better eye was quadratically associated (i.e., increased with the square of the PF outcome) with worse anxiety (b ¼ 0.04 [95% CI, 0.07 to 0.02]; P ¼ 0.002), selfimage (b ¼ 0.03 [95% CI, 0.05 to 0.01]; P ¼ 0.002), and less confidence in health care (b ¼ 0.03 [95% CI, 0.55 to 0.10]; P ¼ 0.002; Table 4). Compared with controls, glaucoma patients in each tertile of VF loss showed significantly worse PF for all outcomes, ranging between 26.0% to 76.1% (all P < 0.001; Table 5; Fig 2, available at www.aaojournal.org). To characterize how PF varied across the spectrum of VA and VF losses, restricted cubic spline analysis was conducted for the above significant associations with VA (Fig 3) and VF loss (Fig 4) in the better eye. Self-image and anxiety decreased linearly with reduced best-corrected VA. Self-image decreased going from 0 to 10 dB but plateaued at 10 dB or more severe levels of VF loss. For self-image, mean adjusted PF scores at >10 and 10 dB levels were 2.32 and 1.08 logits, respectively (PF/MD slope difference, 0.25; P < 0.001). Going from 0 to 12.1 dB, anxiety increased, but it
improved beyond 12.1 dB. For anxiety, the PF/MD slopes at better (>12.1 dB) and poorer MD (12.1 dB) were 0.2 and 0.07, respectively (P ¼ 0.003). Similarly, confidence in health care decreased going from 0 to 9.3 dB and improved beyond 9.3 dB. The PF/MD slopes at better (>9.3 dB) and poorer MD (9.3 dB) were 0.12 and 0.07, respectively (P ¼ 0.008).
Discussion This study characterizes the impact of glaucoma and associated VA and VF losses on specific PF outcomes. Glaucoma patients had 63% greater anxiety, 71% lower self-image, 38% less psychological well-being, and 32% reduced confidence in health care compared with controls. The results indicate that in early stage glaucoma (i.e., at mild VF loss), when VF worsens, anxiety increases, whereas self-image and confidence in health care are reduced (Fig 4). In late-stage glaucoma (i.e., when VA is eventually affected), only anxiety and self-image worsen as VA worsens (Fig 3). Hence, preserving both VA and VF in glaucoma patients is important. This study adds to the limited body of evidence on psychological, social, and emotional functioning in glaucoma.13e22 In this study, there were significant associations between glaucoma and VA and VF losses and reduced PF, namely, increased anxiety, reduced self-image, and less confidence in health care. Several earlier studies found no correlation between VA or VF loss with patient perceptions of treatment,19 anxiety,16 or fear of blindness.18 However, Parrish et al21 and Gutierrez et al22 determined that VF losses
497
Ophthalmology Volume 122, Number 3, March 2015 Table 4. Impact of Bilateral Glaucoma and Visual Acuity and Visual Field Loss in the Better Eye on Psychosocial Functioning Using the Glaucoma Quality of Life 36 Questionnaire Glaucoma (Present)
b (95% CI) Unadjusted Anxiety Self-image Psychological well-being Confidence in health care Adjusted* Anxiety Self-image Psychological well-being Confidence in health care
VA (0.1 logMAR Worse)
b (95% CI)
P
VF Loss, MD (5 dB Worse)
b (95% CI)
P
2.49 2.63 2.55 2.31
(3.06 (3.15 (3.46 (2.89
to to to to
1.92) 2.11) 1.64) 1.74)
<0.001 <0.001 <0.001 <0.001
0.29 0.34 0.33 0.19
(0.46 (0.50 (0.58 (0.36
to to to to
0.13) 0.18) 0.08) 0.02)
0.001y <0.001y 0.009y 0.025y
0.03 0.04 0.31 0.05
(0.05 (0.05 (0.57 (0.07
to to to to
0.01) 0.02) 0.044) 0.03)
3.09 2.79 3.01 1.66
(3.85 (3.40 (4.19 (2.37
to to to to
2.31) 2.18) 1.82) 0.94)
<0.001 <0.001 <0.001 <0.001
0.30 0.27 0.29 0.07
(0.52 (0.45 (0.60 (0.26
to to to to
0.07) 0.08) 0.03) 0.12)
0.009y 0.005y 0.074y 0.493y
0.04 0.03 0.23 0.03
(0.07 (0.05 (0.55 (0.05
to to to to
0.02) 0.01) 0.10) 0.01)
P 0.004z <0.001z 0.022y <0.001z 0.002z 0.002z 0.175y 0.007z
CI ¼ confidence interval; logMAR ¼ logarithm of the minimum angle of resolution; MD ¼ mean deviation; VA ¼ visual acuity; VF ¼ visual field. *Adjusted for age, employment status, educational level, no. of persons in the house, and medical comorbidities. y b and P values indicated for linear relationships between psychosocial functioning domains with VA or VF. z b and P values indicated for quadratic relationships between psychosocial functioning domains with VA or VF.
were at least modestly correlated with vision-specific dependency, role difficulties, social and emotional well-being, and mental health on the National Eye Institute VisionFunction Questionnaire. The different study findings may be due to the dissimilar spectrums of glaucoma severity between studies. Our study shows that 3 of the 4 PF outcomes assessed (anxiety, self-image, and confidence in health care) remain largely unchanged in advanced glaucoma despite worsening VA or VF loss (Figs 1 and 2). This may explain why other studies involving individuals at these more severe disease levels generally found nonsignificant or weak associations. A second reason could be the higher sensitivity of the glaucoma-specific Glau-QoL-36 questionnaire compared with health- or vision-related QoL questionnaires in eliciting concerns directly related to glaucoma. Glaucoma had a significant negative impact on all 4 PF outcomes, even after adjusting for sociodemographic and health factors. This effect could be mediated indirectly through glaucoma-related impairment in key QoL subtraits, such as daily functioning, including use of topical medication and mobility.8e11 For example, the cost and compliance of long-term glaucoma treatment and monitoring could reduce PF. Prior research has indicated that patients with glaucoma have an increased fear of falling, possibly because of physical limitations imposed by visual disability, thereby affecting anxiety, self-image, and psychological well-being outcomes.37 Driving cessation is strongly impacted by glaucoma,8,12 consequently affecting PF. Understanding to what extent PF outcomes are linked with activity limitation may be important because this may suggest that targeted strategies to assist glaucoma patients with daily activities could potentially improve PF. Even mild VF losses were associated with increased anxiety and reduced self-image and confidence in health care (Fig 4). Ophthalmologists managing patients with early glaucoma should recognize that patients may have valid concerns pertaining to intraocular pressure, possibility of blindness, being emotionally diminished, or being treated differently (Table 1) to engage patient trust. In early disease, other visual function aspects (e.g., contrast sensitivity) may
498
already be impaired. The negative effect of mild VF loss on PF is consistent with the Los Angeles Latino Eye Study,7 where small changes in VF resulted in observable reductions in HRQoL. This finding underscores the importance of identifying glaucoma patients early in their disease to preserve existing visual function. Anxiety increased and confidence in health care decreased with worsening VF initially but improved in more severe disease beyond 12.1 and 9.3 dB, respectively. Loss of VF was associated with reduced self-image up to about 10 dB but remained unchanged beyond 10 dB. Such “quadratic” associations between visual function and HRQoL outcomes have not been previously reported because of assumptions that all significant associations followed linear trends.6e12,37 Several hypotheses could explain these results. Crabb et al38 reported that patients with mild to moderate glaucoma commonly perceived blurred or missing patches, whereas individuals with advanced glaucoma perceive a tunnel effect with blurred edges. When early glaucoma progresses, patients may experience greater anxiety and concerns about self-image and loss of confidence in health care, perhaps relating to an increase in the number or depth of scotoma. However, individuals with advanced glaucoma and tunnel vision fields may not notice small changes in VF, especially if progression occurs in the periphery.28,29 Increased adaptation to vision loss associated with glaucoma when glaucoma is advanced could also explain reduced anxiety and self-image concerns. Patients with more severe disease could have had more aggressive intraocular pressureelowering therapy, e.g., glaucoma surgery, possibly accounting for better confidence in health care. Ophthalmologists need to be aware of increased anxiety and reductions in self-image and confidence in health care as patients lose VF up to about 9 to 12 dB. In addition, improved PF in advanced glaucoma may explain why patient motivation toward treatment is reduced, such as reduced medication compliance. Loss of VA was associated with increased anxiety and reduced self-image (Fig 3). Worsening VA suggests increasing involvement of central fixation, as is present in
Chan et al
Psychosocial Functioning in Glaucoma
Figure 3. Relationship between psychosocial functioning and logarithm of the minimum angle of resolution (logMAR) visual acuity in the better eye in bilateral glaucoma in a restricted cubic spline plot for (A) anxiety, (B) self-image, and (C) psychological well-being. (Black line represents the estimate, and the shaded area delineates the upper and lower limits of the 95% confidence intervals (CIs). Knots are set at the 10th, 50th, and 90th percentiles of logMAR visual acuity [0, 0.02, and 0.26)]. Snellen visual acuity is indicated in parentheses on the axes). logit ¼ log of odds unit.
advanced glaucoma.29,30 Monitoring VA changes in glaucoma patients, particularly those with advanced disease, could be important because anxiety and self-image outcomes worsen whenever VA losses occur. These findings suggest that it is important to consider VA loss in addition to VF loss in terms of the overall impact on PF; although anxiety and self-image levels reach a plateau as VF worsens, if VA also worsens, it is likely that PF will be affected. In advanced disease, VA could be a better indicator of PF than VF. Strengths of this study include the use of Rasch analysis to validate 4 specific domains of the Glau-QoL-36 questionnaire and produce linear interval measures for each of the 4 PF domains. This study also used locally weighted scatterplot smoothing analysis to characterize the trend in PF over the spectrum of VA and VF losses. Standard regression techniques used in several QoL studies assume linear relationships and do not adequately capture information over the entire disease trajectory.6e12,37
Figure 4. Relationship between psychosocial functioning and visual field loss in the better eye in bilateral glaucoma in a restricted cubic spline plot for (A) anxiety, (B) self-image, and (C) confidence in health care. (Black line represents the estimate, and the shaded area delineates the upper and lower limits of the 95% confidence intervals. Knots are set at 10th, 50th, and 90th percentiles of visual field loss (0.14, 5.17, and 16.67). Logit ¼ log of odds unit; MD ¼ mean deviation.
Several limitations of this study should be discussed. Because this was a cross-sectional study, investigation of causation was not possible. This study involved patients in a tertiary ophthalmic center and may not accurately represent
499
Ophthalmology Volume 122, Number 3, March 2015 individuals with glaucoma in the population, many of whom are undiagnosed. The control group comprised glaucoma suspects, which could underestimate the impact of glaucoma on PF because the psychological burden of “possibly having glaucoma” in the glaucoma suspect control group may narrow the differences in PF than if a group with only mild eye disease was used as the control. Nevertheless, we still found a significant difference in all PF aspects between the glaucoma and control groups. Glaucoma suspects have been included as controls in similar QoL studies9e11,37 because population-based individuals may have other unmeasured confounding factors or may have “supranormal” PF. The knowledge of “potential glaucoma” is likely to have had only a small effect because VA and VF losses were clearly related to several PF outcomes. There was almost twice the number of female than male glaucoma patients because of the sociodemographic characteristics of our hospital-based sample. However, this is unlikely to bias the findings because independent relationships were present after controlling for gender. Although we considered only better eye VA and VF, rather than an “integrated VF” or binocular VA, strong correlations between monocular and binocular VF39 and VA40 measures still exist. Monocular visual parameters are readily available in the clinical setting and thus are more easily integrated into clinical decision making. In conclusion, this study demonstrates the adverse impact of glaucoma on PF. Losses in VF in early glaucoma significantly worsen anxiety and self-image, emphasizing a need for early identification of glaucoma. In advanced glaucoma, if VA worsens, anxiety and self-image outcomes will be adversely affected. The association between VA and PF supports the use of VA monitoring in glaucoma patients at all stages of severity. Further research is required to determine the efficacy of various therapeutic and rehabilitative strategies in improving PF for glaucoma patients.
7.
8.
9. 10.
11. 12. 13. 14.
15. 16. 17.
Acknowledgments. The authors thank Dr Eva Fenwick, from the University of Melbourne, for her assistance in the preparation of this manuscript.
18.
References
19.
1. Quigley HA, Broman AT. The number of people with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol 2006;90: 262–7. 2. Eye Diseases Prevalence Research Group. Prevalence of openangle glaucoma among adults in the United States. Arch Ophthalmol 2004;122:532–8. 3. Shen SY, Wong TY, Foster PJ, et al. The prevalence and types of glaucoma in Malay people: the Singapore Malay Eye Study. Invest Ophthalmol Vis Sci 2008;49:3846–51. 4. Foster PJ, Oen FT, Machin D, et al. The prevalence of glaucoma in Chinese residents of Singapore: a cross-sectional population survey. Arch Ophthalmol 2000;118:1105–11. 5. Narayanaswamy A, Baskaran M, Zheng Y, et al. The prevalence and types of glaucoma in an urban Indian population: the Singapore Indian Eye Study. Invest Ophthalmol Vis Sci 2013;54:4621–7. 6. Chan EW, Chiang PP, Wong TY, et al. Impact of glaucoma severity and laterality on vision-specific functioning: the
500
20.
21. 22. 23. 24. 25.
Singapore Malay Eye Study. Invest Ophthalmol Vis Sci 2013;54:1169–75. McKean-Cowdin R, Wang Y, Wu J, et al; Los Angeles Latino Eye Study Group. Impact of visual field loss and health-related quality of life in glaucoma: the Los Angeles Latino Eye Study. Ophthalmology 2008;115:941–8. West SK, Rubin GS, Broman AT, et al; SEE Project Team. How does visual impairment affect performance on tasks of everyday life? The SEE Project. Arch Ophthalmol 2002;120: 774–80. Ramulu PY, Maul E, Hochberg C, et al. Real-world assessment of physical activity in glaucoma using an accelerometer. Ophthalmology 2012;119:1159–66. Hochberg C, Maul E, Chan ES, et al. Association of vision loss in glaucoma and age-related macular degeneration with IADL disability. Invest Ophthalmol Vis Sci 2012;53: 3201–6. Ramulu PY, Swenor BK, Jefferys JL, et al. Difficulty with outloud and silent reading in glaucoma. Invest Ophthalmol Vis Sci 2013;54:666–72. Ramulu PY, West SK, Munoz B, et al. Driving cessation and driving limitation in glaucoma: the Salisbury Eye Evaluation Project. Ophthalmology 2009;116:1846–53. Wilson MR, Coleman AL, Yu F, et al. Depression in patients with glaucoma as measured by self-report surveys. Ophthalmology 2002;109:1018–22. Wang SY, Singh K, Lin SC. Prevalence and predictors of depression among participants with glaucoma in a nationally representative population sample. Am J Ophthalmol 2012;154: 436–44. Bramley T, Peeples P, Walt JG, et al. Impact of vision loss on costs and outcomes in Medicare beneficiaries with glaucoma. Arch Ophthalmol 2008;126:849–56. Mabuchi F, Yoshimura K, Kashiwagi K, et al. Risk factors for anxiety and depression in patients with glaucoma. Br J Ophthalmol 2012;96:821–5. Odberg T, Jakobsen JE, Hultgren SJ, Halseide R. The impact of glaucoma on the quality of life of patients in Norway. I. Results from a self-administered questionnaire. Acta Ophthalmol Scand 2001;79:116–20. Janz NK, Wren PA, Lichter PR, et al; CIGTS Group. Quality of life in newly diagnosed glaucoma patients: the Collaborative Initial Glaucoma Treatment Study. Ophthalmology 2001;108:887–97. Jampel HD, Frick KD, Janz NK, et al. Depression and mood indicators in newly diagnosed glaucoma patients. Am J Ophthalmol 2007;144:238–44. Bhargava JS, Patel B, Foss AJ, et al. Views of glaucoma patients on aspects of their treatment: an assessment of patient preference by conjoint analysis. Invest Ophthalmol Vis Sci 2006;47:2885–8. Parrish RK II, Gedde SJ, Scott IU, et al. Visual function and quality of life among patients with glaucoma. Arch Ophthalmol 1997;115:1447–55. Gutierrez P, Wilson RW, Johnson C, et al. Influence of glaucomatous visual field loss on health-related quality of life. Arch Ophthalmol 1997;115:777–84. WHOQOL Group. The World Health Organization quality of life assessment (WHOQOL): position paper from the World Health Organization. Soc Sci Med 1995;41:1403–9. Astrom M, Asplund K, Astrom T. Psychosocial function and life satisfaction after stroke. Stroke 1992;23:527–31. Carels RA, Musher-Eizenman D, Cacciapaglia H, et al. Psychosocial functioning and physical symptoms in heart failure
Chan et al
26. 27. 28. 29. 30. 31.
32.
Psychosocial Functioning in Glaucoma
patients: a within-individual approach. J Psychosom Res 2004;56:95–101. Friedman LC, Kalidas M, Elledge R, et al. Optimism, social support and psychosocial functioning among women with breast cancer. Psychooncology 2006;15:595–603. Ramulu PY. Glaucoma and disability: which tasks are affected, and at what stage of disease. Curr Opin Ophthalmol 2009;20:92–8. Nelson P, Aspinall P, Papasouliotis O, et al. Quality of life in glaucoma and its relationship with visual function. J Glaucoma 2003;12:139–50. Asaoka R. The relationship between visual acuity and central visual field sensitivity in advanced glaucoma [letter]. Br J Ophthalmol 2013;97:1355–6. Portney GL, Hanible JA. Visual acuity and field loss in advanced glaucoma. Ann Ophthalmol 1974;6:1147–50. Bechetoille Arnould B, Bron A, et al. Measurement of healthrelated quality of life with glaucoma: validation of the GlauQOL 36-item questionnaire. Acta Ophthalmol 2008;86: 71–80. Chylack LT Jr, Wolfe JK, Singer DM, et al. Longitudinal Study of Cataract Study Group. The Lens Opacities Classification System III. Arch Ophthalmol 1993;111:831–6.
33. Hodapp E, Parrish RK II, Anderson DR. Clinical Decisions in Glaucoma. St Louis: Mosby; 1993:52–61. 34. Bailey IL, Bullimore MA, Raasch TW, Taylor HR. Clinical grading and the effects of scaling. Invest Ophthalmol Vis Sci 1991;32:422–32. 35. Lamoureux E, Pesudovs K. Vision-specific quality-of-life research: a need to improve the quality. Am J Ophthalmol 2011;151:195–7. 36. Durrleman S, Simon R. Flexible regression models with cubic splines. Stat Med 1989;8:551–61. 37. Ramulu PY, van Landingham SW, Massof RW, et al. Fear of falling and visual field loss from glaucoma. Ophthalmology 2012;119:1352–8. 38. Crabb DP, Smith ND, Glen FC, et al. How does glaucoma look? Patient perception of visual field loss. Ophthalmology 2013;120:1120–6. 39. Arora KS, Boland MV, Friedman DS, et al. The relationship between better-eye and integrated visual field mean deviation and visual disability. Ophthalmology 2013;120:2476–84. 40. Rubin GS, Munoz B, Bandeen-Roche K, West SK. Monocular versus binocular visual acuity as measures of visual impairment and predictors of visual disability. Invest Ophthalmol Vis Sci 2000;41:3327–34.
Footnotes and Financial Disclosures Originally received: January 17, 2014. Final revision: September 19, 2014. Accepted: September 24, 2014. Available online: November 8, 2014.
Financial Disclosure(s): The authors have no proprietary or commercial interest in any materials discussed in this article. Manuscript no. 2014-89.
1
Department of Ophthalmology, National University Health System and National University of Singapore, Singapore.
2
Singapore Eye Research Institute, Singapore National Eye Centre, Singapore.
3
Centre for Eye Research Australia, The Royal Victorian Eye and Ear Hospital, University of Melbourne, Australia.
4
DUKE-NUS Graduate Medical School, Singapore.
Abbreviations and Acronyms: CI ¼ confidence interval; Glau-QoL-36 ¼ Glaucoma Quality of Life 36; HRQoL ¼ health-related quality of life; logits ¼ log of odds units; logMAR ¼ logarithm of the minimum angle of resolution; MD ¼ mean deviation; PF ¼ psychosocial functioning; VA ¼ visual acuity; VF ¼ visual field. Correspondence: Ecosse Lamoureux, MSc, PhD, Office of Clinical Sciences, Duke-NUS Graduate Medical School Singapore, 8 College Road Level 4, Singapore 169857. E-mail: [email protected].
501
Glaucoma is a leading cause of visual impairment and blindness worldwide, and the prevalence is expected to increase.1,2 In Singapore, the prevalence of glaucoma ranges between 2% and 3.4%.3e5 Glaucoma adversely affects a patient’s overall health-related quality of life (HRQoL),6,7 as well as specific activities such as mobility, reading, and driving.8e12 However, the psychological, social, and emotional burdens of glaucoma are less clearly understood because of limited and equivocal evidence.13e22 Psychosocial functioning (PF) represents a key aspect of the World Health Organization’s conceptualization of HRQoL23 and refers to psychological and social effects of the disease on
494
2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
the patient. It distinctly differs from the ability to perform vision-related tasks,6e12 that is, vision-specific functioning, which has been the focus of HRQoL research in ophthalmology to date. Reduced PF is an increasingly recognized clinical consequence of chronic medical illness and negatively impacts an individual’s overall physical and mental health.24e26 Therefore, understanding how PF is affected in glaucoma patients is important and may determine when additional treatment and/or psychosocial therapy is indicated.27 Visual acuity (VA) and visual field (VF) losses are the most commonly used clinical indices to assess visual function in
http://dx.doi.org/10.1016/j.ophtha.2014.09.030 ISSN 0161-6420/14
Chan et al
Psychosocial Functioning in Glaucoma
glaucoma.28 However, although the extent of VF damage has been extensively used to characterize HRQoL in glaucoma,6e12 surprisingly limited corresponding evidence exists with glaucoma-induced VA loss.28e30 Quantitative changes in VA may more readily reflect HRQoL reduction in advanced glaucoma, where only a central island of vision is present.7,29,30 This study aims to investigate the impact of glaucoma and glaucoma-induced VA reduction and VF loss on 4 glaucoma-specific PF outcomes, namely, anxiety, self-image, psychological well-being, and confidence in health care, using the validated Glaucoma Quality of Life 36 (Glau-QoL-36) questionnaire.31
Methods Study Participants This cross-sectional study recruited patients aged 50 years and older from the glaucoma clinics of the Singapore National Eye Centre, a tertiary eye hospital in Singapore. Participants were excluded if they have had any prior laser procedure, hospitalization, or ocular or nonocular surgery in the last 2 months. Participants were also excluded if they had other eye disease, such as age-related macular degeneration, diabetic retinopathy, myopic macular degeneration, or significant cataract. Significant cataract was defined by the presence of nuclear opalescence of 2, cortical changes of 1, and any posterior subcapsular component, based on the Lens Opacity Classification System III in a dilated slit-lamp examination.32 Participants were recruited into 1 of 2 groups: (1) glaucoma cases and (2) controls. Glaucoma was defined based on the Hodapp-Anderson-Parish criteria33: vertical cup-to-disc ratio >0.7 and/or notching of the neuroretinal rim on clinical examination with a 78-diopter noncontact lens attributable to glaucoma associated with a reproducible glaucomatous VF defect on 2 occasions with the following features: (1) a glaucoma hemifield test outside normal limits, (2) a cluster of 3 nonedge, contiguous points not crossing the horizontal meridian on the pattern deviation plot, with a probability of <5% compared with age-matched normals (1 with a probability <1%), and (3) pattern standard deviation of <0.05. A VF was deemed reliable if it satisfied all of the following criteria: fixation losses of <20%, false positives and false negatives each <33%. The control group comprised patients from the same clinic population who were on follow-up solely as “glaucoma suspects” for optic disc evaluation (e.g., presence of a large cup-to-disc ratio [0.7]). None of these individuals had ocular hypertension or received any form of glaucoma-related medical, laser, or surgical treatment. All control participants had no VF defects (consistent with glaucoma or otherwise), had mean deviation (MD) better than 2 dB in each eye, and had logarithm of the minimum angle of resolution (logMAR) of 0.2 (Snellen 30/32). All participants had VA testing within the previous week and reliable VFs conducted in the last 3 months. During follow-up visits, VF testing was based on the Swedish Interactive Testing Algorithm (SITA FAST 24-2 algorithm) using a size III stimulus (Humphrey Visual Field Analyzer; Carl Zeiss Meditec, Inc., Dublin, CA). This study was approved by the Singhealth Institutional Research Board and conducted according to the tenets of the Declaration of Helsinki. Written informed consent was obtained from all participants.
Measurement of Covariates We measured VA monocularly in each eye with Early Treatment Diabetic Retinopathy Study charts at either 1 or 4 m.34 Participants used their habitual visual correction. Participants also underwent a refraction examination by study optometrists. Sociodemographic information (age, race, educational level, employment status, and number of persons residing in the same house) and data on medical comorbidities were collected from face-to-face interviews by trained research assistants.
Assessment of Psychosocial Functioning The Glau-QoL-36 questionnaire comprises 36 questions assessing the following 7 domains: anxiety, self-image, psychological wellbeing, confidence in health care, driving, burden of treatment, and daily life.31 Only the 4 domains pertaining to PF (anxiety, selfimage, psychological well-being, and confidence in health care) were evaluated in this study (Table 1). For all patients, masked research assistants administered the questionnaires before the clinical consult to eliminate the effect of clinician interaction on participants’ responses. All participants were asked to rate their difficulties and emotional responses to items, regardless of whether they were experienced recently. Responses were rated on a scale from 0 (not at all) to 4 (hugely) for anxiety, from 0 (completely true) to 3 (completely untrue) for the self-image domain, 0 (very often) to 4 (never) for the psychological well-being domain, and 0 (completely true) to 3 (completely untrue) in the confidence in health care domain (Table 1). The questionnaire was administered by trained bilingual research assistants in English and/or Mandarin.
Rasch Analysis Rasch analysis was conducted to determine the reliability, validity, and measurement characteristics of the 4 PF domains of the GlauQoL-36 questionnaire. Rasch analysis is a psychometric method that transforms raw ordinal scores into interval-level measurement (expressed in log of odds units [logits]). It calculates item difficulty (item measure) in relation to person ability (person measure) by placing both on the same linear continuum. A high person measure (in logits) indicates that the participant possesses a high level of the assessed latent trait (e.g., anxiety). Rasch analysis assesses the psychometric properties of a scale, such as how well items fit the underlying latent trait being measured (i.e., anxiety, self-image), how well items discriminate between the participants, how well item difficulty targets person ability, and the appropriateness of the response scale used.35 Winsteps (version 3.74; J.M. Linacre, Chicago, IL) was used to perform Rasch analysis on data from the 4 Glau-QoL-36 domains with the Andrich rating scale model. Each of the 4 PF domains of the Glau-QoL-36 scale was assessed for (1) item fit, (2) behavior of response categories (i.e., if higher categories represented better functioning), (3) measurement precision (minimum person separation index of 2.0), (4) unidimensionality, i.e., principal component analysis of >50% and an eigenvalue of <2.0, and (5) targeting items to the patient’s level of the 4 PF traits. Rasch scores (person measures) were generated after the structure of the responses and the items in the 4 PF domains of the Glau-QoL-36 questionnaire satisfied the five criteria of the Rasch model.
Statistical Analysis All statistical analyses were performed using STATA software version 12.1 (Stata Corp., College Station, TX). Characteristics of the study population and risk factors were examined using proportions, means, medians, percentiles, and standard deviations. Chi-square statistics and t tests were used for univariate associations. For each PF outcome, whether a nonlinear function resulted
495
Ophthalmology Volume 122, Number 3, March 2015 Table 1. Glaucoma Quality of Life-36 (Glau-QoL-36) Questionnaire: Dimensions, Item Content, and Response Options for Anxiety, Self-image, Psychological Well-being, and Confidence in Health Care Glau-QoL-36 Dimensions
Item Content
Anxiety
Feeling concerned about ocular pressure Fear of having to undergo an operation because of eye condition Problems in falling asleep because thinking about eye condition Thinking about the possibility of going blind
Self-image
Feeling that you didn’t get enough support when you first became ill Feeling of being treated differently Feeling physically or emotionally diminished Avoiding speaking about eye condition Feeling older because of eye condition Feeling that it is unfair that you are ill, rather than somebody else Feeling of being misunderstood Feeling of discouragement Feeling of frustration Feeling weak or vulnerable Feeling useless Regular check-ups with ophthalmologist being of help Following advice or instructions given by the ophthalmologist Medication being effective Having enough information about the eye disease and treatment
Psychological well-being
Confidence in health care
in a more reasonable fit than a linear trend was determined by inserting a quadratic term and determining significance using the F test. Visual predictors (i.e., glaucoma, VA, and VF MD in the better eye) were separately evaluated in multivariate linear or quadratic regression models adjusting for age, gender, other sociodemographic variables, and medical comorbidity. Where appropriate, b coefficients were converted into adjusted means and presented as percentages along with 95% confidence intervals (CIs) to facilitate interpretation. We used restricted cubic splines with knots corresponding to the 10th, 50th, and 90th percentiles36 of quantitative visual variables (i.e., VA and VF) significantly associated with PF outcomes to further characterize these relationships, namely, linear or quadratic. For quadratic associations, linear splines of the change in PF per unit change in VA or VF (i.e., PF/ logMAR or PF/MD slopes) were used to estimate specific VA or VF thresholds at which PF changed significantly. To ease the reader’s interpretation of logMAR VA, Snellen equivalents are shown (Table 2, available at www.aaojournal.org).
Results There were 192 bilateral glaucoma cases (82.7%) and 40 controls (17.3%). There were 108 primary open-angle glaucoma and 84 angle-closure glaucoma cases. Compared with our normal-sighted control group, those with glaucoma were generally older (mean age, 68.710.3 vs. 54.14.5 years; P < 0.001), were not working (71.5% vs. 40.0%; P < 0.001), reported lower educational levels (primary education or below, 57% vs. 29%; P < 0.001), and had a greater proportion with 2 medical comorbidities (16.4% vs. 2.5%; P ¼ 0.022; Table 3). As anticipated, compared with the control group, individuals with glaucoma had more severe VF loss in the better eye (MD, 8.89 vs. 1.19 dB; P < 0.001) and worse eye (17.70 vs. 2.82 dB; P < 0.001) and poorer logMAR VA in the better eye (0.11 vs. 0.01; P < 0.001) and worse eye (0.40 vs. 0.02; P < 0.001).
Validation of the 4 PF domains of the Glau-QoL-36 in Singapore Asians All participants answered all items in the Glau-QoL-36 questionnaire. For the anxiety and self-image domains, there was evidence
496
Response Options 0¼Not at all 1¼A little 2¼Moderately 3¼A lot 4¼Hugely 0¼Completely true for me 1¼Somewhat true for me 2¼Somewhat untrue for me 3¼Completely untrue for me 0¼Very often 1¼Often 2¼From time to time 3¼Seldom 4¼Never 0¼Completely true 1¼Somewhat true for me 2¼Somewhat untrue for me 3¼Completely untrue for me
of disordered thresholds (a sign that the response categories are not working as intended). Merging adjacent response categories subsequently resulted in ordered thresholds. Items that demonstrated misfit, that is, infit residual values of >1.3, were iteratively removed to improve fit statistics. These were Q13 (“Do you feel like you are the victim of an injustice?”), Q14 (“Do you feel misunderstood?”), and Q15 (“Do you feel discouraged?”) for the psychological well-being scale and Q25 (“Are you afraid of surgery?”) for the anxiety scale. The person separation index ranged from 0.68 (anxiety) to 1.27 (psychological well-being), suggesting that the 4 PF scales could only modestly discriminate between levels of difficulty experienced by the participants. However, the principal component analysis of the residuals for each of the 4 PF scales explained >50% of the raw variance (range, 51.2%e58.8%), and the unexplained variance by the first contrast of the residuals for all scales was <2.0 eigenvalues. These 2 parameters indicate that the 4 assessed PF traits were unidimensional (measuring a single latent trait). The mean value standard deviation of person and item fit residual values ranged from 4.842.82 (psychological well-being) to þ0.271.26 (anxiety). These data suggest that the scale item difficulty in relation to person ability was suboptimal and participants were too able for the level of item difficulty in the scale. There was no evidence of differential item functioning for any items across the 4 PF scales for age, gender, and systemic comorbidities.
Impact of Glaucoma, VA, and VF Losses in the Better Eye on PF Glaucoma cases, compared with controls, had significantly lower mean scores for anxiety (1.801.80 vs. 4.290.68; P < 0.001), self-image (1.461.67 vs. 4.090; P < 0.001), psychological well-being (4.402.92 vs. 6.950.35; P < 0.001), and confidence in health care (3.841.84 vs. 6.150; P < 0.001). In multivariate models adjusting for sociodemographic factors and medical comorbidities (Table 4), glaucoma in the better eye was independently associated with increased anxiety (b ¼ 3.09 [95% CI, 3.85 to 2.31]; P < 0.001), lower self-image (b ¼ 2.79 [95% CI, 3.40 to 2.18]; P < 0.001), psychological well-being
Chan et al
Psychosocial Functioning in Glaucoma
Table 3. Visual, Demographic, and Health Characteristics of Controls and Bilateral Glaucoma Cases Controls (n [ 40) Demographics Age (yrs) Female gender (%) Education level, no. (%) Primary and below Secondary/technical diploma Pre-university or higher Employed, no. (%) Housing Public, no. (%) Private, no. (%) No. of additional persons in same household Health (medical comorbidities) <2 2 Vision (logMAR VA) Worse eye Better eye VF loss MD (dB) Worse Eye Better Eye Glaucoma subtype, no. (%) Primary open-angle glaucoma Primary angle closure glaucoma
54.13 (4.48) 18 (45.0)
Glaucoma (n [ 192) 68.67 (10.29) 67 (34.9)
P <0.001 0.228
(57.0) (28.9) (14.1) (28.5)
<0.001
36 (90.0) 4 (10.0) 3.73 (1.15)
178 (92.7) 14 (7.3) 3.34 (1.40)
0.765 0.116
39 (97.5) 1 (2.5)
158 (83.6) 31 (16.4)
0.022
0.02 (0.04) 0.01 (0.03)
0.40 (0.81) 0.11 (0.11)
<0.001 <0.001
2.82 (5.55) 1.19 (3.02)
17.70 (8.99) 8.89 (6.52)
<0.001 <0.001
11 10 17 24
(29.0) (26.3) (44.7) (60.0)
e e
73 37 18 37
90 (46.9) 84 (43.8)
<0.001
-
Data are mean (standard deviation) unless otherwise indicated. logMAR ¼ logarithm of minimum angle of resolution; MD ¼ mean deviation; VA ¼ visual acuity; VF ¼ visual field.
(b ¼ 3.01 [95% CI, 4.19 to 1.82]; P < 0.001), and confidence in health care (b ¼ 1.66 [95% CI, 2.37 to 0.94]; P < 0.001) compared with controls. In terms of percentage reductions (Table 5, available at www.aaojournal.org), glaucoma was associated with reduced PF compared with controls (63.3% for anxiety, 71.0% for self-image, 38.3% for psychological wellbeing, and 32.4% for confidence in health care). In multivariate models, worse best-corrected VA in the better eye was linearly associated with lower self-image (b ¼ 0.27 [95% CI, 0.45 to 0.08]; P ¼ 0.005) and greater anxiety (b ¼ 0.30 [95% CI, 0.52 to 0.07]; P ¼ 0.009; Table 4). Compared with controls, glaucoma patients in each tertile of VA loss had significantly worse PF for all outcomes, ranging from 26.8% to 81.5% (all P < 0.001; Table 5; Fig 1, available at www.aaojournal.org). Greater VF loss in the better eye was quadratically associated (i.e., increased with the square of the PF outcome) with worse anxiety (b ¼ 0.04 [95% CI, 0.07 to 0.02]; P ¼ 0.002), selfimage (b ¼ 0.03 [95% CI, 0.05 to 0.01]; P ¼ 0.002), and less confidence in health care (b ¼ 0.03 [95% CI, 0.55 to 0.10]; P ¼ 0.002; Table 4). Compared with controls, glaucoma patients in each tertile of VF loss showed significantly worse PF for all outcomes, ranging between 26.0% to 76.1% (all P < 0.001; Table 5; Fig 2, available at www.aaojournal.org). To characterize how PF varied across the spectrum of VA and VF losses, restricted cubic spline analysis was conducted for the above significant associations with VA (Fig 3) and VF loss (Fig 4) in the better eye. Self-image and anxiety decreased linearly with reduced best-corrected VA. Self-image decreased going from 0 to 10 dB but plateaued at 10 dB or more severe levels of VF loss. For self-image, mean adjusted PF scores at >10 and 10 dB levels were 2.32 and 1.08 logits, respectively (PF/MD slope difference, 0.25; P < 0.001). Going from 0 to 12.1 dB, anxiety increased, but it
improved beyond 12.1 dB. For anxiety, the PF/MD slopes at better (>12.1 dB) and poorer MD (12.1 dB) were 0.2 and 0.07, respectively (P ¼ 0.003). Similarly, confidence in health care decreased going from 0 to 9.3 dB and improved beyond 9.3 dB. The PF/MD slopes at better (>9.3 dB) and poorer MD (9.3 dB) were 0.12 and 0.07, respectively (P ¼ 0.008).
Discussion This study characterizes the impact of glaucoma and associated VA and VF losses on specific PF outcomes. Glaucoma patients had 63% greater anxiety, 71% lower self-image, 38% less psychological well-being, and 32% reduced confidence in health care compared with controls. The results indicate that in early stage glaucoma (i.e., at mild VF loss), when VF worsens, anxiety increases, whereas self-image and confidence in health care are reduced (Fig 4). In late-stage glaucoma (i.e., when VA is eventually affected), only anxiety and self-image worsen as VA worsens (Fig 3). Hence, preserving both VA and VF in glaucoma patients is important. This study adds to the limited body of evidence on psychological, social, and emotional functioning in glaucoma.13e22 In this study, there were significant associations between glaucoma and VA and VF losses and reduced PF, namely, increased anxiety, reduced self-image, and less confidence in health care. Several earlier studies found no correlation between VA or VF loss with patient perceptions of treatment,19 anxiety,16 or fear of blindness.18 However, Parrish et al21 and Gutierrez et al22 determined that VF losses
497
Ophthalmology Volume 122, Number 3, March 2015 Table 4. Impact of Bilateral Glaucoma and Visual Acuity and Visual Field Loss in the Better Eye on Psychosocial Functioning Using the Glaucoma Quality of Life 36 Questionnaire Glaucoma (Present)
b (95% CI) Unadjusted Anxiety Self-image Psychological well-being Confidence in health care Adjusted* Anxiety Self-image Psychological well-being Confidence in health care
VA (0.1 logMAR Worse)
b (95% CI)
P
VF Loss, MD (5 dB Worse)
b (95% CI)
P
2.49 2.63 2.55 2.31
(3.06 (3.15 (3.46 (2.89
to to to to
1.92) 2.11) 1.64) 1.74)
<0.001 <0.001 <0.001 <0.001
0.29 0.34 0.33 0.19
(0.46 (0.50 (0.58 (0.36
to to to to
0.13) 0.18) 0.08) 0.02)
0.001y <0.001y 0.009y 0.025y
0.03 0.04 0.31 0.05
(0.05 (0.05 (0.57 (0.07
to to to to
0.01) 0.02) 0.044) 0.03)
3.09 2.79 3.01 1.66
(3.85 (3.40 (4.19 (2.37
to to to to
2.31) 2.18) 1.82) 0.94)
<0.001 <0.001 <0.001 <0.001
0.30 0.27 0.29 0.07
(0.52 (0.45 (0.60 (0.26
to to to to
0.07) 0.08) 0.03) 0.12)
0.009y 0.005y 0.074y 0.493y
0.04 0.03 0.23 0.03
(0.07 (0.05 (0.55 (0.05
to to to to
0.02) 0.01) 0.10) 0.01)
P 0.004z <0.001z 0.022y <0.001z 0.002z 0.002z 0.175y 0.007z
CI ¼ confidence interval; logMAR ¼ logarithm of the minimum angle of resolution; MD ¼ mean deviation; VA ¼ visual acuity; VF ¼ visual field. *Adjusted for age, employment status, educational level, no. of persons in the house, and medical comorbidities. y b and P values indicated for linear relationships between psychosocial functioning domains with VA or VF. z b and P values indicated for quadratic relationships between psychosocial functioning domains with VA or VF.
were at least modestly correlated with vision-specific dependency, role difficulties, social and emotional well-being, and mental health on the National Eye Institute VisionFunction Questionnaire. The different study findings may be due to the dissimilar spectrums of glaucoma severity between studies. Our study shows that 3 of the 4 PF outcomes assessed (anxiety, self-image, and confidence in health care) remain largely unchanged in advanced glaucoma despite worsening VA or VF loss (Figs 1 and 2). This may explain why other studies involving individuals at these more severe disease levels generally found nonsignificant or weak associations. A second reason could be the higher sensitivity of the glaucoma-specific Glau-QoL-36 questionnaire compared with health- or vision-related QoL questionnaires in eliciting concerns directly related to glaucoma. Glaucoma had a significant negative impact on all 4 PF outcomes, even after adjusting for sociodemographic and health factors. This effect could be mediated indirectly through glaucoma-related impairment in key QoL subtraits, such as daily functioning, including use of topical medication and mobility.8e11 For example, the cost and compliance of long-term glaucoma treatment and monitoring could reduce PF. Prior research has indicated that patients with glaucoma have an increased fear of falling, possibly because of physical limitations imposed by visual disability, thereby affecting anxiety, self-image, and psychological well-being outcomes.37 Driving cessation is strongly impacted by glaucoma,8,12 consequently affecting PF. Understanding to what extent PF outcomes are linked with activity limitation may be important because this may suggest that targeted strategies to assist glaucoma patients with daily activities could potentially improve PF. Even mild VF losses were associated with increased anxiety and reduced self-image and confidence in health care (Fig 4). Ophthalmologists managing patients with early glaucoma should recognize that patients may have valid concerns pertaining to intraocular pressure, possibility of blindness, being emotionally diminished, or being treated differently (Table 1) to engage patient trust. In early disease, other visual function aspects (e.g., contrast sensitivity) may
498
already be impaired. The negative effect of mild VF loss on PF is consistent with the Los Angeles Latino Eye Study,7 where small changes in VF resulted in observable reductions in HRQoL. This finding underscores the importance of identifying glaucoma patients early in their disease to preserve existing visual function. Anxiety increased and confidence in health care decreased with worsening VF initially but improved in more severe disease beyond 12.1 and 9.3 dB, respectively. Loss of VF was associated with reduced self-image up to about 10 dB but remained unchanged beyond 10 dB. Such “quadratic” associations between visual function and HRQoL outcomes have not been previously reported because of assumptions that all significant associations followed linear trends.6e12,37 Several hypotheses could explain these results. Crabb et al38 reported that patients with mild to moderate glaucoma commonly perceived blurred or missing patches, whereas individuals with advanced glaucoma perceive a tunnel effect with blurred edges. When early glaucoma progresses, patients may experience greater anxiety and concerns about self-image and loss of confidence in health care, perhaps relating to an increase in the number or depth of scotoma. However, individuals with advanced glaucoma and tunnel vision fields may not notice small changes in VF, especially if progression occurs in the periphery.28,29 Increased adaptation to vision loss associated with glaucoma when glaucoma is advanced could also explain reduced anxiety and self-image concerns. Patients with more severe disease could have had more aggressive intraocular pressureelowering therapy, e.g., glaucoma surgery, possibly accounting for better confidence in health care. Ophthalmologists need to be aware of increased anxiety and reductions in self-image and confidence in health care as patients lose VF up to about 9 to 12 dB. In addition, improved PF in advanced glaucoma may explain why patient motivation toward treatment is reduced, such as reduced medication compliance. Loss of VA was associated with increased anxiety and reduced self-image (Fig 3). Worsening VA suggests increasing involvement of central fixation, as is present in
Chan et al
Psychosocial Functioning in Glaucoma
Figure 3. Relationship between psychosocial functioning and logarithm of the minimum angle of resolution (logMAR) visual acuity in the better eye in bilateral glaucoma in a restricted cubic spline plot for (A) anxiety, (B) self-image, and (C) psychological well-being. (Black line represents the estimate, and the shaded area delineates the upper and lower limits of the 95% confidence intervals (CIs). Knots are set at the 10th, 50th, and 90th percentiles of logMAR visual acuity [0, 0.02, and 0.26)]. Snellen visual acuity is indicated in parentheses on the axes). logit ¼ log of odds unit.
advanced glaucoma.29,30 Monitoring VA changes in glaucoma patients, particularly those with advanced disease, could be important because anxiety and self-image outcomes worsen whenever VA losses occur. These findings suggest that it is important to consider VA loss in addition to VF loss in terms of the overall impact on PF; although anxiety and self-image levels reach a plateau as VF worsens, if VA also worsens, it is likely that PF will be affected. In advanced disease, VA could be a better indicator of PF than VF. Strengths of this study include the use of Rasch analysis to validate 4 specific domains of the Glau-QoL-36 questionnaire and produce linear interval measures for each of the 4 PF domains. This study also used locally weighted scatterplot smoothing analysis to characterize the trend in PF over the spectrum of VA and VF losses. Standard regression techniques used in several QoL studies assume linear relationships and do not adequately capture information over the entire disease trajectory.6e12,37
Figure 4. Relationship between psychosocial functioning and visual field loss in the better eye in bilateral glaucoma in a restricted cubic spline plot for (A) anxiety, (B) self-image, and (C) confidence in health care. (Black line represents the estimate, and the shaded area delineates the upper and lower limits of the 95% confidence intervals. Knots are set at 10th, 50th, and 90th percentiles of visual field loss (0.14, 5.17, and 16.67). Logit ¼ log of odds unit; MD ¼ mean deviation.
Several limitations of this study should be discussed. Because this was a cross-sectional study, investigation of causation was not possible. This study involved patients in a tertiary ophthalmic center and may not accurately represent
499
Ophthalmology Volume 122, Number 3, March 2015 individuals with glaucoma in the population, many of whom are undiagnosed. The control group comprised glaucoma suspects, which could underestimate the impact of glaucoma on PF because the psychological burden of “possibly having glaucoma” in the glaucoma suspect control group may narrow the differences in PF than if a group with only mild eye disease was used as the control. Nevertheless, we still found a significant difference in all PF aspects between the glaucoma and control groups. Glaucoma suspects have been included as controls in similar QoL studies9e11,37 because population-based individuals may have other unmeasured confounding factors or may have “supranormal” PF. The knowledge of “potential glaucoma” is likely to have had only a small effect because VA and VF losses were clearly related to several PF outcomes. There was almost twice the number of female than male glaucoma patients because of the sociodemographic characteristics of our hospital-based sample. However, this is unlikely to bias the findings because independent relationships were present after controlling for gender. Although we considered only better eye VA and VF, rather than an “integrated VF” or binocular VA, strong correlations between monocular and binocular VF39 and VA40 measures still exist. Monocular visual parameters are readily available in the clinical setting and thus are more easily integrated into clinical decision making. In conclusion, this study demonstrates the adverse impact of glaucoma on PF. Losses in VF in early glaucoma significantly worsen anxiety and self-image, emphasizing a need for early identification of glaucoma. In advanced glaucoma, if VA worsens, anxiety and self-image outcomes will be adversely affected. The association between VA and PF supports the use of VA monitoring in glaucoma patients at all stages of severity. Further research is required to determine the efficacy of various therapeutic and rehabilitative strategies in improving PF for glaucoma patients.
7.
8.
9. 10.
11. 12. 13. 14.
15. 16. 17.
Acknowledgments. The authors thank Dr Eva Fenwick, from the University of Melbourne, for her assistance in the preparation of this manuscript.
18.
References
19.
1. Quigley HA, Broman AT. The number of people with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol 2006;90: 262–7. 2. Eye Diseases Prevalence Research Group. Prevalence of openangle glaucoma among adults in the United States. Arch Ophthalmol 2004;122:532–8. 3. Shen SY, Wong TY, Foster PJ, et al. The prevalence and types of glaucoma in Malay people: the Singapore Malay Eye Study. Invest Ophthalmol Vis Sci 2008;49:3846–51. 4. Foster PJ, Oen FT, Machin D, et al. The prevalence of glaucoma in Chinese residents of Singapore: a cross-sectional population survey. Arch Ophthalmol 2000;118:1105–11. 5. Narayanaswamy A, Baskaran M, Zheng Y, et al. The prevalence and types of glaucoma in an urban Indian population: the Singapore Indian Eye Study. Invest Ophthalmol Vis Sci 2013;54:4621–7. 6. Chan EW, Chiang PP, Wong TY, et al. Impact of glaucoma severity and laterality on vision-specific functioning: the
500
20.
21. 22. 23. 24. 25.
Singapore Malay Eye Study. Invest Ophthalmol Vis Sci 2013;54:1169–75. McKean-Cowdin R, Wang Y, Wu J, et al; Los Angeles Latino Eye Study Group. Impact of visual field loss and health-related quality of life in glaucoma: the Los Angeles Latino Eye Study. Ophthalmology 2008;115:941–8. West SK, Rubin GS, Broman AT, et al; SEE Project Team. How does visual impairment affect performance on tasks of everyday life? The SEE Project. Arch Ophthalmol 2002;120: 774–80. Ramulu PY, Maul E, Hochberg C, et al. Real-world assessment of physical activity in glaucoma using an accelerometer. Ophthalmology 2012;119:1159–66. Hochberg C, Maul E, Chan ES, et al. Association of vision loss in glaucoma and age-related macular degeneration with IADL disability. Invest Ophthalmol Vis Sci 2012;53: 3201–6. Ramulu PY, Swenor BK, Jefferys JL, et al. Difficulty with outloud and silent reading in glaucoma. Invest Ophthalmol Vis Sci 2013;54:666–72. Ramulu PY, West SK, Munoz B, et al. Driving cessation and driving limitation in glaucoma: the Salisbury Eye Evaluation Project. Ophthalmology 2009;116:1846–53. Wilson MR, Coleman AL, Yu F, et al. Depression in patients with glaucoma as measured by self-report surveys. Ophthalmology 2002;109:1018–22. Wang SY, Singh K, Lin SC. Prevalence and predictors of depression among participants with glaucoma in a nationally representative population sample. Am J Ophthalmol 2012;154: 436–44. Bramley T, Peeples P, Walt JG, et al. Impact of vision loss on costs and outcomes in Medicare beneficiaries with glaucoma. Arch Ophthalmol 2008;126:849–56. Mabuchi F, Yoshimura K, Kashiwagi K, et al. Risk factors for anxiety and depression in patients with glaucoma. Br J Ophthalmol 2012;96:821–5. Odberg T, Jakobsen JE, Hultgren SJ, Halseide R. The impact of glaucoma on the quality of life of patients in Norway. I. Results from a self-administered questionnaire. Acta Ophthalmol Scand 2001;79:116–20. Janz NK, Wren PA, Lichter PR, et al; CIGTS Group. Quality of life in newly diagnosed glaucoma patients: the Collaborative Initial Glaucoma Treatment Study. Ophthalmology 2001;108:887–97. Jampel HD, Frick KD, Janz NK, et al. Depression and mood indicators in newly diagnosed glaucoma patients. Am J Ophthalmol 2007;144:238–44. Bhargava JS, Patel B, Foss AJ, et al. Views of glaucoma patients on aspects of their treatment: an assessment of patient preference by conjoint analysis. Invest Ophthalmol Vis Sci 2006;47:2885–8. Parrish RK II, Gedde SJ, Scott IU, et al. Visual function and quality of life among patients with glaucoma. Arch Ophthalmol 1997;115:1447–55. Gutierrez P, Wilson RW, Johnson C, et al. Influence of glaucomatous visual field loss on health-related quality of life. Arch Ophthalmol 1997;115:777–84. WHOQOL Group. The World Health Organization quality of life assessment (WHOQOL): position paper from the World Health Organization. Soc Sci Med 1995;41:1403–9. Astrom M, Asplund K, Astrom T. Psychosocial function and life satisfaction after stroke. Stroke 1992;23:527–31. Carels RA, Musher-Eizenman D, Cacciapaglia H, et al. Psychosocial functioning and physical symptoms in heart failure
Chan et al
26. 27. 28. 29. 30. 31.
32.
Psychosocial Functioning in Glaucoma
patients: a within-individual approach. J Psychosom Res 2004;56:95–101. Friedman LC, Kalidas M, Elledge R, et al. Optimism, social support and psychosocial functioning among women with breast cancer. Psychooncology 2006;15:595–603. Ramulu PY. Glaucoma and disability: which tasks are affected, and at what stage of disease. Curr Opin Ophthalmol 2009;20:92–8. Nelson P, Aspinall P, Papasouliotis O, et al. Quality of life in glaucoma and its relationship with visual function. J Glaucoma 2003;12:139–50. Asaoka R. The relationship between visual acuity and central visual field sensitivity in advanced glaucoma [letter]. Br J Ophthalmol 2013;97:1355–6. Portney GL, Hanible JA. Visual acuity and field loss in advanced glaucoma. Ann Ophthalmol 1974;6:1147–50. Bechetoille Arnould B, Bron A, et al. Measurement of healthrelated quality of life with glaucoma: validation of the GlauQOL 36-item questionnaire. Acta Ophthalmol 2008;86: 71–80. Chylack LT Jr, Wolfe JK, Singer DM, et al. Longitudinal Study of Cataract Study Group. The Lens Opacities Classification System III. Arch Ophthalmol 1993;111:831–6.
33. Hodapp E, Parrish RK II, Anderson DR. Clinical Decisions in Glaucoma. St Louis: Mosby; 1993:52–61. 34. Bailey IL, Bullimore MA, Raasch TW, Taylor HR. Clinical grading and the effects of scaling. Invest Ophthalmol Vis Sci 1991;32:422–32. 35. Lamoureux E, Pesudovs K. Vision-specific quality-of-life research: a need to improve the quality. Am J Ophthalmol 2011;151:195–7. 36. Durrleman S, Simon R. Flexible regression models with cubic splines. Stat Med 1989;8:551–61. 37. Ramulu PY, van Landingham SW, Massof RW, et al. Fear of falling and visual field loss from glaucoma. Ophthalmology 2012;119:1352–8. 38. Crabb DP, Smith ND, Glen FC, et al. How does glaucoma look? Patient perception of visual field loss. Ophthalmology 2013;120:1120–6. 39. Arora KS, Boland MV, Friedman DS, et al. The relationship between better-eye and integrated visual field mean deviation and visual disability. Ophthalmology 2013;120:2476–84. 40. Rubin GS, Munoz B, Bandeen-Roche K, West SK. Monocular versus binocular visual acuity as measures of visual impairment and predictors of visual disability. Invest Ophthalmol Vis Sci 2000;41:3327–34.
Footnotes and Financial Disclosures Originally received: January 17, 2014. Final revision: September 19, 2014. Accepted: September 24, 2014. Available online: November 8, 2014.
Financial Disclosure(s): The authors have no proprietary or commercial interest in any materials discussed in this article. Manuscript no. 2014-89.
1
Department of Ophthalmology, National University Health System and National University of Singapore, Singapore.
2
Singapore Eye Research Institute, Singapore National Eye Centre, Singapore.
3
Centre for Eye Research Australia, The Royal Victorian Eye and Ear Hospital, University of Melbourne, Australia.
4
DUKE-NUS Graduate Medical School, Singapore.
Abbreviations and Acronyms: CI ¼ confidence interval; Glau-QoL-36 ¼ Glaucoma Quality of Life 36; HRQoL ¼ health-related quality of life; logits ¼ log of odds units; logMAR ¼ logarithm of the minimum angle of resolution; MD ¼ mean deviation; PF ¼ psychosocial functioning; VA ¼ visual acuity; VF ¼ visual field. Correspondence: Ecosse Lamoureux, MSc, PhD, Office of Clinical Sciences, Duke-NUS Graduate Medical School Singapore, 8 College Road Level 4, Singapore 169857. E-mail: [email protected].
501