Visual impairment may limit individual's performance of everyday tasks such as eating, cooking, reading and writing, traveling from place to place in the community, and communicating with others.1–3 Patients may feel frustrated and experience symptoms of depression when they are unable to perform these activities independently.4–7 Low-vision services may increase functional visual ability, independence, and quality of life. However, programs differ from one another in the range and intensity of services that they provide. Despite the variation in service delivery, these programs may be able to demonstrate successful outcomes but with success limited to the specific services that they provide.8 Clinical trials are needed to compare the low-vision outcomes of different service delivery models.
Four clinical trials conducted outside of the United States compared programs that offer rehabilitation to those that dispense low-vision devices without therapy.9–13 Most recently, Pearce et al.9 studied the efficacy of low-vision device training in a hospital-based United Kingdom low-vision clinic. After their low-vision examination with an optometrist, patients were randomized to an intervention group who received an additional session to review low-vision device use or a control group. Although patient-reported difficulty with daily tasks decreased after a low-vision examination, no further improvement occurred after the second visit to instruct patients in use of low-vision devices. Reeves et al.10 randomized patients from a United Kingdom low-vision clinic to one of three groups: an enhanced low-vision rehabilitation model that provided home-based low-vision rehabilitation, a conventional low-vision rehabilitation model based in a hospital clinic, or a group that received home visits without rehabilitation to control for additional contact time. These investigators found that enhanced low-vision rehabilitation was no more effective than the conventional model. Controlled before and after designs were also used by de Boer et al.11 in the Netherlands and La Grow12 in New Zealand to compare outcomes of basic and multidisciplinary or comprehensive services. Differences in outcomes were not reported in either study.
In the United States, the Veterans Affairs Low-vision Intervention Trial was conducted from September 2010 to July 2014 at nine Veterans Affairs medical facilities to determine if low-vision devices and rehabilitation with a therapist (low-vision rehabilitation) were more effective than low-vision devices dispensed without therapy (basic low vision) for patients with macular diseases and best-corrected visual acuity in the better eye of 20/50 to 20/200.13–15 Both treatments were found to be effective in the trial, but low-vision rehabilitation was more effective only for patients with visual acuity worse than 20/63 to 20/200. The study protocol, outcomes of the trial (baseline to 4 months), and an economic evaluation of the treatments provided were reported previously.13–15
We observed 255 participants from Veterans Affairs Low-vision Intervention Trial II after the trial ended (4 months from baseline) until 1-year follow-up. The present study compares changes in self-reported visual ability for patients randomized to the low-vision rehabilitation and basic low-vision groups (baseline to 4 months, 4 months to 1 year, baseline to 1 year) and examines the clinical or interventional factors that predict these changes between and within treatment groups.
Conduct of the Study
Masked interviewers administered the 48-item Veterans Affairs Low-vision Visual Functioning Questionnaire,16–20 the 36-item Short-Form Health Survey,21 and the EuroQol-5D22 by telephone at baseline before randomization and at 4-month and 1-year follow-ups. During the interviews, patients also measured their near visual acuity using the Lighthouse number card with the low-vision device prescribed for near spot-checking.
The 48-item Veterans Affairs Low-vision Visual Functioning Questionnaire was used to assess visual function.16–20 Patients used the 48-item Veterans Affairs Low-vision Visual Functioning Questionnaire to rate their difficulty performing each of 48 daily activities using the ordered response categories: (1) not difficult, (2) moderately difficult, (3) extremely difficult, and (4) impossible. Patients were also permitted to respond that they do not perform an activity for nonvisual reasons.
The 36-item Short-Form Health Survey and EuroQol-5D were used to assess quality of life. The 36-item Short-Form Health Survey includes scales for physical functioning, physical role limitations, bodily pain, vitality, social functioning, emotional role limitations, mental health, general health, and physical and mental components.21 The EuroQol-5D includes a visual analog scale and descriptive system that comprises five dimensions: anxiety/depression, pain/discomfort, usual activities, self-care, and mobility.22
Fig. 1 presents a flowchart of patient participation in the present study. Originally, 323 participants were randomized into either the low-vision rehabilitation group or the basic low-vision service group. Among those participants, 68 of them did not complete the 1-year follow-up interview owing to death (31), withdrawal of consent (11), lost to follow-up (19), and refusal (7). The present study includes a total of 255 patients who completed the 1-year follow-up: 135 in the low-vision rehabilitation group and 120 in the basic low-vision group.
Patients randomized to the low-vision rehabilitation group received rehabilitation with a therapist; those randomized to the basic low-vision group received low-vision devices dispensed without therapy. After Veterans Affairs Low-vision Intervention Trial II ended (4 months from baseline), patients in both treatment groups were eligible to receive usual Veterans Affairs low-vision care. Between 4 months and 1 year, 3.7% (5/135) of the low-vision rehabilitation group and 5.8% (7/120) of the basic low-vision group were referred to Veterans Affairs blind rehabilitation centers; 22.2% (30/135) of the low-vision rehabilitation group and 40.8% (49/120) of the basic low-vision group received additional low-vision services from clinicians at the participating sites. These services included orientation and mobility evaluation and training, home therapy, additional low-vision optometry evaluation, additional low-vision therapy, and homework assignments.
The baseline demographic and clinical characteristics were compared between the low-vision rehabilitation group and the basic low-vision group using a two-sample t test for continuous variables and the χ2 test for categorical variables. To assess the bias in the population studied, the baseline characteristics were also compared between the patients who were followed up for 1 year and the patients who dropped out before the study ended using the same methods.
Visual ability measured by the 48-item Veterans Affairs Low-vision Visual Functioning Questionnaire was estimated in logits (log odds) by Rasch analysis of responses to all 48 items.13 Rasch analysis of responses to subsets of items was used to estimate visual ability (reading, mobility, visual information processing, visual motor skills, and overall ability) for each administration of the 48-item Veterans Affairs Low-vision Visual Functioning Questionnaire.13 Higher scores indicate more visual ability (i.e., less difficulty performing activities). A 0.14 logit change in visual ability corresponds to the ability change expected from a one-line change in logMAR visual acuity.23
Changes in each functional domain from baseline to 4 months, 4 months to 1 year, and from baseline to 1 year within each group were compared using paired t tests. We compared differences in changes from baseline to 1 year between the low-vision rehabilitation and the basic low-vision group using the two-sample t test. Effect size,24 defined as the difference between groups in the mean changes divided by the pooled standard deviation of the changes, was calculated to compare the relative magnitude of the mean changes in visual ability from baseline to 1 year.
The subscale scores for 36-item Short-Form Health Survey (0 to 100) were generated using the Medical Outcomes Study 36-item Short-Form Health Survey (V2) scoring system, which generates eight subscales and two summary scales.25 The EuroQol-5D health states were converted to a single summary health utility index (0 to 1) using the U.S. value set and a formula that attaches weights to each of the levels in each dimension.26 Higher scores in the 36-item Short-Form Health Survey and higher indices in the EuroQol-5D indicate better quality of life.
Linear regression models with backward selection were used to determine the predictors of change (loss) from 4 months to 1 year in reading ability and overall visual ability. Before modeling, the correlation coefficients were assessed among the potential independent variables including age, treatment group, reading ability score at 4 months, near visual acuity at 4 months, loss of near visual acuity from 4 months to 1 year, use of additional low-vision services after the 4-month follow-up, 36-item Short-Form Health Survey physical and mental summary scores at 4 months, and EuroQol-5D utility index at 4 months. We included age, low-vision treatment group, 4-month reading ability score, loss of near visual acuity from 4 months to 1 year, use of additional services, and 4-month EuroQol-5D utility index in the models. The 36-item Short-Form Health Survey summary scores were excluded because of significant correlation with the EuroQol-5D utility index. The final models were generated after model selection procedures were completed. The predictors of loss in overall ability from 4 months to 1 year were determined using the same modeling procedures.
All analyses were two-sided. SAS software (version 9.4; SAS Institute Inc., Cary, NC) was used to perform all analyses.27
Table 1 presents two comparisons of the baseline demographic and clinical characteristics: low-vision rehabilitation group versus basic low-vision group among the patients who completed the 1-year follow-up and patients who completed the 1-year follow-up interview versus the patients who dropped out of the study before the end of the 1-year follow-up. There were no significant differences in the baseline characteristics between those in low-vision rehabilitation and basic low-vision treatment groups for those who completed the 1-year follow-up study except race and the 36-item Short-Form Health Survey mental component score. Mean (standard deviation) age was 79.0 (10.9) years for those in low-vision rehabilitation and 79.9 (9.7) years for those in basic low vision. Approximately 97% of the patients in both groups were male. The mean (standard deviation) best-corrected distance visual acuity in the better eye was 0.6 (0.2) logMAR (20/80 Snellen equivalent) for both groups, and the mean (standard deviation) near visual acuity was 0.3 (0.3) logMAR in the low-vision rehabilitation group and 0.3 (0.4) logMAR in the basic low-vision group. Diagnoses of patients in the low-vision rehabilitation group were 31.1% (42/135) nonexudative macular degeneration and 31.1% (42/135) exudative macular degeneration; diagnoses of patients in the basic low-vision group were 31.7% (38/120) nonexudative macular degeneration and 40.0% (48/120) exudative macular degeneration. The comparison of baseline characteristics between patients who completed the 1-year follow-up versus the patients who did not complete the 1-year follow-up showed no significant differences except for race, walking assistance, and problems with memory.
Table 2 shows the mean (standard deviation) within-group changes in visual function from baseline to 4 months and from 4 months to 1 year. Visual ability improved significantly in all domains for the low-vision rehabilitation group and in all domains except for mobility for the basic low-vision group from baseline to 4 months. However, significant losses occurred from 4 months to 1 year in reading ability in both groups (low-vision rehabilitation group, −0.64 [1.2] logit; 95% confidence interval, −0.84 to −0.44 logit; P < .0001; basic low-vision group, −0.63 [1.4] logit; 95% confidence interval, −0.88 to −0.38 logit; P < .0001) and the low-vision rehabilitation group lost overall visual ability (−0.20 [0.8] logit; 95% confidence interval, −0.34 to −0.06 logit; P = .005).
Table 3 presents the mean (standard deviation) changes in visual function measured using the 48-item Veterans Affairs Low-vision Visual Functioning Questionnaire from baseline to 1 year within each group and the comparison of differences in mean changes from baseline to 1 year between the low-vision rehabilitation and basic low-vision groups. Both groups experienced significant improvement in all measures of visual ability from baseline to 1 year. Compared with the basic low-vision group, the low-vision rehabilitation group had greater improvement in visual information processing (0.38 logit; 95% confidence interval, 0.10 to 0.65; P = .01). The magnitude of the treatment effects was small, ranging from 0.09 to 0.34; the treatment effect is 0.34 for visual information processing (Table 3).
Near Visual Acuity
The mean (standard deviation) near visual acuity was 0.3 (0.3) logMAR in the low-vision rehabilitation group and 0.3 (0.4) logMAR in the basic low-vision group at baseline. There were no significant changes in mean near visual acuity from baseline to 4 months in either group. From 4 months to 1-year follow-up, both groups experienced a significant loss in near visual acuity. The mean reduction in near visual acuity (standard deviation) was 0.25 (0.52) logMAR in the low-vision rehabilitation group (P < .001) and 0.14 (0.48) logMAR in the basic low-vision group (P < .01). However, the difference between the two groups was not statistically significant.
Predictors of Loss in Visual Reading Ability and Overall Visual Ability
The predictors of losses in reading ability and overall visual ability from 4 months to 1 year were assessed for all patients and by treatment groups. Patients who had lower 4-month reading ability scores and utility indices initially and those who lost near visual acuity from 4 months to 1 year were more likely to lose their reading ability at 1 year. Being healthier had the effect of increasing the reading ability score by 1.09 logit (P < .001) (Table 4A). Lower reading ability at 4 months and loss of near visual acuity from 4 months to 1 year would reduce the reading ability scores by 0.35 logit (P < .0001) and 0.48 logit (P < .001), respectively (Table 4A). The predictors of loss in reading ability did not differ by treatment groups. For overall visual ability, the poor 4-month overall ability score had the effect of decreasing the overall ability score by 0.3 logit (P < .0001) (Table 4B). The findings were consistent when the modeling was conducted for each group.
This study compared changes in self-reported visual ability for patients randomized to the low-vision rehabilitation and basic low-vision groups (baseline to 4 months, 4 months to 1 year, baseline to 1 year) and examined the clinical or interventional factors that predict these changes between and within treatment groups. Both treatment groups demonstrated significant improvements in functional visual ability (reading, mobility, visual information processing, visual motor, and overall) from baseline to 1-year follow-up. When the differences in mean changes of 48-item Veterans Affairs Low-vision Visual Functioning Questionnaire scores between groups were compared, significant differences were only found for visual information processing. Both the low-vision rehabilitation and the basic low-vision groups lost visual reading ability from 4 months to 1 year, and the low-vision rehabilitation group also lost overall visual ability. This loss of visual reading ability is independently predicted by near visual acuity losses (logMAR) and lower 4-month reading ability and utility index scores; the loss in overall visual ability in the low-vision rehabilitation group is only predicted by a lower overall visual ability score at 4 months. Although some Veterans Affairs Low-vision Intervention Trial II participants received additional low-vision services from the participating programs or referral to blind rehabilitation centers during the observation period, use of additional services did not predict changes in reading domain scores or in overall visual ability measured from 4 months to 1 year.
Many different outcomes measures are used in low-vision research. The primary outcome measure in this study is patients' self-report of their functional visual ability using the difficulty ratings on the 48-item Veterans Affairs Low-vision Visual Functioning Questionnaire. Self-reported visual ability was selected because patient perception of the difficulty experienced performing activities in their home and community is an important component of low-vision device use. Patients may demonstrate improved performance with low-vision devices during examinations but abandon these devices when it comes to everyday activities.
Ideally, clinical trials that compare outcomes of different low-vision service delivery models would compare both self-reported visual function and reading performance measured in the clinic. Minnesota Low-Vision Reading Test (MNREAD) measures including reading acuity, maximum reading speed, and critical print size were administered in the Veterans Affairs Low-vision Intervention Trial II at baseline and at the end of treatment. However, the patients did not have a low-vision follow-up visit at the end of the trial (4 months from baseline) or at the end of the observation period (8 months after the trial ended), and it was not possible to measure reading speed. This is a limitation of the observational study.
Although this study addresses rehabilitation goals shared by most patients with visual impairment, it is difficult to compare low-vision outcome studies. Differences in measurement resolution and scoring algorithms of various self-report instruments; differences in visual impairment severity, patient traits, or treatment protocols; and differences in the acquisition of low-vision devices and/or the types of low-vision devices dispensed influenced the outcomes of rehabilitation that have been reported in the literature.28
The Veterans Affairs Low-vision Intervention Trial II was conducted in Veterans Affairs medical centers. The U.S. veterans military population is mostly male, whereas the low-vision population served by the private sector is mostly female. Veterans Affairs health insurance provides funding for low-vision devices and rehabilitation for all veterans participating in Veterans Affairs rehabilitation programs at no cost to the patient.29 These Veterans Affairs policies differ from Medicare, where therapy is provided by occupational therapists based upon medical necessity and International Statistical Classification of Diseases, 10th Revision, Clinical Modification codes and low-vision devices are not routinely covered,30 although they may be covered through some Medicare Advantage programs. Patients without other supplemental insurance coverage may also have copayments. Because of these differences, the results of this study cannot be compared with studies conducted in other service delivery settings in the United States or to health care systems in other countries.
Although both treatment groups experienced significant improvement in all measures of visual ability from baseline to 1 year, there was a significant decline in visual reading ability in both treatment groups and overall visual ability in the low-vision rehabilitation group during the observation period. These losses reduced the benefit of the initial services provided. We have no way of knowing, without further follow-up, whether the improvement at 1 year will be maintained or eroded further. Because patients lost visual reading ability and visual acuity decreased from baseline to 1 year in both treatment groups, yearly low-vision follow-up is recommended to assess functional changes associated with disease progression and to modify the patient's treatment plan, if indicated.
1. Crews JE, Campbell VA. Health Conditions, Activity Limitations and Participation Restrictions among Older People Reporting Visual Impairment. J Vis Impair Blind 2001;95:453–67.
2. Haymes SA, Johnson AW, Heyes AD. Relationship between Visual Impairment and Ability to Perform Activities of Daily Living. Ophthalmic Physiol Opt 2002;22:79–81.
3. Lamoureux EL, Hassell JB, Keeffe JE. The Determinants of Participation in Activities of Daily Living in People with Impaired Vision. Am J Ophthalmol 2004;137:265–70.
4. Burmedi D, Becker S, Heyl V, et al. Emotional and Social Consequences of Age-related Low Vision. Visual Impair Res 2002;4:47–71.
5. Rovner BW, Casten RJ, Tasman WS. Effect of Depression on Visual Function in Age-related Macular Degeneration. Arch Ophthalmol 2002;120:1041–4.
6. Brody BL, Gamst AC, Williams RA, et al. Depression, Visual Acuity, Comorbidity, and Disability Associated with Age-related Macular Degeneration. Ophthalmology 2001;108:1893–900.
7. Rovner BW, Ganguli M. Depression and Disability Associated with Impaired Vision: The MOVES Project. J Am Geriatr Soc 1998;46:617–9.
8. Stelmack JA, Stelmack TR, Massof RW. Measuring Low-vision Rehabilitation Outcomes with the NEI VFQ-25. Invest Ophthalmol Vis Sci 2002;43:2859–68.
9. Pearce E, Crossland MD, Rubin GS. The Efficacy of Low Vision Training in a Hospital-based Low Vision Clinic. Br J Ophthalmol 2011;95:105–8.
10. Reeves BC, Harper RA, Russel WB. Enhanced Low Vision Rehabilitation for People with Age-related Macular Degeneration: A Randomized Controlled Trial. Br J Ophthalmol 2004;88:1443–9.
11. de Boer MR, Twisk J, Moll AC, et al. Outcomes of Low-vision Services Using Optometric and Multidisciplinary Approaches: A Non-randomized Comparison. Ophthal Physiol Opt 2006;26:535–44.
12. La Grow SJ. The Effectiveness of Comprehensive Low Vision Services for Older Persons with Visual Impairments in New Zealand. J Vis Impair Blind 2004;98:679–92.
13. Stelmack JA, Tang C, Wei Y, et al. Outcomes of the Veterans Affairs Low Vision Intervention Trial II (LOVIT II). JAMA Ophthalmol 2017;135:96–104.
14. Stelmack JA, Tang XC, Reda DJ, et al. VA LOVIT II: A Protocol to Compare Low-vision Rehabilitation and Basic Low Vision. Ophthalmic Physiol Optics 2012;32:461–71.
15. Stroupe KT, Stelmack JA, Tang XC, et al. Economic Evaluation of Low-vision Rehabilitation for Veterans with Macular Diseases in the US Department of Veterans Affairs. JAMA Ophthalmol 2018;136:524–31.
16. Stelmack JA, Massof RW. Using the VA LV VFQ-48 and LV VFQ-20 in Low-vision Rehabilitation. Optom Vis Sci 2007;84:705–17.
17. Stelmack JA, Szlyk JP, Stelmack TR, et al. Measuring Outcomes of Low-vision Rehabilitation with the Veterans Affairs Low Vision Visual Functioning Questionnaire (VA LV VFQ-48). Invest Ophthalmol Vis Sci 2006;47:3253–61.
18. Stelmack J, Szlyk J, Stelmack T, et al. Use of Rasch Person Item Map in Exploratory Data Analysis: A Clinical Perspective. J Rehabil Res Dev 2004;41:233–42.
19. Szlyk JP, Stelmack J, Massof RW, et al. Performance of the Veterans Affairs Low-vision Visual Functioning Questionnaire. J Vis Impair Blind 2004;98:261–75.
20. Stelmack JA, Szlyk JP, Stelmack TR, et al. Psychometric Properties of the Veterans Affairs Low-vision Visual Functioning Questionnaire. Invest Ophthalmol Vis Sci 2004;45:3919–28.
21. Ware JE Jr., Sherbourne CD. The MOS 36-item Short-Form Health Survey (SF-36). Med Care 1992;30:473–83.
22. Shaw JW, Johnson JA, Coon SJ. US Valuation of the EQ-5D Health States: Development and Testing of the D1 Valuation Model. Med Care 2005;43:203–20.
23. Massof RW, Fletcher DC. Evaluation of the NEI Visual Functioning Questionnaire as an Interval Measure of Visual Ability in Low Vision. Vision Res 2001;41:397–13.
24. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence A. Erlbaum; 1988.
25. Ware JE, Snow K, Kosinski M, et al. SF-36 Health Survey Manual and Interpretation Guide. Boston, MA: Nimrod Press; 1993.
26. Szende A, Oppe M, Devlin N. EQ-5D Value Sets: Inventory, Comparative Review and User Guide. London, UK: Springer; 2007.
27. SAS Institute. SAS Procedures Guide Version 8. Gary, NC: SAS Institute; 1999.
28. Binns AM, Bunce C, Dickinson C, et al. How Effective Is Low Vision Service Provision? A Systematic Review. Surv Ophthalmol 2012;57:34–65.
29. Department of Veterans Affairs, Veterans Health Administration. Blind Rehabilitation Service. In: Coordinated Services for Blinded Veterans 1B vol. 11–59. Washington, DC: Department of Veterans Affairs; 1996.