To determine the change in vision-related quality-of-life scores after providing eyeglasses to American Indian/Alaska Natives with undercorrected refractive error.
Prospective, comparative (nonrandomized) interventional study.
We compared a group with undercorrected refractive error to a control group who did not need a change in eyeglasses. Undercorrected refractive error was defined as distance visual acuity 20/40 or worse in the better-seeing eye that could be improved by at least 2 lines in Snellen visual acuity. Intervention was the provision of new glasses to the undercorrected refractive error group members, based on results of manifest refraction. The main outcome measures were the differences in the 25-Item National Eye Institute Visual Functioning Questionnaire (NEI VFQ-25) scores from baseline (Time 1) to the time after providing eyeglasses (Time 2).
The NEI VFQ-25 median Composite score at Time 1 was significantly lower in those with undercorrected refractive error when compared to the control group (75 vs 92, P = .001). The median Composite score for the undercorrected refractive error group improved to 96 ( P < .001) at Time 2 when compared to Time 1, while the control group remained stable at 93 ( P = .417). The undercorrected refractive error group showed significantly greater improvement than the control group in 8 of 12 subscale scores and in the overall Composite score (all P values ≤ .05). A multivariate linear regression analysis, which controlled for differences in age, percent self-identified American Indian/Alaskan Native, and best-corrected visual acuity between the undercorrected refractive error and control group, showed eyeglasses to be significantly associated with improvement in NEI VFQ-25 composite score.
Visual impairment from undercorrected refractive error is common in American Indian/Alaskan Natives. Providing eyeglasses results in a large, significant increase in vision-related quality of life.
The National Health Interview Survey shows that visual impairment affects more than 21 million noninstitutionalized Americans over the age of 18. Uncorrected refractive error is a leading cause of visual impairment in adults. Visual impairment is approximately 2 times more prevalent in American Indian/Alaskan Natives when compared to age-matched white or African-American individuals. In American Indian/Alaskan Native populations, refractive error is believed to be common because of a high prevalence of astigmatism, and studies demonstrate a higher need for eyeglasses in American Indian/Alaskan Natives when compared to other ethnic groups.
Little information exists about the changes in vision-related quality of life with the provision of eyeglasses. Such studies have focused primarily on patients who are elderly or have other comorbidities, such as macular degeneration, depression, diabetes, or cataracts. We were interested in the amount of improvement in vision-related quality of life after providing adult American Indian/Alaskan Natives with eyeglasses.
The Indian Health Service was established to provide health care for the members of the more than 569 federally recognized tribes in the United States. However, Indian Health Service funding is limited. Eyeglasses are not typically provided to tribal members via federal funding. While eyeglasses are a relatively safe and inexpensive treatment for visual impairment, knowledge regarding the amount of change in vision-related quality of life is important information, especially when considering the relative value of eyeglasses to other competing treatments for visual impairment, such as cataract and refractive surgery.
We enrolled American Indian/Alaskan Native participants who were 18 years or older from 2 locations in the Northwest region of the United States and 1 location from the Midwest. We recruited participants into 1 of 2 groups: the undercorrected refractive error group (if presenting distance visual acuity was 20/40 or worse in the better-seeing eye and manifest refraction showed an improvement of at least 2 lines), or the control group (if presenting distance vision was either better than 20/40 or could not be improved by at least 2 lines on a visual acuity chart). Similar definitions for undercorrected refractive error have been used previously. We excluded candidates who were unable to complete subjective testing.
We used the 25-Item National Eye Institute Visual Function Questionnaire (NEI VFQ-25) to assess self-reported vision-related quality of life. We calculated a Composite score for vision-related quality of life, according to published methods, at baseline (Time 1) and at follow-up after treatment (Time 2). We modified 1 item of the Near Vision Activities subscale of the NEI VFQ-25, adding the term “bead-working,” to better reflect the cultural activities of the American Indian/Alaskan Native population.
We interviewed participants using the NEI VFQ-25 at Time 1 and repeated the questionnaire at Time 2, approximately 3 months later. We waited 3 months to limit test-retest bias for both groups as well as to allow the undercorrected refractive error group sufficient time to obtain and adjust to their new eyeglasses.
We used a modified Behavioral Risk Factor Surveillance System survey to determine American Indian/Alaskan Native heritage, ocular history, and risk factors for medical and eye diseases. We also added a previously validated single question regarding self-reported depression to a subset of the study cohort (n = 37 at Time 1; n = 70 at Time 2). This question (“How often have you felt downhearted and blue in the last 4 weeks?”) has 6 response options: 1 = “All the time,” 2 = “Most of the time,” 3 = “A good bit of the time,” 4 = “Some of the time,” 5 = “A little of the time,” and 6 = “None of the time.” All participants spoke English, and the Behavioral Risk Factor Surveillance Survey and NEI VFQ-25 questionnaires were administered in this language.
Visual Acuity Testing
We recorded presenting Snellen distance vision using standard Early Treatment of Diabetic Retinopathy Study charts, trans-illuminated with a chart illuminator (Lighthouse Low Vision Products, New York, New York, USA). We converted these measurements into a logMAR (logarithm of the minimal angle of resolution) scale of distance visual acuity, using published methods. The main outcome measure was the difference in NEI VFQ-25 scores from Time 1 to Time 2, in the undercorrected refractive error group compared to the control group.
We provided distance or bifocal spectacles to the undercorrected refractive error group (n = 26); we provided bifocal spectacles to 9 participants, while the remainder received distance correction only. Those that only required reading glasses were not included in this study.
We used nonparametric tests (Mann-Whitney U and Kruskal-Wallis tests) when the data were not normally distributed (subscale and composite NEI VFQ-25 scores). NEI VFQ-25 score change from Time 1 to Time 2 was normally distributed, and we used a t test to evaluate the change between control and undercorrected refractive error groups. We also used a multivariate linear regression model to control for differences in visual acuity, age, gender, and self-reported percent American Indian/Alaskan Native heritage between the undercorrected refractive error and control groups. All statistical analyses were performed using SPSS (v16.0, SPSS, Inc, Chicago, Illinois, USA) and R (Available at: www.R-project.org . Accessed August 8, 2010) statistical softwares. We checked the internal consistency of subscales by Cronbach alpha statistics. We employed mixed-effects models with a random intercept to account for potential autocorrelation within a person within a testing location using the R “nlme” library.
We determined whether our sample sizes were adequate based on observed group means and effect sizes. Prior to the study, we determined a sample size of 84 people per group was needed to detect a 10-point difference in Composite scores, with an alpha of 0.05 ( t test), power of 0.80, and intertemporal correlation between scores of 0.60. We further calculated that a 20-point difference would require 21 persons in each group. The amount of change in the undercorrected refractive error group exceeded 20 points in the majority of the subscales and in the Composite score; therefore, we stopped enrolling participants after recruiting 26 undercorrected refractive error participants.
We enrolled 114 participants: 76 into the control group, 26 into the undercorrected refractive error group, and 12 participants (11%) who did not complete the NEI VFQ-25 questionnaire at Time 2. Therefore, we report data for 102 (89%) participants. We found no statistically significant differences ( P > .05) in age, sex, percent American Indian/Alaskan Native heritage, Time 1 Composite scores, measures of self-reported depression, or visual acuity between those who completed and those who did not complete the questionnaire at Time 2 (data not shown).
Table 1 shows the baseline characteristics of the control and undercorrected refractive error groups. Significant differences between the groups existed in age, percent self-reported American Indian/Alaskan Native heritage, level of visual impairment, and median visual acuity. Control group participants showed an improvement of less than 1 line of acuity (Time 1 mean = 0.09, median 0.00; Time 2 mean = 0.06, median = 0.00) after manifest refraction (0.30 lines), while those in the undercorrected refractive error group improved nearly 6 lines (Time 1 mean = 0.60, median = 0.50; Time 2 mean = 0.03, median = 0.00) after manifest refraction (5.7 lines).