Frequency distribution of the National Eye Institute Visual Function Questionnaire-25.
Independent effect of visual acuity on the total 25-item National Eye Institute Visual Function Questionnaire score while controlling for age, retinopathy level, loss of tactile sensation, pack-years smoked, SF-36 (36-Item Short-Form Health Survey) physical component summary, and SF-36 mental component summary.
Klein R, Moss SE, Klein BEK, Gutierrez P, Mangione CM. The NEI-VFQ-25 in People With Long-term Type 1 Diabetes MellitusThe Wisconsin Epidemiologic Study of Diabetic Retinopathy. Arch Ophthalmol. 2001;119(5):733-740. doi:10.1001/archopht.119.5.733
To examine the association of the 25-item National Eye Institute Visual Function Questionnaire (NEI-VFQ-25) overall and specific scale scores with visual acuity, diabetic retinopathy, and other characteristics, in a cohort of persons with type 1 diabetes.
Population-based cohort study.
An 11-county area in southern Wisconsin.
Six hundred two persons with diabetes whose conditions were diagnosed when they were younger than 30 years and who were currently taking insulin participated in baseline, 4-year, 10-year, and 14-year follow-up examinations.
Main Outcome Measures
An interview that consisted of the 25-item NEI-VFQ was completed. Visual acuity was measured by the Early Treatment of Diabetic Retinopathy Study (ETDRS) protocol and the presence and severity of retinopathy and macular edema were detected by masked grading of stereoscopic color fundus photographs using the modified Airlie House classification and the ETDRS retinopathy severity scheme.
Univariate analyses revealed that the total NEI-VFQ-25 score was lower in persons who were older, had a longer duration of diabetes, higher glycosylated hemoglobin, were in renal failure, had a history of cardiovascular disease, hypertension, or amputation of a lower limb, had poorer visual acuity, more severe diabetic retinopathy, macular edema, glaucoma, cataract, abnormalities in tactile sensation or temperature sensitivity, smoked more total pack-years, led a more sedentary lifestyle, and had poor peak expiratory flow. In multivariate analyses, while controlling for the physical and mental component scores from the Medical Outcomes Survey 36-Item Short-Form Health Survey as measures of comorbidity, lower total NEI-VFQ-25 scores were independently associated with poorer visual acuity, more severe retinopathy, older age, history of loss of tactile sensation, and more total pack-years of cigarettes smoked.
In this cross-sectional study, the 25-item NEI-VFQ seems to be strongly associated with vision, independent of severity of retinopathy and other complications associated with type 1 diabetes. It may be a useful measure of health-related quality of life as it relates to vision in epidemiological studies and clinical trials in persons with diabetes.
THE NATIONAL Eye Institute Visual Function Questionnaire (NEI-VFQ) was designed to measure vision-related functioning and the influence of vision problems on health-related quality of life in persons with several eye conditions, such as glaucoma, diabetic retinopathy, age-related cataract, age-related macular degeneration, cytomegalovirus retinitis, and low vision.1 The questions constituting the instrument were developed from focus groups. The reliability and validity of the instrument when used among people with common eye conditions was tested and confirmed.2 However, to date there have been no population-based cohorts of people with diabetes in which the associations of retinopathy and visual acuity with the NEI-VFQ scales have been examined. The purpose of this report is to examine the association of the NEI-VFQ-25 composite scores and specific scale scores with visual acuity, diabetic retinopathy, and other characteristics, in the cohort of persons with type 1 diabetes participating in the 14-year follow-up of the Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR).
The population, which has been described in previous reports,3- 7 consisted of a sample selected from 10 135 diabetic patients who received primary care in an 11-county area in southern Wisconsin from 1979 to 1980. This sample was composed of "younger-onset" persons and "older-onset" persons. These analyses were limited to the group of younger-onset persons, all of whom were taking insulin and the conditions of all had been diagnosed when they were younger than 30 years (n = 1210). There were 996 persons in this group who participated in the baseline examination (1980-1982),3891 in the 4-year follow-up,5 765 in the 10-year follow-up,6 and 654 in the 14-year follow-up(including 20 who had missed earlier follow-up examinations).7 The reasons for nonparticipation and comparisons between participants and nonparticipants at baseline and the 4-, 10-, and 14-year follow-ups have been presented elsewhere.3,5- 7 Mean ± SD and median lengths of time between the baseline and 14-year follow-up examinations were 14.4 ± 0.5 years and 14.3 years, respectively.
The baseline and follow-up examinations were performed in a mobile examination van in or near the cities where the participants resided. All examinations followed a similar protocol that was approved by the institutional human subjects committee of the University of Wisconsin–Madison. The pertinent parts of the ocular and physical examinations included measuring refractive error and best-corrected visual acuity for distance using a modified Early Treatment Diabetic Retinopathy Study (ETDRS) protocol in which the charts were reduced in size for a 2-m distance8,9;measuring weight, height, blood pressure,10 and intraocular pressure; dilating the pupils; taking stereoscopic color fundus photographs of 7 standard fields8,9;performing a semiquantitative determination of glucose, ketone, and protein levels in the urine using Labstix (Ames, Elkhart, Ind); and determining blood glucose and glycosylated hemoglobin A1 levels from a capillary(baseline) or venous (follow-up) blood sample.11,12
A structured interview was conducted by the examiners including questions about specific medications for control of hyperglycemia and blood pressure, the use of diuretic agents, the number of aspirin tablets taken during the 30 days before the baseline examination, and smoking. If there was any question about medication use, it was verified by a physician's report. Other questions included whether patients had experienced loss of sensation in their hands or feet and decreased ability to feel the hotness or coldness of things they touched since having diabetes.
After the interview, a measure of peak expiratory flow was obtained for 3 trials. The measurement used in the analyses reflects the maximum peak expiratory flow of the 3 trials.
From July 1997 to November 1998 (mean ± SD, 27 ± 6 months; median, 28 months after the last eye examination), the NEI-VFQ-25 was administered by telephone to 602 WESDR younger-onset subjects who had participated in the 1995-1996 examination.
For each eye, the best-corrected visual acuity was recorded as the number of letters read correctly from 0 (≤20/250) to 70 (20/10).13 For eyes with visual acuity worse than 20/250, 1 of 6 levels of visual acuity was recorded: 20/320, 20/400, 20/800, hand motions, light perception, and no light perception. This represents a continuation of the scale wherein a loss of 15 letters is equivalent to a doubling of the visual angle or a significant reduction in vision. These levels were assigned values on the visual acuity scale of −5, −10, −25, −40, −55, and −70, respectively. In this study, visual acuity was defined as the visual acuity in the better eye, and blindness was defined as a visual acuity of 20/200 or worse in the better eye.
The grading protocol has been described in detail elsewhere and is a modification of the ETDRS adaptation of the modified Airlie House classification of diabetic retinopathy.6,14- 17 For each eye, the maximum grade in any of the 7 standard photographic fields was determined for each of the lesions used in defining the "retinopathy levels." The retinopathy level for a participant was derived by concatenating the levels for the 2 eyes, giving the eye with the higher level greater weight. This scheme provided a 15-step scale (10/10, 21/<21, 21/21, 31/<31, 31/31, 37/<37, 37/37, 43/<43, 43/43, 47/<47, 47/47, 53/<53, 53/53, 60+/<60+, and 60+/60+) when all levels of proliferative retinopathy were grouped as one level. For purposes of classification, if the retinopathy severity could not be graded in an eye, it was considered to have a score equivalent to that in the other eye. For purposes of tabulation, the number of severity levels for individuals was reduced to 4: none if the retinopathy level was 10/10; minimal nonproliferative diabetic retinopathy (levels 21/<21 to 37/37); moderate to severe nonproliferative diabetic retinopathy (levels 43/<43 to 53/53); and proliferative diabetic retinopathy (levels 60+/<60+ to 60+/60+). Panretinal photocoagulation was documented by the grading of fundus photographs.
Macular edema was defined as thickening of the retina with or without partial loss of transparency within 1 disc diameter from the center of the macula6 or the presence of focal photocoagulation scars in the macular area associated with a history of development of macular edema as documented by stereoscopic fundus photographs. Clinically significant macular edema was based on the detailed gradings and was defined as the presence of any 1 of the following: retinal thickening at or within 500 µm of the center of the macula; and/or hard exudates at or within 500 µm of the center of the macula if associated with thickening of the adjacent retina; and/or a zone or zones of retinal thickening 1 disc area in size, at least part of which was within 1 disc diameter of the center.18 Whenever we found new signs of photocoagulation scars in the macular area in the absence of macular edema and we had not previously documented macular edema by grading fundus photographs taken at an earlier examination, we obtained fundus photographs from the participant's ophthalmologist. In the absence of fundus photographs, we obtained medical records documenting that macular edema due to diabetes had been present prior to the focal (or grid) photocoagulation. For situations in which participants gave a history of laser photocoagulation but there were no signs of treatment burns, we requested information from the treating ophthalmologist to verify that such treatment had been done and to ascertain whether macular edema had been present prior to focal laser treatment. For purposes of analysis, macular edema was considered to be present in an eye if graded as present in fundus photographs or if there was a documented prior history of macular edema with focal photocoagulation treatment. If macular edema could not be graded in an eye, the individual was assigned the score of the other eye.
Age was defined as the age at the time of the 14-year follow-up examination in 1995-1996. Age at diagnosis of diabetes was defined as the age at the time the diagnosis was first recorded by a physician on the patient's medical record. The duration of diabetes was that period between the age at diagnosis and the age at the 14-year follow-up.
The means of both systolic and diastolic blood pressures were the averages of the last 2 of 3 measurements according to the protocol of the Hypertension Detection and Follow-up Program.8,9 Hypertension was defined as a mean systolic blood pressure greater than or equal to 160 mm Hg and/or a mean diastolic blood pressure greater than or equal to 95 mm Hg or a history of antihypertensive medication use at the time of examination in individuals aged 25 years or older or a mean systolic blood pressure greater than or equal to 140 mm Hg and/or a mean diastolic blood pressure greater than or equal to 90 mm Hg and/or a history of antihypertensive medication use at the time of examination in younger persons.
Cigarette smoking status was determined as follows: a person was classified as having never smoked if he or she had smoked fewer than 100 cigarettes in his or her lifetime, as being an ex-smoker if he or she smoked more than this number of cigarettes in his or her lifetime but had stopped smoking before the examination, or as currently smoking if he or she had not stopped. For purposes of analysis, 2 dichotomous variables were defined: 1 to compare persons who had formerly smoked with those who had never smoked and 1 to compare persons who currently smoked with those who had never smoked. Pack-years smoked was calculated as the number of cigarettes smoked per day divided by 20, multiplied by the number of years of smoking. Proteinuria was defined as a urine protein concentration of 0.30 g/L or greater as measured by Labstix. End-stage renal disease was defined as having had a kidney transplant or receiving renal dialysis. History of lower extremity amputation was ascertained by questionnaire or direct observation. The 36-Item Short-Form Health Survey (SF-36) was administered during the study.19,20 This was reduced to the physical component summary and mental component summary.20 These summary measures reduced the 8 subscales of the SF-36 to the physical and mental dimensions of health status. To adjust for the independent effects of overall physical and mental health on the NEI-VFQ scores, each multivariate model included the physical and the mental component summary scores from the SF-36. In addition, all subjects were asked whether their health was excellent, very good, good, fair, or poor (EVGFP scale). In rating their physical activity, subjects were asked how many times per week they engaged in a regular activity long enough to work up a sweat. A sedentary lifestyle was defined as engaging in fewer than 3 such activities per week.
The composite NEI-VFQ scale used in the analyses was a modified version in which the general health question was the only item excluded. These analyses gave results similar to those for the composite NEI-VFQ scale using the full 25 items. For univariate analyses of the 25-item composite scale, mean values of the scale were calculated for categories of the independent variables. Statistical significance of the associations was tested by the significance of the Spearman correlation. Because the scale was skewed (Figure 1), a logarithmic transform was performed by the formula TVFQ25 = ln (101-VFQ25), in which TVFQ25 and VFQ25 are the transformed and untransformed values of the 25-item composite scale, respectively. The transformed scale was used as the dependent variable in multiple linear regression.21 For ease of interpretation, the results are reported in the original untransformed scale. Significant variables in the final multiple linear regression model were examined for nonlinear effects by the addition of quadratic terms. Interactions of age with other factors were also tested. Univariate analyses of NEI-VFQ-25 subscales were performed similarly. Because these scales are more discrete in nature and do not lend themselves to normalizing transformations, multivariate analyses were performed by ordinal logistic regression.22 The general vision, near vision, far vision, and driving scales were divided into 3 groups, and the mental health scale was divided into 4 groups.
The differences between responders and nonresponders (n = 52) are presented in Table 1. Those who did not respond were more likely to be older, have longer duration of diabetes, have poorer visual acuity, and have more complications associated with diabetes.
The distributions of the NEI-VFQ-25 and its subscales are presented in Table 2. Few participants had scale scores of 0, while a sizable proportion had scale scores of 100. Scores for general health, general vision, near vision, far vision, and driving were lowest, and those for peripheral vision, color vision, and visual pain were highest.
Univariate analyses revealed that the total score was lower in persons who were older, had a longer duration of diabetes, higher glycosylated hemoglobin, were in renal failure, had a history of cardiovascular disease, hypertension, or amputation of a lower limb, had poorer visual acuity, more severe diabetic retinopathy, macular edema, glaucoma, cataract, abnormalities in tactile sensation or temperature sensitivity, smoked more pack-years, led a more sedentary lifestyle, and had poorer peak expiratory flow (Table 3 and Table 4). There was no difference between men and women. Similar associations were found for the general, near, and far vision subscales and the mental health score (data not shown) and driving subscale scores. The associations with visual pain, social functioning, role functioning, dependency, peripheral vision, and color vision were not examined because deficits on these subscales were uncommon.
While controlling for retinopathy severity level, age, pack-years smoked, SF-36 physical component summary, SF-36 mental component summary, and the loss of tactile sensation, the association of visual acuity with the composite NEI-VFQ-25 score is shown in Figure 2. Persons with visual acuity of 20/10 had a composite NEI-VFQ-25 score of 96.1 while those with visual acuity of 20/800 had a total score of 62.8. Because of the transformation of the NEI-VFQ-25, the effect of the independent variables was greater at higher values with the exception of visual acuity, the effect of which was greater with poorer vision. However, visual acuity and the SF-36 physical component summary each had a significant quadratic term. For visual acuity, the effect of this term was to decrease the effect of lower vision; for the SF-36 physical component summary, the quadratic term increased the effect at higher values. Visual acuity explained 27.1% of the variance (R2). Additional statistically significant associations were found for retinopathy severity level (3.8%), age (0.8%), pack-years smoked(1.6%), SF-36 physical component summary (12.8%), SF-36 mental component summary(3.7%), and loss of tactile sensation (1.4%). These factors and visual acuity explain 51% of the variance (Table 5).
Multivariate analyses of general, near, and far vision subscale scores and the mental health and driving subscale scores again showed visual acuity to be strongly associated with each of the subscale scores (data not shown). While controlling for age, visual acuity, and use of glaucoma drops, persons with panretinal photocoagulation in 1 or both eyes were more likely to report a peripheral vision deficit than persons without panretinal photocoagulation(odds ratio, 2.35; 95% confidence interval, 1.31-4.21).
The NEI-VFQ-25 has been shown to be a reliable and valid questionnaire for patients with 5 chronic eye conditions or low vision from any cause.23 The data presented herein extend these findings in a population-based cohort of persons with long-term type 1 diabetes in whom visual acuity and diabetic retinopathy were measured using standardized protocols. The NEI-VFQ-25 composite score and subscale scores associated most closely with central vision (general, near, and far vision subscale scores and the driving subscale score) are strongly and independently correlated with visual acuity. These findings are consistent with those reported in 123 patients with moderate to severe diabetic retinopathy who were participants in the NEI-VFQ Field Test designed to assess the reliability and validity of the 51-item Field Test Version of the NEI-VFQ.2
In the WESDR, after controlling for visual acuity, there was a minimal effect of more severe retinopathy on the NEI-VFQ-25 composite score (a reduction from a total score of 93.7 in those without retinopathy to 88.8 in those with proliferative retinopathy in both eyes). In our study, the effect of age-related cataract on the NEI-VFQ-25 was also small once visual acuity was controlled. These findings provide further evidence that the NEI-VFQ-25, independent of the presence and severity of retinopathy or macular edema or cataract, is a valid measure of vision-targeted health-related quality of life in epidemiological studies and clinical trials of persons with diabetes.
Although this is the first report of the influence of diabetic eye disease on multidimensional vision-targeted health-related quality of life as captured by the NEI-VFQ-25, others have reported similar correlations between self-reported measures of visual functioning and measured visual acuity among persons with retinal disease in general,24 age-related macular degeneration,25 glaucoma,26 and optic neuritis.27 Additionally, in this cohort with diabetes, we have previously reported on the negative influence of hyperglycemia and diabetic complications on general health-related quality of life as measured by the SF-36.28,29 In a practice-based sample, Hanninen et al30 also found that patients with type 2 diabetes had poorer general health-related quality of life than age- and sex-matched controls and that these decrements were primarily attributable to diabetes-related complications, such as visual impairment and coronary artery disease. A recent study that measured health utilities to estimate the influence of visual loss attributable to diabetic retinopathy identified substantial decrements in quality of life attributable to diabetic eye disease.31
Care must be exercised when interpreting the results of this study. First, the NEI-VFQ-25 was administered from July 1997 through November 1998, a period of 1 to 3 years after completion of the study examinations. Changes in visual acuity (both worsening and improvement) could weaken the association reported. In addition, the lag period between the last eye examination and the NEI-VFQ-25 interview might also explain some of the ceiling effect noted on some of the subscales. This may have resulted because many of the nonresponders, who had more severe disease, would be the ones most likely to have lower VFQ scores. Second, the results are limited to persons with long-term (≥16 years) type 1 diabetes. Although the content of the NEI-VFQ-25 was driven by information provided during focus groups with persons with chronic eye diseases, it is possible that the participants with diabetes had decrements in domains of vision-targeted health-related quality of life that are not included in the NEI-VFQ-25.
In summary, strong associations are reported between best-corrected visual acuity and the NEI-VFQ-25 composite and subscale scores associated with central vision. Longitudinal data and information on people with type 2 diabetes are needed for further evaluation of the generalizability of these findings.
Accepted for publication September 25, 2000.
This research was supported by grants EYO3083 (Drs R. Klein and B. E. K. Klein) and EY12198 (Drs R. Klein and B. E. K. Klein) from the National Institutes of Health, Bethesda, Md; and in part by the Senior Scientific Investigator Award from Research to Prevent Blindness Inc, New York, NY (Dr R. Klein).
We thank the 452 Wisconsin physicians and their staffs who participated in and supported this study. We also thank the local hospitals that provided supportive services for the mobile van, and the State of Wisconsin Division of Health, Madison, for donating the van.
Corresponding author: Ronald Klein, MD, MPH, Department of Ophthalmology and Visual Sciences, University of Wisconsin–Madison, 610 N Walnut St, 460 WARF, Madison, WI 53705-2397.