Purpose: We developed a multiattribute instrument to assess utility loss from glaucoma progression based upon the National Eye Institute Visual Function Questionnaire (NEI-VFQ) to enable assessment of utility loss due to modest yet clinically significant changes in vision.
Method: We analyzed NEI-VFQ data from 1,699 participants with ocular hypertension or glaucoma to identify items responsive to changes in disease status. We incoporated these items (attributes) into a conjoint analysis interview administered to 48 participants with glaucoma to assess item importance. From these results we constructed a web-based interview using standard gamble and visual analog scales to elicity utility from a community sample. These results were analyzed using both multiattribute and regression methods. We present preliminary regression results from 135 participants.
Results: The five attributes most important to people with glaucoma were: ability to read, driving at night, ability to leave home, needing help with activities and ability to accomplish tasks. We added seeing objects to the side, since loss of peripheral vision is associated with glaucoma. The Table provides demographic characteristics and utility loss associated with full loss of the attribute. The health state reflecting loss of all attributes (“Pits”) had a utility loss of 0.33 (sd=0.34). In multivariable regression analysis with Pits as the dependent variable and other attributes as the independent, the attributes with the strongest contribution to disutility were the inability to read, inability to leave home and loss of peripheral vision.
Conclusion: We identified the items from the NEI-VFQ most associated with loss of quality of life on the policy scale. Our current report is based upon full loss of the attribute. Further analyses will provide estimates less degree of loss. This will allow use of previously collected NEI-VFQ data to estimate the loss of utility on the policy scale across the spectrum of glaucoma.
| Mean or % | Standard Deviation |
| Age | 43.2 | 15.8 |
| Race |
|
|
| White | 80.4% |
|
| African-American | 9.0% |
|
| Other | 10.6% |
|
| Education Level |
|
|
| Less than college | 6.1% |
|
| College graduate | 57.1% |
|
| Post graduate | 36.8% |
|
| Health Status |
|
|
| Excellent | 31.6% |
|
| Good | 40.6% |
|
| Fair | 24.8% |
|
| Poor | 3.0% |
|
| Utility loss for total loss of attribute |
|
|
| Reading | 0.189 | 0.249 |
| Driving at night | 0.190 | 0.248 |
| Seeing to the side | 0.164 | 0.235 |
| Leaving home | 0.266 | 0.286 |
| Accomplishing less | 0.218 | 0.270 |
| Needing help | 0.306 | 0.303 |
See more of: The 32nd Annual Meeting of the Society for Medical Decision Making