Dear Editor,

We read with interest the article by Gupta and colleagues describing their findings in an investigation of the utility loss associated with glaucoma. When considered in light of previous investigations this work makes an important contribution to our limited understanding of the influence of culture, socio-economic status, and ethnic background on health state preference. It is particularly gratifying to see this done in glaucoma, a disease where there has been very limited work done towards utility estimate.

There are some aspects of the results of this study that bear clarification. First, we ask that they comment on staging glaucoma using visual acuity rather than visual field loss. Clinical staging of glaucoma by loss of visual field is the method preferred by glaucoma specialists,[1] and both utility loss (as measured by the EQ-5D)[2] and cost of care[3] have been shown to be responsive to this measure. Second, it would be good to report the number of study participants who refused to trade any time (or risk blindness or death due to surgery). In previous work, this proportion has substantial.[4] Finally, we ask that they comment more fully what some may consider an inconsistency in the findings. They found that on average people with glaucoma are willing to accept a 14% risk of death to eliminate their disease, but only a 3% risk of blindness, leading us to assume that among the people in this sample, being blind is worse than being dead. This has not been found in previous studies of the utility associated with blindness.

Just as Dr. Gupta and his colleagues speculate that there might be a cultural or socio-economic basis for the substantially higher utility loss associated with glaucoma he found in his sample (when contrasted with the work of Jampel4), it is possible that these apparently “inconsistent” findings might be the result of differing views of visual disability in our industrialized society (with nearly universal access to—although not necessarily utilization of—health care and rehabilitative services), versus that found in a developing country. If these findings are indeed valid and supported by additional research, it would make an important contribution to our understanding of preference based measures by providing evidence that preferences for health states (and thus variance in reported utility) may vary based upon socio-economic factors. While some have suggested that this is not the case,[5] such a finding would be consistent with recent reports that reported utility varies by race and other factors.[6,7]

References:

1. Hodapp E, Parrish RK, Anderson DR. Clinical Decisions in Glaucoma. St. Louis: Mosby, 1993.

2. Alm A, Kobelt G, Bergstrom A, Chen E, Linden C. Measuring Utility in Glaucoma. 2005 Annual Meeting Association for Research in Vision and Ophthalmology Fort Lauderdale, Florida.

3. Lee PP, Walt J, Doyle JJ, Kotak SV, Evans SJ, Budenz DL, et al. A Multi -center, Retrospective Pilot Study of Resource Utilization and Costs Associated with Severity of Disease in Glaucoma. Archives of Ophthalmology 2005;Accepted for Publication.

4. Jampel HD, Schwartz A, Pollack I, Abrams D, Weiss H, Miller R. Glaucoma Patients' Assessment of Their Visual Function and Quality of Life. Journal of Glaucoma 2002;11(2):154-63.

5. Brown GC, Brown MM, Sharma S, Beauchamp GR, Hollands H. The reproducibility of ophthalmic utility values. Transactions of the American Ophthamological Society 2001;99:199-204.

6. Bravata DM, Nelson LM, Garber AM, Goldstein MK. Invariance and Inconsistency in Utility Ratings. Medical Decision Making 2005;25(2):158- 67.

7. Wittenberg E, Divi N, Halpern E, Araki SS, Prosser L, Weeks JC. The Effect of Age, Race and Gender on Utility Values for Hypothetical Health States. 2004 Annual Meeting Society for Medical Decision Making Atlanta Georgia.