Disability weights for vision disorders in Global Burden of Disease study

Abstract

As members of the Vision Loss Expert Group, we recognise and applaud the time and effort that went into the 2010 updated Global Burden of Disease (GBD) study. Overall, the methods are excellent and the amount and scope of work breathtaking. The outcome was worth the wait. However, this work also included a systematic reassessment of the disability weights for an extended range of disorders (Dec 15, p 2129).1Salomon J Vos T Hogan D et al.Common values in assessing health outcomes from disease and injury: disability weights measurement study for the Global Burden of Disease Study 2010.Lancet. 2012; 380: 2129-2143Summary Full Text Full Text PDF PubMed Scopus (911) Google Scholar We are concerned about the apparently anomalous disability weights given to some disorders such as vision loss and blindness. Although the approach seems methodologically sound, the results simply do not pass the common-sense test. They seem to represent a deductive false-cause fallacy. The cause of this systematic error seems to be in the formulation of the lay questions used in the survey. Although the vision group did see the lay definitions before the survey questions were finalised, we were unable to compare their content or format with the descriptions prepared for other disorders. It seems that some aspects of the lay descriptions might have been misunderstood by respondents, or inaccurately described the relative consequences of each disorder. For example, severe vision loss (visual acuity <6/60, or legal blindness in most countries) only “causes difficulty in daily activities and some emotional impact”. Other disorders have “great”, “severe”, or “extreme” included in their definition. Vision and hearing loss are notable outliers in the new analysis. The disability weight for blindness has dropped from 0·60 to 0·195, making it little worse than moderate skin disfigurement with itch or pain (0·187), and way behind mild alcoholism (0·259), moderate rheumatoid arthritis (0·292), or even a pain in the neck (0·221)! We are also concerned that the sample populations used for the survey (internet-based or in urban settings) did not adequately represent the visually impaired population, which is larger in rural areas of developing countries where there is little access to education or the internet (to see or to complete the survey) or for people to benefit from rehabilitative services or devices for blindness or visual impairment. The GBD Vision Loss Expert Group was not made aware of the dramatic reduction in weights until the analyses were essentially completed. The large difference merits investigation and explanation before its widespread adoption. The failure of the disability weight assigned to blindness to pass the common-sense test unfortunately threatens the creditability of the whole undertaking. We declare that we have no conflicts of interest. Common values in assessing health outcomes from disease and injury: disability weights measurement study for the Global Burden of Disease Study 2010This study represents the most extensive empirical effort as yet to measure disability weights. By contrast with the popular hypothesis that disability assessments vary widely across samples with different cultural environments, we have reported strong evidence of highly consistent results. Full-Text PDF Disability weights for vision disorders in Global Burden of Disease study – Authors' replyHugh Taylor and colleagues comment on the new disability weights for vision loss and blindness reported for the Global Burden of Disease (GBD) Study 2010.1 Three relevant comparisons are indicated. Full-Text PDF