Lee Bryan S, Kymes Steven M, Nease Robert F, Sumner Walton, Siegfried Carla J, Gordon Mae O
Department of Ophthalmology and Visual Sciences, Washington University School of Medicine, St. Louis, Missouri 63110, USA.
Ophthalmology. 2008 May;115(5):898-903.e4. doi: 10.1016/j.ophtha.2007.06.008. Epub 2007 Sep 12.
To elicit utilities on a perfect health and perfect vision scale for 5 common eye diseases.
Cross-sectional observational preference study.
We included 434 patients: 58 with diabetic retinopathy, 99 with glaucoma, 44 with age-related macular degeneration (AMD), 124 with cataract; 109 with refractive error.
Standard gamble utilities were estimated using a computer-based preference assessment interview platform.
Standard gamble utilities, a quality-of-life measure that examines the willingness to accept a risk of death or unilateral blindness in return for perfect health or perfect vision.
Using the standard policy scale, where health equivalent to death is 0 and perfect health is 1, participants with asymptomatic diabetic retinopathy had a utility of 0.93. By comparison, symptomatic diabetics had a further utility loss of 0.14. Asymptomatic glaucoma participants had a utility of 0.92 with a decrease of 0.03 for early field loss and a further decrease of 0.03 with central field loss. Participants with AMD who had > or =20/100 better-eye visual acuity reported a utility of 0.89, whereas those with more severe AMD reported 0.76. However, neither clinical cataract opacity score nor refractive error correlated with utility. Adjustment for age and comorbidity did not alter these relationships. For the same participants, utilities measured with different anchor points-monocular blindness as 0 and perfect vision as 1-were lower, especially among participants with increased disease severity. The difference between utility assessed on this perfect vision-blindness scale and the perfect health-death scale ranged from 0.04 for those with severe refractive error to 0.19 for symptomatic diabetics and 0.37 for those with severe AMD.
This paper elicits utilities with different anchor points from a previously unreported sample of 434 patients. Lower utility scores normally imply greater benefit with successful treatment or prevention of disease, but switching from the conventional policy scale to the perfect vision scale also consistently results in lower scores. Because most previous ophthalmic studies have used perfect vision as the upper anchor, the resulting utilities may not have been accurate.
得出5种常见眼病在完美健康和完美视力量表上的效用值。
横断面观察性偏好研究。
我们纳入了434例患者，其中58例患有糖尿病性视网膜病变，99例患有青光眼，44例患有年龄相关性黄斑变性（AMD），124例患有白内障，109例患有屈光不正。
使用基于计算机的偏好评估访谈平台估计标准博弈效用值。
标准博弈效用值，这是一种生活质量测量指标，用于考察为换取完美健康或完美视力而接受死亡或单眼失明风险的意愿。
使用标准量表，即健康等同于死亡为0，完美健康为1，无症状糖尿病性视网膜病变患者的效用值为0.93。相比之下，有症状的糖尿病患者效用值进一步损失0.14。无症状青光眼患者的效用值为0.92，早期视野缺损时降低0.03，中央视野缺损时再降低0.03。最佳矫正视力大于或等于20/100的AMD患者效用值为0.89，而病情更严重的AMD患者效用值为0.76。然而，临床白内障混浊程度评分和屈光不正均与效用值无关。对年龄和合并症进行校正并未改变这些关系。对于相同的参与者，以单眼失明为0、完美视力为1的不同锚定点测量的效用值较低，尤其是在疾病严重程度增加的参与者中。在这种完美视力-失明量表上评估的效用值与完美健康-死亡量表之间的差异，从重度屈光不正患者的0.04到有症状糖尿病患者的0.19以及重度AMD患者的0.37不等。
本文从434例此前未报道的患者样本中得出了具有不同锚定点的效用值。较低的效用值通常意味着成功治疗或预防疾病会带来更大益处，但从传统量表转换为完美视力量表也会始终导致得分降低。由于此前大多数眼科研究都将完美视力作为上限锚定点，因此得出的效用值可能并不准确。
