Douzdjian V, Ferrara D, Silvestri G
Department of Surgery, and the Center for Health Care Research, Medical University of South Carolina, Charleston, USA.
Am J Kidney Dis. 1998 May;31(5):794-802. doi: 10.1016/s0272-6386(98)70048-4.
Clinical decision analysis has become an important tool for evaluating specific clinical scenarios and exploring public health policy issues. A decision analysis model that incorporates patient preferences regarding various outcomes, as well as cost, may be particularly informative in patients with type I diabetes and end-stage renal disease (ESRD). Such a model that includes pancreas transplantation as a treatment choice has not been performed and is presented in this study. The decision tree consisted of a choice between four possible treatment strategies: dialysis, kidney-alone transplant from a cadaver (KA-CAD) or living donor (KA-LD), and simultaneous pancreas-kidney (SPK) transplant. The analysis was based on a 5-year model, and the measures of outcome used in the model were cost and cost adjusted for quality of life. The measure of preference for quality of life was obtained using the "Standard Reference Gamble" method in 17 SPK transplant recipients who underwent transplantation between January, 1992 and June, 1996 at our center. The measures for various outcome states (mean +/- 1 SD) were dialysis-free/insulin-free = 1, dialysis-free/insulin-dependent = 0.6 (0.4 to 0.8), dialysis-dependent/insulin-free = 0.5 (0.36 to 0.64), dialysis-dependent/insulin-dependent = 0.4 (0.21 to 0.59), and death = 0. The expected 5-year costs for each of the treatment strategies in the model were dialysis, $216,068; KA-CAD transplant, $214,678; KA-LD transplant, $210,872; and SPK transplant, $241,207. The expected cost per quality-adjusted year for each of the treatment strategies in the model were dialysis, $317,746; KA-CAD transplant, $156,042; KA-LD transplant, $123,923; and SPK transplant, $102,422. SPK transplantation remained the optimal strategy after varying survival probabilities, costs, and utilities over plausible ranges by means of one-way sensitivity analysis. In conclusion, according to the 5-year cost-utility model presented in this study, SPK transplantation is the most cost-effective treatment strategy for a patient with type I diabetes and ESRD. From a policy standpoint, looking at the cost alone of pancreas transplantation is deceiving. In these patients, who may view various outcome states differently, it would be important to take into account cost adjusted for quality of life when evaluating this procedure.
临床决策分析已成为评估特定临床情况和探讨公共卫生政策问题的重要工具。一个纳入患者对各种结果的偏好以及成本的决策分析模型，对于I型糖尿病和终末期肾病（ESRD）患者可能特别有参考价值。本研究提出了一个包含胰腺移植作为治疗选择的此类模型，但此前尚未进行过。决策树由四种可能的治疗策略之间的选择组成：透析、来自尸体供体的单独肾脏移植（KA-CAD）或活体供体的单独肾脏移植（KA-LD）以及胰肾联合移植（SPK）。该分析基于一个5年模型，模型中使用的结果衡量指标是成本以及根据生活质量调整后的成本。生活质量偏好的衡量指标是通过“标准参考赌博”方法，在1992年1月至1996年6月期间于我们中心接受移植的17名SPK移植受者中获得的。各种结果状态的衡量指标（均值±1标准差）分别为：无透析/无胰岛素 = 1，无透析/胰岛素依赖 = 0.6（0.4至0.8），透析依赖/无胰岛素 = 0.5（0.36至0.64），透析依赖/胰岛素依赖 = 0.4（0.21至0.59），死亡 = 0。模型中每种治疗策略的预期5年成本分别为：透析，216,068美元；KA-CAD移植，214,678美元；KA-LD移植，210,872美元；SPK移植，241,207美元。模型中每种治疗策略每质量调整年的预期成本分别为：透析，317,746美元；KA-CAD移植，156,042美元；KA-LD移植，123,923美元；SPK移植，102,422美元。通过单因素敏感性分析在合理范围内改变生存概率、成本和效用后，SPK移植仍然是最优策略。总之，根据本研究提出的5年成本效用模型，SPK移植是I型糖尿病和ESRD患者最具成本效益的治疗策略。从政策角度来看，仅看胰腺移植的成本是有误导性的。在这些可能对各种结果状态看法不同的患者中，在评估该手术时考虑根据生活质量调整后的成本非常重要。
