Abstract：
Lung cancer screening with low-dose computed tomography (LDCT) in high-risk populations has been shown in randomised controlled trials to lead to early diagnosis and reduced lung cancer mortality. However, investment into screening will largely depend on the outcomes of cost-effectiveness analyses that demonstrate acceptable costs for every quality-adjusted life year (QALY) gained. The methods used to apply utility values to measure QALYs can significantly impact the outcomes of cost-effectiveness analyses and if applied inaccurately can lead to unreliable estimates. We reviewed the use of utility values in 26 cost-effectiveness analyses of lung screening with LDCT conducted between 2005 and 2021, and found considerable variation in methods. Specifically, authors made different assumptions made relating to (i) baseline quality-of-life among screening participants, (ii) potential harms from screening, (iii) utilities and disutilities applied to lung cancer health states, and (iv) quality-of-life for lung cancer survivors. We discuss how each of these assumptions can influence incremental cost-effectiveness ratios. Key recommendations for future evaluations are (i) that modelling studies should justify the choice of baseline utilities, especially if patients are assumed to recover fully after curative treatment; (ii) the impact of false positive scans on quality-of-life should be modelled, at least in sensitivity analyses; (iii) modellers should justify assumptions relating to post-operative recovery, preferably based on knowledge of local practices; (iv) utilities applied to a lung cancer diagnosis should be appropriately sourced and calculated; and (v) adjustment for age-related declines in quality-of-life should be considered, especially for models that examine lifetime horizons.
摘要：
在高风险人群中使用低剂量计算机断层扫描（LDCT）进行肺癌筛查已在随机对照试验中显示，可导致早期诊断并降低肺癌死亡率。然而,对筛查的投资将在很大程度上取决于成本效益分析的结果，该分析证明了每个质量调整生命年（QALY）的可接受成本。用于应用效用值衡量质量调整寿命周期的方法可能会显著影响成本效益分析的结果，如果应用不准确，可能导致不可靠的估计。我们回顾了2005年至2021年间进行的26项LDCT肺筛查成本效益分析中效用值的使用，发现方法存在很大差异。具体来说,作者对以下方面做出了不同的假设：(i)筛查参与者的基线生活质量，(ii)筛查的潜在危害，(iii)适用于肺癌健康状态的公用事业和废品，和（iv）肺癌幸存者的生活质量。我们讨论了这些假设中的每一个如何影响增量成本效益比。未来评估的主要建议是：（i）建模研究应证明选择基准实用程序是合理的，特别是如果患者被认为在治愈治疗后完全康复；（ii）假阳性扫描对生活质量的影响应建模，至少在敏感性分析中；（iii）建模人员应证明与术后恢复有关的假设，最好基于当地实践的知识；(iv)应用于肺癌诊断的效用应得到适当的来源和计算；(v)应考虑对与年龄相关的生活质量下降进行调整，特别是对于检查寿命的模型。
