Department of Public Health, College of Medicine, National Cheng Kung University, Tainan, Taiwan.
Department of Surgery, National Cheng Kung University Hospital, Tainan, Taiwan.
J Formos Med Assoc. 2022 Mar;121(3):633-642. doi: 10.1016/j.jfma.2021.06.013. Epub 2021 Jul 8.
BACKGROUND/PURPOSE: We estimated loss-of-life expectancy (LE) and lifetime medical expenditures (LME) stratified by stages to evaluate the cost-effectiveness of breast cancer (BC) screening in Taiwan.
We interlinked four national databases- Cancer Registry, Mortality Registry, National Health Insurance Claim, and Mammography Screening. A cohort of 123,221 BC was identified during 2002-2015 and followed until December 31, 2017. We estimated LE and loss-of-LE by rolling extrapolation algorithm using age-, sex-, and calendar-year-matched referents simulated from vital statistics. LME was estimated by multiplying monthly cost with survival probability and adjusted for annual discount rate. We calculated incremental cost-effectiveness ratio (ICER) by comparing the loss-of-LE of those detected by screening versus non-screening after accounting for administration fees and radiation-related excess BC.
The LEs of stages I, II, III, and IV were 31.4, 27.2, 20.0, and 5.2 years, respectively, while the loss-of-LEs were 1.2, 4.9, 11.7, and 25.0 years with corresponding LMEs of US$ 73,791, 79,496, 89,962, and 66,981, respectively. The difference in LE between stages I and IV was 26.2 years while that of loss-of-LE was 23.8 years, which implies that a potential lead time bias may exist if diagnosis at younger ages for earlier stages were not adjusted for. The ICER of mammography seemed cost-saving after the coverage exceeded half a million.
Mammography could detect BC early and be cost-saving after adjustment for different distributions of age and calendar year of diagnosis. Future studies exploring healthcare expenditure and impaired quality of life for false-positive cases are warranted.
背景/目的：我们通过分层评估各期别来估算生命损失预期寿命（LE）和终生医疗支出（LME），以评估台湾地区乳腺癌（BC）筛查的成本效益。
我们将四个国家数据库（癌症登记处、死亡率登记处、全民健康保险理赔和乳房 X 光筛查）进行了链接。我们在 2002 年至 2015 年间确定了一个 123221 例 BC 队列，并随访至 2017 年 12 月 31 日。我们使用从生命统计中模拟的年龄、性别和日历年匹配的参考人群的滚动外推算法来估算 LE 和 LE 的损失。LME 通过将每月成本乘以生存概率并根据年度贴现率进行调整来估算。我们通过比较筛查组和非筛查组的 LE 损失，计算出增量成本效益比（ICER），并考虑了管理费用和与辐射相关的超额 BC。
I 期、II 期、III 期和 IV 期的 LE 分别为 31.4、27.2、20.0 和 5.2 年，LE 损失分别为 1.2、4.9、11.7 和 25.0 年，相应的 LME 分别为 73791 美元、79496 美元、89962 美元和 66981 美元。I 期和 IV 期的 LE 差异为 26.2 年，LE 损失差异为 23.8 年，这意味着如果不调整年龄和诊断日历年的不同分布，可能存在潜在的领先时间偏倚。在覆盖范围超过 50 万之后，乳房 X 光检查的 ICER 似乎具有成本效益。
在考虑到不同年龄分布和诊断日历年的校正后，乳房 X 光检查可以更早地发现乳腺癌，并且具有成本效益。需要进一步研究来探索医疗支出和假阳性病例对生活质量的影响。
