Chen Yu-Ching, Wang Wei-Ming, Lin Boniface J, Wang Jung-Der, Ku Li-Jung Elizabeth
Department of Public Health, College of Medicine, National Cheng Kung University, Tainan, Taiwan.
Department of Healthcare Administration and Medical Informatics, Kaohsiung Medical University, Kaohsiung, Taiwan.
Diabetes Metab J. 2025 Sep;49(5):1051-1063. doi: 10.4093/dmj.2024.0507. Epub 2025 Mar 26.
Diabetes is associated with reduced lifespan. To explore pay-for-performance (P4P) program and life expectancy (LE), we investigated the impact of interval between diabetes diagnosis and enrollment in P4P program on loss-of-LE among patients with diabetes in Taiwan.
From diabetes mellitus health database, which collected all newly-diagnosed patients with diabetes by calendar year, we selected patients, aged 40 to 64, with 503,662 in P4P group and 450,071 in non-P4P group, respectively, from 2004 to 2015, and followed them until the end of 2018 using Kaplan-Meier survival analysis. We simulated age-, gender-, and calendar yearmatched referents for each group through Monte Carlo method from Taiwan's vital statistics. We constructed a restricted cubic spline model on logit-transformed relative survival between each group and its corresponding matched referents, and applied a rolling-over algorithm month-by-month to extrapolate the survival function of each index group to lifetime to estimate the LE, which was subtracted from that of matched referents to obtain the loss-of-LE.
We found stratified analysis by interval showed that the earlier the enrollment, the lower the loss-of-LE, namely, 0.06±0.72 years for interval <1 year, 0.05±0.59 years for interval 1-4 years, 10.01±0.11 years for interval 5-9 years, and 12.77±0.14 years for interval 10-15 years, respectively (P<0.001), compared with 2.60±0.14 years for non-P4P group.
Early enrollment in the P4P program was associated with reduced loss-of-LE, indicating P4P might gain life if implemented early after diabetes diagnosis.
糖尿病与寿命缩短有关。为了探究绩效薪酬（P4P）计划与预期寿命（LE）之间的关系，我们调查了台湾糖尿病患者从糖尿病诊断到加入P4P计划的时间间隔对LE损失的影响。
从糖尿病健康数据库中，该数据库按日历年收集所有新诊断的糖尿病患者，我们分别选取了2004年至2015年年龄在40至64岁之间的患者，P4P组有503,662例，非P4P组有450,071例，并使用Kaplan-Meier生存分析对他们进行随访直至2018年底。我们通过蒙特卡罗方法从台湾人口动态统计数据中为每组模拟年龄、性别和日历年匹配的对照。我们在每组与其相应匹配对照之间的对数转换相对生存上构建了一个受限立方样条模型，并逐月应用滚动算法将每个指标组的生存函数外推至终生以估计LE，从匹配对照的LE中减去该LE以获得LE损失。
我们发现按时间间隔进行的分层分析表明，加入时间越早，LE损失越低，即时间间隔<1年时为0.06±0.72年，时间间隔1 - 4年时为0.05±0.59年，时间间隔5 - 9年时为10.01±0.11年，时间间隔10 - 15年时为12.77±0.14年（P<0.001），而非P4P组为2.60±0.14年。
早期加入P4P计划与降低LE损失相关，表明P4P计划如果在糖尿病诊断后早期实施可能会延长寿命。
