Department for Neurobiology, Care Sciences and Society, Division of Neurogeriatrics, Karolinska Institutet, Solna, Sweden.
Department of Psychiatry and Neuropsychology, Alzheimer Centre Limburg, School for Mental Health and Neurosciences, Maastricht University, Maastricht, the Netherlands.
Alzheimers Dement. 2020 May;16(5):789-796. doi: 10.1002/alz.12050. Epub 2020 Mar 22.
Loss to follow-up in dementia studies is common and related to cognition, which worsens over time. We aimed to (1) describe dropout and missing cognitive data in the Swedish dementia registry, SveDem; (2) identify factors associated with dropout; and (3) estimate propensity scores and use them to adjust for dropout.
Longitudinal cognitive data were obtained from 53,880 persons from the SveDem national quality dementia registry. Inverse probability of censoring weights (IPCWs) were estimated using a logistic regression model on dropout.
The mean annualized rate of change in Mini-Mental State Examination (MMSE) in those with a low MMSE (0 to 10) was likely underestimated in the complete case analysis (+1.5 points/year) versus the IPCW analysis (-0.3 points/year).
Handling dropout by IPCWs resulted in plausible estimates of cognitive decline. This method is likely of value to adjust for biased dropout in longitudinal cohorts of dementia.
痴呆症研究中的失访很常见，且与认知功能相关，认知功能会随时间恶化。我们旨在：(1)描述瑞典痴呆症登记处 SveDem 的脱落和缺失认知数据；(2)确定与脱落相关的因素；(3)估计倾向评分并将其用于调整脱落。
从 SveDem 全国质量痴呆症登记处的 53880 名患者中获得纵向认知数据。使用逻辑回归模型对脱落进行反向概率 censoring 权重（IPCW）估计。
在简易精神状态检查（MMSE）得分较低（0 到 10）的患者中，完全案例分析（每年 MMSE 变化率+1.5 分/年）与 IPCW 分析（每年 MMSE 变化率-0.3 分/年）相比，MMSE 年化率的估计值可能被低估。
通过 IPCW 处理脱落导致认知下降的估计值更合理。这种方法可能对调整痴呆纵向队列中因偏倚导致的脱落有用。
