Program in Public Health, Renaissance School of Medicine, Stony Brook University, Stony Brook, New York, USA.
Department of Family, Population, and Preventive Medicine, Renaissance School of Medicine, Stony Brook University, Stony Brook, New York, USA.
Alzheimers Dement. 2024 Aug;20(8):5540-5550. doi: 10.1002/alz.14053. Epub 2024 Jul 12.
Limited observational windows lead to conflicting results in studies examining educational differences in Alzheimer's disease and related dementias (ADRD) risk, due to observational window bias relative to onset of accelerated cognitive decline. This study tested a novel model to address observational window bias and tested for the presence and sources of disparities in accelerated cognitive declines due to ADRD.
The sample examined 167,314 cognitive assessments from 32,441 Health and Retirement Study participants. We implemented a parametric non-linear nested longitudinal regression and reported multivariable-adjusted nodal incidence ratios (aNIR).
University degrees were associated with lower incidence (aNIR = 0.253, 95% confidence interval [CI] = [0.221 to 0.289], p < 0.001), while black participants had a higher incidence (aNIR = 1.995, [1.858 to 2.141], p < 0.001) of accelerated cognitive decline, adjusting for demographic, sociobehavioral, and medical risk factors. Sex-stratified analyses identified diminished educational returns for women and increased incidence among minoritized women.
Addressing observational window bias reveals large social inequalities in the onset of accelerated cognitive declines indicative of ADRD.
This study identifies observational window bias as a source of conflicting results among previous studies of educational achievement in Alzheimer's disease and related dementias (ADRD) disparities. The study locates preclinical accelerated cognitive decline, which is indicative of ADRD while occurring 10+ years prior to symptom onset, as a site to study ADRD disparities that mitigates observational window bias. A novel method, nested non-linear regression, is developed to test for differences in the onset of accelerated cognitive decline. Educational and racial/ethnic disparities are demonstrated in the onset of accelerated cognitive decline, as are their intersecting differences with sex/gender.
由于观察窗口偏倚与加速认知衰退的发病有关，因此研究阿尔茨海默病和相关痴呆症（ADRD）风险中教育差异的观察性研究窗口有限，导致结果相互矛盾。本研究测试了一种新的模型来解决观察性窗口偏差，并测试了由于 ADRD 导致加速认知衰退的存在和差异的来源。
该样本检查了 32441 名健康与退休研究参与者的 167314 项认知评估。我们实施了参数非线性嵌套纵向回归，并报告了多变量调整后的节点发病率比（aNIR）。
大学学历与较低的发病率相关（aNIR=0.253，95%置信区间[CI]=[0.221 至 0.289]，p<0.001），而黑人参与者的加速认知衰退发病率更高（aNIR=1.995，[1.858 至 2.141]，p<0.001），调整了人口统计学、社会行为和医疗风险因素。按性别分层分析确定了女性教育回报的减少和少数族裔女性发病率的增加。
解决观察窗口偏差揭示了加速认知衰退发病方面的巨大社会不平等，这表明与 ADRD 相关。
本研究确定观察窗口偏差是以前关于阿尔茨海默病和相关痴呆症（ADRD）差异中教育成就研究结果相互矛盾的一个来源。该研究确定了发生在症状出现前 10 多年的、预示 ADRD 的临床前加速认知衰退，作为研究 ADRD 差异的一个位置，从而减轻观察窗口偏差。开发了一种新方法，即嵌套非线性回归，用于测试加速认知衰退发病的差异。在加速认知衰退的发病中表现出教育和种族/民族差异，以及与性别/性别相关的交叉差异。
