Department of Mathematical Sciences, Bentley University, Waltham, Massachusetts, United States of America.
School of Computing and Data Science, Wentworth Institute of Technology, Boston, Massachusetts, United States of America.
PLoS One. 2024 Aug 14;19(8):e0306270. doi: 10.1371/journal.pone.0306270. eCollection 2024.
Mild cognitive impairment (MCI) is a pre-clinical stage of Alzheimer's disease (AD). Understanding the transition probabilities across the disease continuum of AD, ranging from MCI to AD to Mortality is crucial for the economic modeling of AD and effective planning of future interventions and healthcare resource allocation decisions. This study uses the Multi-state Markov model to quantify the transition probabilities along the disease progression and specifically investigates medications as modifiable risk factors of AD associated with accelerated or decelerated transition times from MCI to AD, MCI to mortality, and AD to mortality.
Individuals with MCI were identified from the National Alzheimer's Coordinating Center between September 2005 and May 2021. A three-state Markov model was postulated to model the disease progression among three states: MCI, AD, and mortality with adjustment for demographics, genetic characteristics, comorbidities and medications. Transition probabilities, the total length of stay in each state, and the hazard ratios of the use of medications for diabetes, hypertension, and hypercholesterolemia (the known modifiable risk factors of AD) were evaluated for these transitions.
3,324 individuals with MCI were identified. The probability of developing AD after one year since the initial diagnosis of MCI is 14.9%. After approximately 6 years from the initial diagnosis of MCI, the probability of transitioning to AD increases to nearly 41.7% before experiencing a subsequent decline. The expected total lengths of stay were 5.38 (95% CI: 0.002-6.03) years at MCI state and 7.61 (95%CI: 0.002-8.88) years at AD state. Patients with active use of lipid-lowering agents were associated with significantly lower hazards of transitioning from MCI to AD (HR: 0.83, 95%CI:0.71-0.96), MCI to mortality (HR: 0.51, 95%CI:0.34-0.77), and AD to mortality (HR: 0.81, 95%CI:0.66-0.99).
Results suggest that lipid-lowering agents may confer a protective effect, delaying the onset of AD. Additionally, lipid-lowering agents indicate a favorable association with a longer survival time.
轻度认知障碍（MCI）是阿尔茨海默病（AD）的临床前阶段。了解 AD 疾病连续体（从 MCI 到 AD 再到死亡）的转移概率对于 AD 的经济建模以及未来干预措施和医疗资源分配决策的有效规划至关重要。本研究使用多状态马尔可夫模型来量化疾病进展过程中的转移概率，并专门研究药物作为与 MCI 向 AD、MCI 向死亡和 AD 向死亡的加速或减速转移时间相关的 AD 可改变的风险因素。
从 2005 年 9 月至 2021 年 5 月，从国家阿尔茨海默病协调中心识别出 MCI 个体。提出了一个三状态马尔可夫模型来模拟三种状态之间的疾病进展：MCI、AD 和死亡，并调整了人口统计学、遗传特征、合并症和药物。评估了这些转移的转移概率、每个状态的总停留时间以及用于糖尿病、高血压和高胆固醇血症（AD 的已知可改变风险因素）的药物使用的危险比。
确定了 3324 名 MCI 个体。在初始 MCI 诊断后一年，发展为 AD 的概率为 14.9%。从 MCI 的初始诊断大约 6 年后，向 AD 的转移概率增加到近 41.7%，然后才再次下降。预期的总停留时间分别为 MCI 状态下的 5.38（95%CI：0.002-6.03）年和 AD 状态下的 7.61（95%CI：0.002-8.88）年。积极使用降脂药的患者从 MCI 向 AD（HR：0.83，95%CI：0.71-0.96）、MCI 向死亡（HR：0.51，95%CI：0.34-0.77）和 AD 向死亡（HR：0.81，95%CI：0.66-0.99）的转移风险显著降低。
结果表明，降脂药可能具有保护作用，延迟 AD 的发作。此外，降脂药与更长的生存时间呈有利关联。
