Zoccali Carmine, Tripepi Giovanni
Renal Research Institute, New York, USA.
Institute of Molecular Biology and Genetics (Biogem), Ariano Irpino, Italy.
J Nephrol. 2025 Mar;38(2):435-444. doi: 10.1007/s40620-024-02167-4. Epub 2024 Dec 7.
Longitudinal cohort studies are pivotal in medical research for understanding disease progression over time. These studies track a group of individuals across multiple time points, enabling the identification of risk factors and the evaluation of interventions. Traditional methods like linear mixed models, generalized estimating equations, and survival analysis often fall short in capturing the complex, non-linear patterns of disease progression. Trajectory analysis, a statistical technique that identifies distinct paths within longitudinal data, offers a more nuanced approach. This review delves into the methodological foundations of trajectory analysis, including data preparation, model selection, parameter estimation, model evaluation, and interpretation. It highlights the advantages of trajectory analysis, such as its ability to capture heterogeneity, handle various data types, and enhance predictive power. The application of trajectory analysis in nephrology, particularly in chronic kidney disease and diabetic nephropathy, demonstrates its utility in identifying distinct subgroups with different disease trajectories. Studies have shown that trajectory analysis can uncover patterns of renal function decline and proteinuria progression, providing insights that inform personalized treatment strategies. Despite its strengths, trajectory analysis requires advanced statistical knowledge, computational resources, and large sample sizes, which can be barriers for some researchers. Nevertheless, its ability to reveal complex disease patterns and improve predictive accuracy makes it a valuable tool in longitudinal studies. This review underscores the potential of trajectory analysis to enhance our understanding of disease progression and improve patient outcomes in nephrology and beyond.
纵向队列研究在医学研究中对于理解疾病随时间的进展至关重要。这些研究在多个时间点跟踪一组个体，从而能够识别风险因素并评估干预措施。诸如线性混合模型、广义估计方程和生存分析等传统方法在捕捉疾病进展的复杂、非线性模式方面往往存在不足。轨迹分析是一种在纵向数据中识别不同路径的统计技术，提供了一种更细致入微的方法。本综述深入探讨了轨迹分析的方法学基础，包括数据准备、模型选择、参数估计、模型评估和解释。它强调了轨迹分析的优势，例如其捕捉异质性、处理各种数据类型以及增强预测能力的能力。轨迹分析在肾脏病学中的应用，特别是在慢性肾脏病和糖尿病肾病中，证明了其在识别具有不同疾病轨迹的不同亚组方面的实用性。研究表明，轨迹分析可以揭示肾功能下降和蛋白尿进展的模式，为个性化治疗策略提供见解。尽管轨迹分析具有优势，但它需要先进的统计知识、计算资源和大样本量，这可能对一些研究人员来说是障碍。然而，其揭示复杂疾病模式和提高预测准确性的能力使其成为纵向研究中的一个有价值的工具。本综述强调了轨迹分析在增强我们对疾病进展的理解以及改善肾脏病学及其他领域患者结局方面的潜力。
