Mohs R C, Schmeidler J, Aryan M
Department of Psychiatry, Mount Sinai School of Medicine, New York, New York 10029, USA.
Stat Med. 2000;19(11-12):1401-9. doi: 10.1002/(sici)1097-0258(20000615/30)19:11/12<1401::aid-sim432>3.0.co;2-x.
This paper reviews data on the natural history of symptoms in patients with Alzheimer's disease (AD) and describes some of the problems encountered in analysing longitudinal data in this population. Data on cognition, functional ability and psychiatric or behavioural symptoms have all been obtained from AD patients. Because of attrition, the length of follow-up is not uniform for all patients and neither is the frequency of evaluation. Furthermore, patients enter longitudinal studies with a wide range of symptom severity and longitudinal decline in cognition and function is distinctly non-linear. Behavioural symptoms do not progress regularly in AD but are episodic phenomena not closely related to cognition or function. Strengths and limitations of various analytic techniques used for hypothesis testing with these longitudinal data are described.
本文回顾了阿尔茨海默病（AD）患者症状自然史的数据，并描述了在分析该人群纵向数据时遇到的一些问题。认知、功能能力以及精神或行为症状的数据均来自AD患者。由于失访，并非所有患者的随访时间长度都一致，评估频率也不一致。此外，患者进入纵向研究时症状严重程度差异很大，认知和功能的纵向衰退明显呈非线性。AD患者的行为症状并非规律性进展，而是与认知或功能无密切关系的发作性现象。本文还描述了用于对这些纵向数据进行假设检验的各种分析技术的优缺点。
