Zielhuis G A, Straatman H, Rach G H, van den Broek P
Department of Epidemiology, Nijmegen University, The Netherlands.
Int J Epidemiol. 1990 Dec;19(4):1037-44. doi: 10.1093/ije/19.4.1037.
Methods of analysing and presenting longitudinal data on health characteristics that have a fluctuating course are by no means standard. Several possibilities are discussed and tried out on longitudinal data on Otitis Media with Effusion (OME) in Dutch children. If the purpose of a study is descriptive, a graphical presentation based on an idea from Fiellau-Nikolajsen is preferred. For estimation purposes on the other hand, hazard functions are the most suitable. Further reduction of the data does not lead to a clear view into the complex nature of the course of a disease such as OME.
分析和呈现具有波动病程的健康特征纵向数据的方法绝非标准方法。本文讨论了几种可能性,并在荷兰儿童中耳积液(OME)的纵向数据上进行了试验。如果研究目的是描述性的,基于菲劳 - 尼古拉耶森的想法进行图形呈现是首选。另一方面,对于估计目的,风险函数是最合适的。对数据的进一步简化并不能清晰地洞察诸如OME这种疾病病程的复杂本质。