摘要
目的 研究一种简单、实用处理零乱的任意点追踪观测数据的统计分析方法 .方法 用 SAS软件 NL IN过程拟合 2 0例正常儿童与 2 8例单纯性肥胖症患儿摄入液体试验餐后胃内残留率的动态变化指数曲线模型 y=aebx,然后采用Simpson法进行曲线下面积的计算 ,考察了曲线下的面积、胃排空平均速度等反应特征 .结果 两组儿童液体胃排空曲线、曲线下面积以及曲线参数无显著差别 . 0～ 1 2 0 min期间儿童液体胃排空平均速度为 (0 .6 2± 0 .0 3) %· min- 1 ,1 0 % ,5 0 %以及 90 %胃排空时间分别为 (8.7± 3.4) ,(45 .9± 1 2 .7)和 (1 4 7.9± 39.8) min.结论 反应特征分析能有效地保证零乱的任意点追踪观测资料信息的有用性 .
AIM To explore a convenient and practical statistical method of analyzing arbitrary time longitudinal data. METHODS The exponential curve model y = a e bx was fitted by SAS software PROC NLIN, which described the dynamic percentage changes of remnants in stomach after ingestion of liquid test meal in 20 normal children and 28 children with simple obesity. Then the area under the curve was calculated by Simpson method and response features such as area under the curve and the mean rate of gastric emptying were studied. RESULTS No significant difference was found in the liquid gastric emptying curves, area under the curve and curve parameters of both groups. The mean rate of liquid gastric emptying was (0.62±0.03)%/min. 10%, 50% and 90% gastric emptying time were (8.7±3.4) min, (45.9±12.7) min and (147.9±39.8) min respectively. CONCLUSION Response feature analysis will effectively ensure the availability of arbitrary time longitudinal data information.
出处
《第四军医大学学报》
2000年第1期15-17,共3页
Journal of the Fourth Military Medical University
基金
国家自然科学基金!( 3 93 70 63 7)
关键词
肥胖
追踪数据
反应特征分析
儿童
obesity
longitudinal data
response feature ana lysis
