Age, period, and cohort effects - Influences on differences between cross-sectional and longitudinal pulmonary function results
Age, period, and cohort effects - Influences on differences between cross-sectional and longitudinal pulmonary function results
复制标题
DOI：
10.1164/ajrccm/154.6_pt_2.s273
复制
发表时间：
1996-12-01
影响因子：
24.7
通讯作者：
Lebowitz, MD
中科院分区：
文献类型：
--
作者：
Lebowitz, MD
方法首先检查不同时期肺功能与年龄(和身体习惯)关系的横断面结果之间的差异，以及同一人群中横断面和纵向结果之间的差异。这些分析将利用图森AOD纵向研究(1-12、24、25)。Tucson对AOD的研究的总体目标包括：功能发育和衰退的生理特征、对吸烟作用的更好理解、确定儿童呼吸系统疾病的长期影响、研究环境/职业暴露和社会因素的影响，以及更好地预测AOD。这项研究利用了一个多阶段、分层的大约3800名非墨西哥裔美国白人(盎格鲁人)的分层抽样(5，24)。自注册以来(1972-73)，家庭中的新生儿和婚姻伴侣已经被登记，使登记的总人数达到约5600人。每次评估包括标准呼吸道症状和其他问卷(5，24)和标准最大呼气流量(MEFV)曲线(2，3，9，11，25，26)。盎格鲁人的纵向地位已经在前面介绍过了(5)；他们仍然被认为是图森人口的代表。正如预期的那样，死亡主要发生在高龄、男性、吸烟者和确诊为AOD的人中。在参与者中，男性比例和每个性别的平均年龄在调查中没有显著差异。然而，随着时间的推移，现在吸烟者的比例下降，而曾经吸烟者的比例增加。大多数这些比较将使用“健康”的受试者(无症状的从不吸烟者)(2，9，25)。我们评估了初始肺功能作为肺功能纵向变化率的决定因素和存活率对观察到的差异的影响。身体习惯和一些危险因素也将被用来评估可能的差异。作为预测指标和终点的症状将被视为肺功能本身的次要评估(1、4、5、8、9、12)。将评估将这些结果与其他人群和时期的结果进行比较的潜在益处。将审查评估频率、测量误差、变异来源和回归到平均值的潜在影响，并将评估作为差异的潜在基础的方法学方法(2、4、6-9、12、IS、19、20、25-30)。
METHODSDifferences/discrepancies will be examined first between cross-sectional results of lung function relationships with age (and body habitus) at different periods and between cross-sectional and longitudinal results in the same population. The Tucson Longitudinal Study of AOD will be utilized for these analyses (1-12, 24, 25). The overall objectives of the Tucson Study of AOD include: the physiologic characterization of growth and decline of function, an improved understanding of the role of smoking, determination of the long-term effects of childhood respiratory troubles, studies of the impacts of environmental/occupational exposures and social factors, and better prediction of AOD. The study utilizes a multistage, stratified cluster sample of about 3,800 non-Mexican American whites (" Anglo") described previously (5, 24). Since enrollment (1972-73), newborns and marital partners in the families have been enrolled, bringing the total enrolled to about 5,600. The evaluations during every survey included standard respiratory symptom and other questionnaires (5, 24) and standard maximal expiratory flowvolume (MEFV) curves (2, 3, 9, 11, 25, 26). The longitudinal status of the Anglo subjects has beenpresented previously (5); they are still considered representative of the Tucson population. As expected, deaths occurred preferentially in older ages, males, smokers, and those with diagnosed AOD. Within participants, the proportion of males, and the average age within each gender did not differ significantly by survey. However, the proportions of current smokers decreased and of ex-smokers increased over time.Most of these comparisons will be performed utilizing the" healthy" subjects (asymptomatic never smokers)(2, 9, 25). We have evaluated the influence of initial lung function as a determinant of the longitudinal rate of change in lung function and of survivorship on the differences observed. Body habitus and some of the risk factors also will be used to evaluate the possible differences. Symptoms as predictors and as endpoints will be looked at secondarily to the evaluation of lung function itself (1, 4, 5, 8, 9, 12). The potential benefit of comparing those findings to results in other populations and periods will be appraised. Potential effects of frequency of evaluation, measurement errors, sources of variation, and regression-to-the-mean will be examined, and the methodologic approaches as potential bases for differences will be evaluated (2, 4, 6-9, 12, IS, 19, 20, 25-30).