Monteiro C A
Department of Nutrition, School of Public Health, University of São Paulo, Brazil.
Bull World Health Organ. 1991;69(6):761-6.
Two methods for estimating the prevalence of growth retardation in a population are evaluated: the classical method, which is based on the proportion of children whose height is more than 2 standard deviations below the expected mean of a reference population; and a new method recently proposed by Mora, which is based on the whole height distribution of observed and reference populations. Application of the classical method to several simulated populations leads to the conclusion that in most situations in developing countries the prevalence of growth retardation is grossly underestimated, and reflects only the presence of severe growth deficits. A second constraint with this method is a marked reduction of the relative differentials between more and less exposed strata. Application of Mora's method to the same simulated populations reduced but did not eliminate these constraints. A novel method for estimating the prevalence of growth retardation, which is based also on the whole height distribution of observed and reference populations, is also described and evaluated. This method produces better estimates of the true prevalence of growth retardation with no reduction in relative differentials.
经典方法，该方法基于身高比参考人群预期均值低2个以上标准差的儿童比例；以及莫拉最近提出的新方法，该方法基于观察人群和参考人群的整体身高分布。将经典方法应用于多个模拟人群得出的结论是，在发展中国家的大多数情况下，生长发育迟缓的患病率被严重低估，且仅反映了严重生长缺陷的存在。该方法的另一个局限性是，暴露程度较高和较低的阶层之间的相对差异显著减小。将莫拉的方法应用于相同的模拟人群，减少但并未消除这些局限性。还描述并评估了一种同样基于观察人群和参考人群的整体身高分布来估算生长发育迟缓患病率的新方法。该方法能更好地估算出生长发育迟缓的真实患病率，且不会减小相对差异。
