New weight guidelines for Americans: justified or injudicious?
New weight guidelines for Americans: justified or injudicious?
复制标题
美国人的新体重指南：合理还是不明智？
DOI：
10.1093/ajcn/53.5.1102
复制
发表时间：
1991
期刊：
影响因子：
--
通讯作者：
VanItallie,T
中科院分区：
文献类型：
--
作者：
Willett,WC;Stampfer,M;Manson,J;VanItallie,T
美国农业部在最新出版的《美国人膳食指南》中发布了修订后的成人建议体重(1)。这些体重指南得到了广泛的宣传，特别是因为它们比以前的建议有了很大的提高。最值得注意的是，与19-34岁的人相比，35岁及以上的人的建议体重有所增加。例如，对于身高6英尺(183厘米)的人，19-34岁的建议体重范围为140-184磅(63.6-83.6公斤)，35岁的建议体重范围为155-199磅(70.5-90.5公斤)。相比之下，1959年大都会人寿保险公司的表格(2)中中等身材男性的建议体重上限为168磅(76.4公斤)，并未显示体重随着年龄的增长而增加。新的建议权重据说是根据国家研究委员会(NRC)1989年的报告(3)得出的。然而，NRC的报告没有具体的建议体重，只提出了非常笼统的建议；该报告不建议增加理想体重。因此，这些新建议的基础尚不清楚。它们很可能来自Andres等人(4)基于对人寿保险数据的特定年龄分析的出版物。正如我们之前指出的，这些分析存在致命的缺陷，因为吸烟的影响没有得到控制。吸烟者往往比不吸烟者体重轻得多，但由于吸烟，许多原因导致的死亡率很高。这种现象往往会人为地增加体重较轻的人的死亡率，并将明显的最佳相对体重向更高的水平移动。因为吸烟的不良影响往往会累积几十年，这种偏见主要出现在老年人身上，35岁的人中只有很小的程度。其结果是，随着年龄的增长，人们明显倾向于更高的最佳相对权重。此外，从青少年时期就开始大量吸烟的这一代吸烟者最近才进入癌症和冠心病高发年龄(6)，因此目前的精算数据比以前的报告更有偏见。吸烟对体重和死亡率之间的关系的影响是显而易见的，因为对于体重较轻的人来说，超额死亡率的很大一部分是由于吸烟引起的肺癌[6]。许多研究中的第二个重要偏差是没有考虑疾病对体重的影响。许多疾病，甚至在确诊之前，就会导致体重减轻，并增加死亡风险。观察到体重不足的人的超额死亡率往往主要发生在随访的早期，而与超重相关的死亡率则延迟[7]，这一观察结果支持了这种存在
The US Department ofAgriculture(USDA) has issued revised suggested weights for adults in the newly published Dietary Guidelinesfor Americans(1). These weight guidelines received widespread publicity, particularly because they represent a substantial increase from earlier recommendations. Most notable were the increased suggested weights for people 35 y and older compared with those 19-34 y. For example, for a person(man or woman) 6 ft (183 cm) tall, the suggested range is 140-184 lb (63.6-83.6 kg) for ages 19-34 y and 155-199 lb (70.5-90.5 kg) for ages 35 y. This compares with an upper limit of 168 lb (76.4 kg) for a man of medium frame from the 1959 Metropolitan Life Insurance Company tables (2), with no suggested weight gain with increasing age. The new suggested weights are said to be derived from the National Research Council (NRC) Report of 1989 (3). However, the NRC report contains no specific suggested weights and makes only very general recommendations; no increase in ideal weight is recommended in that report. Thus the basis of these new recommendations is unclear. It is likely that they are derived from the publication by Andres et al (4) based on age-specific analyses of life-insurance data. As we (5) pointed out before, these analyses are fatally flawed because the effects of cigarette smoking were not controlled for. Cigarette smokers tend to weigh appreciably less than do nonsmokers but have high rates of death from many causes because oftheir smoking. This phenomenon tends to create an artifactual increase in mortality for lighter people and to shift the apparent optimal relative weight toward higher levels. Because the adverse effects of smoking tend to accumulate over many decades, this bias will be seen primarily for older people and to a very minimal degree in people< 35 y of age. The result is an apparent shift toward higher optimal relative weights as people grow older. Moreover, the generation of smokers who began smoking heavily as teen-agers has only recently entered the age of high rates of cancer and coronary heart disease (6), so that the more current actuarial data are more seriously biased than the earlier reports. The smoking effect on the relation between body weight and mortality is clearly evident in that a substantial proportion of the excess mortality for lighter individuals is due to cigarette-induced lung cancer (6).A second important bias in many studies is a failure to consider the effect ofdisease on body weight. Many diseases, even before diagnosis, can cause weight loss and also increase the risk of death. The observation that excess mortality for underweight individuals tends to occur primarily early in follow-up whereas overweight-related mortality is delayed(7) supports the presence