利用尿样估算人群盐摄入量的平均变化。

Estimating mean change in population salt intake using spot urine samples.

机构信息

George Institute for Global Health, University of Sydney, Camperdown, NSW, Australia.

Deakin University, Geelong, Australia, Institute for Physical Activity and Nutrition Research, Locked Bag 20000, Waurn Ponds, Geelong, VIC 3220, Australia.

出版信息

Int J Epidemiol. 2017 Oct 1;46(5):1542-1550. doi: 10.1093/ije/dyw239.

DOI:10.1093/ije/dyw239

PMID:28338706

Abstract

BACKGROUND

Spot urine samples are easier to collect than 24-h urine samples and have been used with estimating equations to derive the mean daily salt intake of a population. Whether equations using data from spot urine samples can also be used to estimate change in mean daily population salt intake over time is unknown. We compared estimates of change in mean daily population salt intake based upon 24-h urine collections with estimates derived using equations based on spot urine samples.

METHODS

Paired and unpaired 24-h urine samples and spot urine samples were collected from individuals in two Australian populations, in 2011 and 2014. Estimates of change in daily mean population salt intake between 2011 and 2014 were obtained directly from the 24-h urine samples and by applying established estimating equations (Kawasaki, Tanaka, Mage, Toft, INTERSALT) to the data from spot urine samples. Differences between 2011 and 2014 were calculated using mixed models.

RESULTS

A total of 1000 participants provided a 24-h urine sample and a spot urine sample in 2011, and 1012 did so in 2014 (paired samples n = 870; unpaired samples n = 1142). The participants were community-dwelling individuals living in the State of Victoria or the town of Lithgow in the State of New South Wales, Australia, with a mean age of 55 years in 2011. The mean (95% confidence interval) difference in population salt intake between 2011 and 2014 determined from the 24-h urine samples was -0.48g/day (-0.74 to -0.21; P < 0.001). The corresponding result estimated from the spot urine samples was -0.24 g/day (-0.42 to -0.06; P = 0.01) using the Tanaka equation, -0.42 g/day (-0.70 to -0.13; p = 0.004) using the Kawasaki equation, -0.51 g/day (-1.00 to -0.01; P = 0.046) using the Mage equation, -0.26 g/day (-0.42 to -0.10; P = 0.001) using the Toft equation, -0.20 g/day (-0.32 to -0.09; P = 0.001) using the INTERSALT equation and -0.27 g/day (-0.39 to -0.15; P < 0.001) using the INTERSALT equation with potassium. There was no evidence that the changes detected by the 24-h collections and estimating equations were different (all P > 0.058). Separate analysis of the unpaired and paired data showed that detection of change by the estimating equations was observed only in the paired data.

CONCLUSIONS

All the estimating equations based upon spot urine samples identified a similar change in daily salt intake to that detected by the 24-h urine samples. Methods based upon spot urine samples may provide an approach to measuring change in mean population salt intake, although further investigation in larger and more diverse population groups is required.

摘要

背景

与 24 小时尿液样本相比，随机尿样更容易收集，并且已经用于通过估计方程来得出人群的平均每日盐摄入量。使用来自随机尿样的数据的方程是否也可以用于估计随时间变化的人群平均每日盐摄入量的变化尚不清楚。我们比较了基于 24 小时尿液收集的平均每日人群盐摄入量变化的估计值与使用基于随机尿样的方程得出的估计值。

方法

在澳大利亚的两个人群中，于 2011 年和 2014 年收集了个体的配对和非配对 24 小时尿液样本和随机尿样。从 24 小时尿液样本中直接获得了 2011 年至 2014 年每日平均人群盐摄入量变化的估计值，并通过应用已建立的估计方程（Kawasaki、Tanaka、Mage、Toft、INTERSALT）来对随机尿样中的数据进行应用。使用混合模型计算 2011 年和 2014 年之间的差异。

结果

共有 1000 名参与者于 2011 年提供了 24 小时尿液样本和随机尿样，其中 1012 名参与者于 2014 年提供了（配对样本 n=870；非配对样本 n=1142）。参与者为居住在维多利亚州或新南威尔士州利特高镇的社区居民，2011 年的平均年龄为 55 岁。通过 24 小时尿液样本确定的 2011 年至 2014 年人群盐摄入量的平均（95%置信区间）差异为-0.48g/天（-0.74 至-0.21；P<0.001）。从随机尿样中估计的相应结果为-0.24g/天（-0.42 至-0.06；P=0.01）使用 Tanaka 方程，-0.42g/天（-0.70 至-0.13；p=0.004）使用 Kawasaki 方程，-0.51g/天（-1.00 至-0.01；P=0.046）使用 Mage 方程，-0.26g/天（-0.42 至-0.10；P=0.001）使用 Toft 方程，-0.20g/天（-0.32 至-0.09；P=0.001）使用 INTERSALT 方程，-0.27g/天（-0.39 至-0.15；P<0.001）使用 INTERSALT 方程和钾。没有证据表明 24 小时采集和估计方程检测到的变化不同（所有 P>0.058）。对非配对和配对数据的单独分析表明，只有在配对数据中才可以通过估计方程检测到变化。