Marks C, Katch V
Med Sci Sports Exerc. 1986 Aug;18(4):485-8.
Ten subjects were studied to determine within-subject variance (Swi2), and its components, biovariability (Si2) and technological error (Se2) of residual lung volume (RLV). Each subject's RLV was determined 18 times using the oxygen dilution method (O2D) over a 5-day period. Se2 was measured using a 9.0 l spirometer as a simulated subject and estimating its added space by the O2D method for 20 repeat trials. The mean +/- SD of RLV across subjects was 1.408 +/- 0.3143 l. The average Swi2 for the 18 trials was 0.0054 (Swi = +/- 0.0735 l). The variance associated with a trend effect (variance of mean trial scores, Sc2) was 0.00050 (Sc = +/- 0.0224 l). Se2 was found to be 0.00104 (Se = +/- 0.0323 l). Si2 calculated by subtracting Se2 and Sc2 from Swi2, was 0.0039 (Si = +/- 0.0621 l). It was demonstrated that body fat weight calculations can vary because of the within-subject variability in RLV. The calculated reliability of RLV across subjects was r = 0.95. It is concluded that RLV determined by the O2D method is very reliable. Moreover, biological variance accounts for 72% of Swi2 while technological variance and trend effect variance account for 19 and 9% of Swi2 respectively.
对10名受试者进行了研究，以确定残气量（RLV）的受试者内方差（Swi2）及其组成部分，即生物变异性（Si2）和技术误差（Se2）。在5天的时间内，使用氧稀释法（O2D）对每位受试者的RLV进行了18次测定。使用一台9.0升的肺量计模拟受试者，并通过O2D方法对其增加的空间进行估计，进行20次重复试验来测量Se2。受试者的RLV均值±标准差为1.408±0.3143升。18次试验的平均Swi2为0.0054（Swi =±0.0735升）。与趋势效应相关的方差（平均试验得分的方差，Sc2）为0.00050（Sc =±0.0224升）。发现Se2为0.00104（Se =±0.0323升）。通过从Swi2中减去Se2和Sc2计算得出的Si2为0.0039（Si =±0.0621升）。结果表明，由于RLV的受试者内变异性，身体脂肪重量的计算可能会有所不同。计算得出的受试者间RLV可靠性为r = 0.95。得出的结论是，通过O2D方法测定的RLV非常可靠。此外，生物方差占Swi2的72%，而技术方差和趋势效应方差分别占Swi2的19%和9%。
