我正在尝试用5个标准肥胖(3个可能的二元结果),胰岛素抵抗(3个可能的二元结果),血脂异常TGC(3个可能的二元结果),血脂异常HDL(2个可能的二元结果)和高血压(4个可能的二进制结果)。如果受试者在这5个标准中的任何3个中均为阳性,则将其视为代谢综合征阳性。
考虑到任何对4或5个标准呈阳性的受试者都将由代理覆盖,我尝试将它们合并到C5,3中。但是我尝试覆盖任何可能的组合时,我的代码太大了。是否可以使用运算符优先级来使代码更小以使其更紧凑?
METSYN <- array (NA,dim = dim(BancoTOTAL)[1] )
for (i in 1:791){
METSYN[i] <- ifelse ( #OID1 OID2
BancoTOTAL$sexo.x[i] == 0 && BancoTOTAL$cintura.x[i] > 90
&& BancoTOTAL$Glic[i] >= 100
&& BancoTOTAL$TRIG[i] > 150
|BancoTOTAL$sexo.x[i] == 1 && BancoTOTAL$cintura.x[i] > 80
&& BancoTOTAL$Glic[i] >= 100
&& BancoTOTAL$TRIG[i] > 150
|BancoTOTAL$IMC[i] > 30
&& BancoTOTAL$Glic[i] >= 100
&& BancoTOTAL$TRIG[i] > 150
and soon
, 1, 0)
}
这是我使用中间分类变量重做的代码
METSYN <- array(NA, dim = dim(BancoTOTAL)[1])
for (i in 1:(dim(BancoTOTAL)[1])){
METSYN[i] <- ifelse(
#OID1
BancoTOTAL$obesity[i] == 1
&& BancoTOTAL$insulinR[i] == 1
&& BancoTOTAL$dyslipidemiaTGC[i] == 1
#OID2
|BancoTOTAL$obesity[i] == 1
&& BancoTOTAL$insulinR[i] == 1
&& BancoTOTAL$dyslipidemiaHDL[i] == 1
#OIH
|BancoTOTAL$obesity[i] == 1
&& BancoTOTAL$insulinR[i] == 1
&& BancoTOTAL$HBP[i] == 1
#OD1D2
|BancoTOTAL$obesity[i] == 1
&& BancoTOTAL$dyslipidemiaTGC[i] == 1
&& BancoTOTAL$dyslipidemiaHDL[i] == 1
#OD1H
|BancoTOTAL$obesity[i] == 1
&& BancoTOTAL$dyslipidemiaTGC[i] == 1
&& BancoTOTAL$HBP[i] == 1
#OD2H
|BancoTOTAL$obesity[i] == 1
&& BancoTOTAL$dyslipidemiaHDL[i] == 1
&& BancoTOTAL$HBP[i] == 1
#ID1D2
|BancoTOTAL$obesity[i] == 1
&& BancoTOTAL$dyslipidemiaTGC[i] == 1
&& BancoTOTAL$dyslipidemiaHDL[i] == 1
#ID1H
|BancoTOTAL$insulinR[i] == 1
&& BancoTOTAL$dyslipidemiaTGC[i] == 1
&& BancoTOTAL$HBP[i] == 1
#ID2H
|BancoTOTAL$insulinR[i] == 1
&& BancoTOTAL$dyslipidemiaHDL[i] == 1
&& BancoTOTAL$HBP[i] == 1
#D1D2H
|BancoTOTAL$dyslipidemiaTGC[i] == 1
&& BancoTOTAL$dyslipidemiaHDL[i] == 1
&& BancoTOTAL$HBP[i] == 1
, 1, 0 )
}
答案 0 :(得分:0)
这对您有用吗? (使用您的中间类别变量)
METSYN <- unlist(lapply(1:dim(BancoTOTAL)[1], function (x){
(BancoTOTAL$obesity[x]+BancoTOTAL$insulinR[x]+BancoTOTAL$dyslipidemiaTGC[x]+BancoTOTAL$dyslipidemiaHDL[x]+BancoTOTAL$HBP[x])>=3
}))