使用来自 Epitools 的oddsratio() 计算 R 中的优势比时出错

Question

我正在尝试计算暴露于焦虑率风险因素水平的优势比，数据由以下人员生成：

data<-structure(list(exposure = structure(c(3L, 1L, 2L), .Label = c("mild", 
"severe", "unaffected"), class = "factor"), count_not_anxious = c(108, 
337, 203), count_anxious = c(8, 48, 76)), .Names = c("exposure", 
"count_not_anxious", "count_anxious"), row.names = c(NA, -3L), class = "data.frame") 

看起来像：

    exposure count_not_anxious count_anxious
1 unaffected               108             8
2       mild               337            48
3     severe               203            76

我将数据专门用于“epitools”包中的oddsratio() 函数，但是oddsratio(data) 生成以下错误消息：

Error in nrow(x) : object 'x' not found

这是为什么？我是不是误解了应该如何为这个函数安排数据？

提前感谢您的任何建议！

Answer 1

看起来您可以通过将数据从三列数据框转换为两行矩阵来做到这一点...

> dd <- as.matrix(data[,-1])  ## drop first column
> rownames(dd) <- data[,1]
> oddsratio(dd)

结果：

$data
           count_not_anxious count_anxious Total
unaffected               108             8   116
mild                     337            48   385
severe                   203            76   279
Total                    648           132   780

$measure
                        NA
odds ratio with 95% C.I. estimate    lower    upper
              unaffected 1.000000       NA       NA
              mild       1.891296 0.911011  4.46757
              severe     4.956366 2.432386 11.56705

$p.value
            NA
two-sided      midp.exact fisher.exact   chi.square
  unaffected           NA           NA           NA
  mild       9.001770e-02 1.289249e-01 9.505682e-02
  severe     1.579224e-06 1.755645e-06 6.787096e-06

$correction
[1] FALSE

attr(,"method")
[1] "median-unbiased estimate & mid-p exact CI"