Unmatched case-control OR

Formula: OR = ad/bc from the 2x2

Usage

Ccdsgn(cases, controls, exposed = NULL, unexposed = NULL, conf = 0.95)

Arguments

cases

Either (a, b) counts, or a 0/1 exposure indicator.

controls

Either (c, d) counts, or a 0/1 exposure indicator.

exposed, unexposed

Ignored; kept for backward compatibility.

conf

Confidence level.

Value

List with estimate, a, b, c, d, log_or, se_log, ci_low, ci_high, chisq, significant, n, method.

Details

a exposed cases, b unexposed cases, c exposed controls, d unexposed controls. In a case-control design the risks themselves are not identified – sampling is on the outcome – but the odds ratio is, which is Cornfield's point. The interval uses Woolf's standard error on the log scale, sqrt(1/a + 1/b + 1/c + 1/d).

References

Cornfield (1951), Proc. 2nd Berkeley Symp. 4:135-148; Woolf (1955), Ann. Human Genetics 19(4):251-253.

Examples

Ccdsgn(cases = list(a = 1, b = 2), controls = list(a = 1, b = 2))
#> $estimate
#> [1] 1
#> 
#> $a
#> [1] 1
#> 
#> $b
#> [1] 2
#> 
#> $c
#> [1] 1
#> 
#> $d
#> [1] 2
#> 
#> $log_or
#> [1] 0
#> 
#> $se_log
#> [1] 1.732051
#> 
#> $ci_low
#> [1] 0.0335487
#> 
#> $ci_high
#> [1] 29.80742
#> 
#> $chisq
#> [1] 0
#> 
#> $significant
#> [1] 0
#> 
#> $n
#> [1] 6
#> 
#> $method
#> [1] "unmatched case-control odds ratio"
#> 
#> attr(,"class")
#> [1] "morie_rich_result" "list"