[R-meta] Deeks’ funnel plot asymmetry test for publication bias

[Guido Schwarzer]

Am 10.08.20 um 12:06 schrieb Tobias Saueressig:

> Hello,
> I am performing a diagnostic accuracy meta-analysis. As a check for 
> publication bias/small study effects I want to use the Deeks' funnel 
> plot asymmetry test for publication bias. I could not find a 
> implementation in a software program. Does anybody know if there is a 
> package that contains this test?

Probably, a little bit late, however, for the record and future queries. ;-)

I implemented the Deeks' test in R package *meta*. In order to use it 
one has to conduct a meta-analysis with /metabin()/ using the diagnostic 
odds ratio as summary measure. Then, R function /metabias()/ conducts 
the linear regression test using the inverse of the square root of the 
effective study size.

Using Wolfgang's example dataset:

dat <- read.table(header = TRUE, text = "
tp np  tn  nn
10 25  47  53
15 24  29  32
10 14  44  47
13 23 123 123
10 32  55  56
16 61 184 186
12 49 137 138
13 26  66  74
16 80  NA  NA
24 42  55  57
17 27  85  91
  5 22  45  46
18 31 240 245
24 28 104 104
26 55  81  89")

library(meta)
settings.meta(digits = 2)

## Drop study with missing information and conduct meta-analysis
##
dat.nomiss <- subset(dat, !is.na(tn))
##
m <- metabin(tp, np, nn - tn, nn, data = dat.nomiss,
              sm = "DOR", addincr = TRUE)

## Funnel plot with inverse of square root of effective study size on y-axis
##
funnel(m, yaxis = "ess", xlim = c(1, 1100), ylim = c(0.15, 0), studlab = 
TRUE)
## same result with meta, > 4.15-1:
## funnel(m, xlim = c(1, 1100), studlab = TRUE)

## Deeks' test for funnel plot asymmetry
##
metabias(m)

Best wishes, Guido


	[[alternative HTML version deleted]]