[R-meta] Ratio of Means

James Pustejovsky jepusto at gmail.com
Thu Jul 20 00:18:47 CEST 2017


I don't know the answer to your question exactly, but I would offer two
suggestions that might help you find a good approach for meta-analysis:

First, many clinical studies are woefully sloppy about distinguishing
between sample standard deviations versus standard errors (which might be
reported if the sample means are estimated with adjustment for baseline
covariates). If you have not already done so, it might be worth double
checking that your data are truly the raw sample standard deviations. If
some of them are actually SEs, then the effect size variances will be
under-estimated by a substantial factor, and this in turn might account for
the high degree of observed heterogeneity.

Second, are the outcomes (average drug doses) on a scale that is directly
comparable across studies? If so, either an unstandardized mean difference
or a response ratio could be an appropriate effect size metric. To
determine which metric to use, you might try creating a scatterplot with
control group means on the x axis and treatment group means on the y axis.
If the scatter resembles a 45-degree line that does not intercept the
origin, then the unstandardized mean difference might be a better summary.
If the scatter resembles a line (with higher or lower slope) through the
origin, then the response ratio might be a better summary.


On Wed, Jul 19, 2017 at 5:05 PM, Nathan Pace <n.l.pace at utah.edu> wrote:

> The outcomes in my example are average drug doses under an experimental
> and a control condition, so ratio scale measurements.
> In a RoM meta analysis there is very high heterogeneity (I2 = 94%).
> There is interest in whether a control group mean value as a moderator
> will improve the model and reduce the heterogeneity.
> Adding this moderator does not improve the model or reduce heterogeneity.
> It is generally recommended to not use a control group value as a
> moderator in meta regression of the mean difference (various papers by
> Sharp, etc).
> Does this recommendation still apply when the study values have been
> transformed into the RoM and not using the mean difference?
> Does this require a different model or a Bayesian meta regression?
> Your insights are much appreciated.
> Nathan
> Subject: RE: Ratio of Means
>     I would say the canonical reference on the log transformed ratio of
> means (often called the 'response ratio' in the field of ecology) is:
>     Hedges, L. V., Gurevitch, J., & Curtis, P. S. (1999). The
> meta-analysis of response ratios in experimental ecology. Ecology, 80(4),
> 1150-1156.
>     If you are using metafor, the data used in that paper can be found
> under dat.curtis1998 (i.e., help(dat.curtis1998) for more details).
>     This measure is used quite a lot in meta-analyses in that field since
> the measurements in those studies tend to be ratio scale measurements
> (e.g., the weight/mass/height of plants grown under different experimental
> conditions). And in fact, that is the big caveat: This outcome measure is
> only applicable for ratio scale measurements. Note that the examples in
> Friedrich, Adhikari, & Beyene (2008) are also all based on things that
> could be considered ratio scale measurements (urine output, serum
> creatinine, and creatinine clearance).
>     So, as long as you are dealing with such measurements, I think the
> response ratio is a perfectly fine outcome measure for a meta-analysis.
>     Minor sidenote: In rma.uni(), you should use test="knha" instead of
> knha=TRUE. The latter still works, but is now undocumented.
>     Best,
>     Wolfgang
>     --
>     Wolfgang Viechtbauer, Ph.D., Statistician | Department of Psychiatry
> and
>     Neuropsychology | Maastricht University | P.O. Box 616 (VIJV1) | 6200
> MD
>     Maastricht, The Netherlands | +31 (43) 388-4170 |
> http://www.wvbauer.com
>     -----Original Message-----
>     From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-
> bounces at r-project.org] On Behalf Of Nathan Pace
>     Sent: Wednesday, July 19, 2017 07:19
>     To: r-sig-meta-analysis at r-project.org
>     Subject: [R-meta] Ratio of Means
>     Hi All,
>     In two papers (J Clin Epidemiol 2011;64:556–564. BMC Med Res Method
> 2008;8(32)DOI: 10.1186/1471-2288-8-32)
>     Friedrich, Adhikari, and Beyene proposed methods for the meta analysis
> of the Ratio of Means of continuous outcomes.
>     Using the reported means, standard deviations, and sample sizes of the
> experimental and control groups, the log of the ratio of means and the
> associated standard error using a first order delta methods are estimated.
> This meta analysis then uses the generic inverse variance method.
>     I have a k = 23 meta analysis. I used the metafor function rma.uni
> with REML and knha = T; a moderator was included.
>     Any comments?
>     Any unexpected pitfalls?
>     Nathan
> _______________________________________________
> R-sig-meta-analysis mailing list
> R-sig-meta-analysis at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis

	[[alternative HTML version deleted]]

More information about the R-sig-meta-analysis mailing list