[R-meta] doubt about meta analysis

[Viechtbauer, Wolfgang (SP)]

Dear Lorenzo,

It isn't clear to me what exactly it is that you are after. As Michael said, only the primary studies can provide adjusted ORs (or if you had the raw data and would do an 'individual patient data meta-analysis'). Of course, one can include predictors ('moderators') in a meta-regression model, but that isn't the same thing as adjusting ORs based on individual patient data.

To use the example of using publication year as a moderator:

### load metafor package
library(metafor)

### calculate log odds ratios and corresponding sampling variances
dat <- escalc(measure="OR", ai=tpos, bi=tneg, ci=cpos, di=cneg, data=dat.bcg)

### meta-regression model with 'year' as moderator
res <- rma(yi, vi, mods = ~ year, data=dat)
res

One can compute predicted ORs at various values of 'year' now. For example:

### predicted OR in 1966
predict(res, newmods=1966, transf=exp)

I wouldn't call this an 'adjusted' OR though. It is simply the predicted OR in 1966 based on this model. 

Best,
Wolfgang

-----Original Message-----
From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org] On Behalf Of Martin Lobo
Sent: Monday, 18 March, 2019 16:52
To: Michael Dewey; r-sig-meta-analysis using r-project.org
Subject: Re: [R-meta] doubt about meta analysis

Dear Michael, thank you for answering, this technique has been used for several years, and in general the original publications do not include the adjusted values. Imagine that it would be very difficult for all publications to include adjustment for the variables that interest us. Even some meta-analyzes adjust for years of publication, something interesting due to changes in the way patients are treated year after year.

Lorenzo Martín Lobo MTSAC, FACC, FESC
Especialista Jerarquizado en Cardiología
Ex Jefe de Unidad Coronaria
Hospital Militar Campo de Mayo
Miembro Titular de la Sociedad Argentina de Cardiología
Fellow American College of Cardiology
Fellow European Society of Cardiology
Miembro del Area de Investigación de la SAC
Director del Consejo de Aterosclerosis y Trombosis de la SAC
Ex Director  y actual Miembro Asesor del Consejo de Epidemiología y Prevención Cardiovascular de la SAC

Instructor de ACLS de la AHA para la Fundación Cardiológica Argentina

________________________________
De: Michael Dewey <lists using dewey.myzen.co.uk>
Enviado: domingo, 17 de marzo de 2019 13:13
Para: Martin Lobo; r-sig-meta-analysis using r-project.org
Asunto: Re: [R-meta] doubt about meta analysis

Dear Martin

Unless either (a) the primary studies have published the adjusted OR, or
(b) you have the original data and can calculate it then I think the
answer is that waht you want to do is not possible.

Michael

On 17/03/2019 06:41, Martin Lobo wrote:
> Hi, my name is martin lobo. I�m cardiologist from argentina.
> I would like to ask you a question. I am wanting to do a meta analysis with the meta and metafor packages in R. I have not managed to adjust the odds ratio for certain covariates. That is, I need part of the crude OR, an adjusted OR for example, by age, cholesterol, sex, etc. Could I be oriented to achieve this type of adjustment in the meta analysis?
>  From already thank you very much
>
> Lorenzo Mart�n Lobo MTSAC, FACC, FESC
> Especialista Jerarquizado en Cardiolog�a
> Ex Jefe de Unidad Coronaria
> Hospital Militar Campo de Mayo
> Miembro Titular de la Sociedad Argentina de Cardiolog�a
> Fellow American College of Cardiology
> Fellow European Society of Cardiology
> Miembro del Area de Investigaci�n de la SAC
> Director del Consejo de Aterosclerosis y Trombosis de la SAC
> Ex Director  y actual Miembro Asesor del Consejo de Epidemiolog�a y Prevenci�n Cardiovascular de la SAC
>
> Instructor de ACLS de la AHA para la Fundaci�n Cardiol�gica Argentina