How to Meta analyze using the Least Square Mean ?

[Richard Kirubakaran]

Dear All Members,

I am currently involved in performing a Meta analysis for a  Systematic Review. The included RCT in this review have reported least square mean and Standard Error for the each respective group.

How do I use this Information and what would be the ideal effect measure for analysis?

Please advise.

Any help would be highly appreciated.

Thanks & Regards 

--
Richard Kirubakaran
Biostatistician
South Asian Cochrane Network & Centre
Prof BV Moses & ICMR Advanced Centre for Research and Evidence-Informed Health Care
Carman Block II Floor
Christian Medical College
Vellore 632002
Tamil Nadu, India
Hand phone: 9500 292922
web: www.cochrane-sacn.org
The Cochrane Library is now available free to all in INDIA!
Evidence for Healthcare decision-making

[Pedro Emmanuel Alvarenga Americano do Brasil]

Well

I allways see rct meta-analysis as pooled riskratios or odds ratios. I believe it is possible to do with continuous measures and the package metafor in R and metan in stata are able to do so. Take a look at their documentation.

Regards

Pedro Brasil
via Android (:)=

--
--
To post a new thread to MedStats, send email to MedS...@googlegroups.com .
MedStats' home page is http://groups.google.com/group/MedStats .
Rules: http://groups.google.com/group/MedStats/web/medstats-rules
 
---
You received this message because you are subscribed to the Google Groups "MedStats" group.
To unsubscribe from this group and stop receiving emails from it, send an email to medstats+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Richard Kirubakaran]

Thanks for your reply. I have done meta analysis for the continuous data using Mean Difference or Standardize Mean Difference as an effect measure.

In this specific situation ,say we have 5 included studies for the Meta analysis. Suppose 3 studies have reported  Least Square Mean score and the remaining 2 studies have reported Mean score for each groups. To combine these different data will give a false estimate.

Now the problem is 

If I considered doing a subgroup analysis using Least Square Mean and Mean score data separately then how to draw a conclusion form both the results. Both the data are measuring the same outcomes.

Is there any other thing can be done like 

Converting the Least square mean to arithmetic mean or .......

please advice..

Thanks.

Richard

PS:

Description of the Included study which has reported Least Square Mean:

Let me explain the study, Parallel Group design.  Arm 1 receives a Drug A and the Arm 2 receives Drug B.Treatment duration was 6 weeks. Outcomes are measured at the beginning (baseline evaluation) there after every 

week  measurement has been made to access the performance of the participants.

So the data have 6 time point. The  performed analysis was mixed-effects model

repeated measures (MMRM) using SAS which produced the adjusted mean (Least Square Mean) along with the estimated coefficient for the model parameters.

[Frank Harrell]

"Least squares mean" is a construct of SAS that is not otherwise standard in statistics.  State the exact contrast you wish to compute and people can probably be more helpful.

Frank

[Pedro Emmanuel Alvarenga Americano do Brasil]

Richard,

Frank has a point. Also, I insist you to check what kind of mean or continuous measure the functions you will use expect. Suppose the function expect a geometric mean and you give a arithmetic mean, the conclusion may be totally wrong, although the output is there for you. 

In systematic reviews one should meta-analyse only thing one consider results that could be combined. You are dealing with different measures of same effect, but another common example is trying to combine similar measures of different effects. It may not be appropriate for example to combine effects of certain medicine of trials of patients with diabetes and trials of patients without diabetes. This is not a statistical issue however. Systematic reviews may have good results without meta-analysis. 

You may contact the authors of the trials and ask them to report you the desired mean and SD. 

Regards,

Dr. Pedro Emmanuel A. A. do Brasil

Curriculum Lattes:  http://lattes.cnpq.br/6597654894290806 
Instituto de Pesquisa Clínica Evandro Chagas
Fundação Oswaldo Cruz
Rio de Janeiro - Brasil
Av. Brasil 4365,
CEP 21040-360,
Tel 55 21 3865-9648
email: pedro....@ipec.fiocruz.br
email: emmanue...@gmail.com

---Apoio aos softwares livres
www.zotero.org - gerenciamento de referências bibliográficas.
www.broffice.org ou www.libreoffice.org - textos, planilhas ou apresentações.
www.epidata.dk - entrada de dados.
www.r-project.org - análise de dados.
www.ubuntu.com - sistema operacional